1
0

Compare commits

...

938 Commits

Author SHA1 Message Date
Matthew Hodgson 5ea4d5d38a fix NPE 2018-01-07 23:08:54 +00:00
Matthew Hodgson f512e5c137 entirely blindly implement ?ts for AS. untested 2018-01-04 19:10:35 +00:00
Richard van der Hoff 3f9f1c50f3 Merge pull request #2683 from seckrv/fix_pwd_auth_prov_typo
synapse/config/password_auth_providers: Fixed bracket typo
2017-12-18 22:37:21 +00:00
Richard van der Hoff 48fa4e1e5b Merge pull request #2435 from silkeh/listen-ipv6-default
Adapt the default config to bind on both IPv4 and IPv6 on all platforms
2017-12-18 22:34:50 +00:00
Silke df0f602796 Implement listen_tcp method in remaining workers
Signed-off-by: Silke <silke@slxh.eu>
2017-12-18 20:00:42 +01:00
Silke 26cd3f5690 Remove logger argument and do not catch replication listener
Signed-off-by: Silke <silke@slxh.eu>
2017-12-18 20:00:42 +01:00
Silke Hofstra ed48ecc58c Add methods for listening on multiple addresses
Add listen_tcp and listen_ssl which implement Twisted's reactor.listenTCP
and reactor.listenSSL for multiple addresses.

Signed-off-by: Silke Hofstra <silke@slxh.eu>
2017-12-17 13:15:48 +01:00
Silke Hofstra 37d1a90025 Allow binds to both :: and 0.0.0.0
Binding on 0.0.0.0 when :: is specified in the bind_addresses is now allowed.
This causes a warning explaining the behaviour.
Configuration changed to match.

See #2232

Signed-off-by: Silke Hofstra <silke@slxh.eu>
2017-12-17 13:10:31 +01:00
Willem Mulder 3e59143ba8 Adapt the default config to bind on IPv6.
Most deployments are on Linux (or Mac OS), so this would actually bind
on both IPv4 and IPv6.

Resolves #1886.

Signed-off-by: Willem Mulder <willemmaster@hotmail.com>
2017-12-17 13:07:37 +01:00
Erik Johnston ba24576f2f Merge pull request #2717 from matrix-org/erikj/createroom_content
Fix wrong avatars when inviting multiple users when creating room
2017-12-07 20:00:34 +00:00
Erik Johnston d8a6c734fa Merge branch 'develop' of github.com:matrix-org/synapse into erikj/createroom_content 2017-12-07 14:24:01 +00:00
Erik Johnston ef045dcd71 Copy dict in update_membership too 2017-12-07 14:17:15 +00:00
Matthew Hodgson 33cb7ef0b7 Merge pull request #2723 from matrix-org/matthew/search-all-local-users
Add all local users to the user_directory and optionally search them
2017-12-05 11:09:47 +00:00
Matthew Hodgson cdc2cb5d11 fix StoreError syntax 2017-12-05 11:09:31 +00:00
Richard van der Hoff 16ec3805e5 Fix error when deleting devices
This was introduced in d7ea8c4 / PR #2728
2017-12-05 09:49:22 +00:00
Richard van der Hoff 8529874368 Merge pull request #2729 from matrix-org/rav/custom_ui_auth
support custom login types for validating users
2017-12-05 09:43:43 +00:00
Richard van der Hoff da1010c83a support custom login types for validating users
Wire the custom login type support from password providers into the UI-auth
user-validation flows.
2017-12-05 09:43:30 +00:00
Richard van der Hoff cc58e177f3 Merge pull request #2728 from matrix-org/rav/validate_user_via_ui_auth
Factor out a validate_user_via_ui_auth method
2017-12-05 09:42:46 +00:00
Richard van der Hoff d7ea8c4800 Factor out a validate_user_via_ui_auth method
Collect together all the places that validate a logged-in user via UI auth.
2017-12-05 09:42:30 +00:00
Richard van der Hoff aa6ecf0984 Merge pull request #2727 from matrix-org/rav/refactor_ui_auth_return
Refactor UI auth implementation
2017-12-05 09:40:38 +00:00
Richard van der Hoff d5f9fb06b0 Refactor UI auth implementation
Instead of returning False when auth is incomplete, throw an exception which
can be caught with a wrapper.
2017-12-05 09:40:05 +00:00
Matthew Hodgson c22e73293a speed up the rate of initial spam for users 2017-12-04 18:05:28 +00:00
Matthew Hodgson b11dca2025 better doc 2017-12-04 17:51:33 +00:00
Matthew Hodgson 7b86c1fdcd try make tests work a bit more... 2017-12-04 17:10:03 +00:00
Richard van der Hoff 58ebdb037c Merge pull request #2716 from matrix-org/rav/federation_client_post
federation_client script: Support for posting content
2017-12-04 17:04:08 +00:00
Matthew Hodgson 95f8a713dc erik told me to 2017-12-04 16:56:25 +00:00
Matthew Hodgson 74e0cc74ce fix pep8 and tests 2017-12-04 15:11:38 +00:00
Matthew Hodgson 1bd40ca73e switch to a simpler 'search_all_users' button as per review feedback 2017-12-04 14:58:39 +00:00
Matthew Hodgson f397153dfc Merge branch 'develop' into matthew/search-all-local-users 2017-11-30 01:51:38 +00:00
Matthew Hodgson 5406392f8b specify default user_directory_include_pattern 2017-11-30 01:45:34 +00:00
Matthew Hodgson f61e107f63 remove null constraint on user_dir.room_id 2017-11-30 01:43:50 +00:00
Matthew Hodgson 4b1fceb913 fix alternation operator for FTS4 - how did this ever work!? 2017-11-30 01:34:03 +00:00
Matthew Hodgson a4bb133b68 fix thinkos galore 2017-11-30 01:17:15 +00:00
Matthew Hodgson cd3697e8b7 kick the user_directory index when new users register 2017-11-29 18:33:34 +00:00
Matthew Hodgson 3241c7aac3 untested WIP but might actually work 2017-11-29 18:27:05 +00:00
Richard van der Hoff 624c46eb06 Merge pull request #2721 from matrix-org/rav/get_user_by_access_token_comments
Improve comments on get_user_by_access_token
2017-11-29 17:57:06 +00:00
Richard van der Hoff 7a48a6b63e Merge pull request #2722 from matrix-org/rav/delete_device_on_logout
Delete devices and pushers on logouts etc
2017-11-29 17:56:46 +00:00
Matthew Hodgson 47d99a20d5 Add user_directory_include_pattern config param to expand search results to additional users
Initial commit; this doesn't work yet - the LIKE filtering seems too aggressive.
It also needs _do_initial_spam to be aware of prepopulating the whole user_directory_search table with all users...
...and it needs a handle_user_signup() or something to be added so that new signups get incrementally added to the table too.

Committing it here as a WIP
2017-11-29 16:46:45 +00:00
Richard van der Hoff ad7e570d07 Delete devices in various logout situations
Make sure that we delete devices whenever a user is logged out due to any of
the following situations:

 * /logout
 * /logout_all
 * change password
 * deactivate account (by the user or by an admin)
 * invalidate access token from a dynamic module

Fixes #2672.
2017-11-29 16:44:35 +00:00
Richard van der Hoff ae31f8ce45 Move set_password into its own handler
Non-functional refactoring to move set_password. This means that we'll be able
to properly deactivate devices and access tokens without introducing a
dependency loop.
2017-11-29 16:44:35 +00:00
Richard van der Hoff 7ca5c68233 Move deactivate_account into its own handler
Non-functional refactoring to move deactivate_account. This means that we'll be
able to properly deactivate devices and access tokens without introducing a
dependency loop.
2017-11-29 16:44:35 +00:00
Richard van der Hoff 2c6d63922a Remove pushers when deleting access tokens
Whenever an access token is invalidated, we should remove the associated
pushers.
2017-11-29 16:44:35 +00:00
Richard van der Hoff 97d1a1dc01 Merge pull request #2718 from matrix-org/rav/notify_logcontexts
Clear logcontext before starting fed txn queue runner
2017-11-29 16:01:46 +00:00
Richard van der Hoff 8b45de90a4 Merge pull request #2719 from matrix-org/rav/handle_missing_hashes
Fix 500 when joining matrix-dev
2017-11-29 16:01:33 +00:00
Richard van der Hoff 7303ed65e1 Fix 500 when joining matrix-dev
matrix-dev has an event (`$/6ANj/9QWQyd71N6DpRQPf+SDUu11+HVMeKSpMzBCwM:zemos.net`)
which has no `hashes` member.

Check for missing `hashes` element in events.
2017-11-29 16:00:46 +00:00
Richard van der Hoff da562bd6a1 Improve comments on get_user_by_access_token
because I have to reverse-engineer this every time.
2017-11-29 15:52:41 +00:00
Richard van der Hoff d4fb4f7c52 Clear logcontext before starting fed txn queue runner
These processes take a long time compared to the request, so there is lots of
"Entering|Restoring dead context" in the logs. Let's try to shut it up a bit.
2017-11-28 15:26:14 +00:00
Erik Johnston dfbc45302e PEP8 2017-11-28 15:23:26 +00:00
Erik Johnston c4c1d170af Fix wrong avatars when inviting multiple users when creating room
We reused the `content` dictionary between invite requests, which meant they could end up reusing the profile info for a previous user
2017-11-28 15:19:15 +00:00
Richard van der Hoff fd04968f32 federation_client script: Support for posting content 2017-11-28 11:59:24 +00:00
Luke Barnard c2a1194424 Merge pull request #2715 from matrix-org/luke/group-guest-access
Allow guest access to group APIs for reading
2017-11-28 11:58:54 +00:00
Luke Barnard ab1b2d0ff2 Allow guest access to group APIs for reading 2017-11-28 11:23:00 +00:00
Richard van der Hoff 5a4da5bf78 Merge pull request #2697 from matrix-org/rav/fix_urlcache_index_error
Fix error on sqlite 3.7
2017-11-27 12:25:48 +00:00
Richard van der Hoff 84b31a3e7a Merge pull request #2713 from matrix-org/rav/no_upsert_forever
Avoid retrying forever on IntegrityError
2017-11-27 12:19:35 +00:00
Richard van der Hoff df6c72ede3 Merge pull request #2711 from matrix-org/rav/fix_dns_errhandler
Fix error handling on dns lookup
2017-11-27 12:19:18 +00:00
Richard van der Hoff 04bb79f139 Merge pull request #2710 from matrix-org/rav/remove_dead_code
Tiny code cleanups
2017-11-27 12:15:44 +00:00
Richard van der Hoff e828a7380a Merge pull request #2708 from matrix-org/rav/replication_logcontext_leaks
Fix some logcontext leaks in replication resource
2017-11-27 12:15:33 +00:00
Richard van der Hoff 7ef22a41a3 Merge pull request #2707 from matrix-org/rav/fix_urlpreview
Fix OPTIONS on preview_url
2017-11-27 12:15:14 +00:00
Richard van der Hoff 96387bd26f Merge pull request #2705 from matrix-org/rav/improve_tracebacks
Improve tracebacks on exceptions
2017-11-27 12:07:52 +00:00
Richard van der Hoff 6be01f599b Improve tracebacks on exceptions
Use failure.Failure to recover our failure, which will give us a useful
stacktrace, unlike the rethrown exception.
2017-11-27 12:05:58 +00:00
Richard van der Hoff 63ccaa5873 Avoid retrying forever on IntegrityError 2017-11-27 12:00:07 +00:00
Richard van der Hoff 8b38096a89 Fix error handling on dns lookup
pass the right arguments to the errback handler

Fixes "TypeError('eb() takes exactly 2 arguments (1 given)',)"
2017-11-24 16:47:48 +00:00
Richard van der Hoff 795b0849f3 Add a comment which might save some confusion 2017-11-24 00:34:56 +00:00
Richard van der Hoff 7f14f0ae38 Remove dead sync_callback
This is never used; let's remove it to stop confusing things.
2017-11-24 00:32:04 +00:00
Richard van der Hoff 0edf085b68 Fix some logcontext leaks in replication resource
The @measure_func annotations rely on the wrapped function respecting the
logcontext rules. Add the necessary yields to make this work.
2017-11-23 23:19:43 +00:00
Richard van der Hoff 8132a6b7ac Fix OPTIONS on preview_url
Fixes #2706
2017-11-23 17:52:31 +00:00
Richard van der Hoff 6b48b3e277 fix sql fails 2017-11-22 18:06:24 +00:00
Richard van der Hoff 2908f955d1 Check database in has_completed_background_updates
so that the right thing happens on workers.
2017-11-22 18:02:15 +00:00
Richard van der Hoff 79eba878a7 Merge pull request #2701 from matrix-org/rav/one_mediarepo_to_rule_them_all
Try to avoid having multiple PreviewUrlResource instances
2017-11-22 16:46:09 +00:00
Richard van der Hoff 68ca864141 Add config option to disable media_repo on main synapse
... to stop us doing the cache cleanup jobs on the master.
2017-11-22 16:20:27 +00:00
Richard van der Hoff e1fd4751de Build MediaRepositoryResource as a homeserver dependency
This avoids the scenario where we have four different PreviewUrlResources
configured on a single app, each of which have their own caches and cache
clearing jobs.
2017-11-22 16:19:49 +00:00
Richard van der Hoff 148c113fbe Merge pull request #2700 from matrix-org/rav/worker_docs
* Improve documentation of workers

Fixes https://github.com/matrix-org/synapse/issues/2554
2017-11-21 18:28:40 +00:00
Richard van der Hoff a0c6688976 Improve documentation of workers
Fixes https://github.com/matrix-org/synapse/issues/2554
2017-11-21 18:28:13 +00:00
Richard van der Hoff d5a7c56ef9 Merge pull request #2698 from matrix-org/rav/remove_dead_dependencies
Clean up dependency list
2017-11-21 17:38:42 +00:00
Richard van der Hoff 0b4aa2dc21 Merge pull request #2689 from matrix-org/rav/unlock_account_data_upsert
Avoid locking account_data tables for upserts
2017-11-21 13:39:14 +00:00
Matthew Hodgson 3ab2cfec47 sanity checks 2017-11-21 12:10:20 +00:00
Richard van der Hoff 7298ed7c51 Clean up dependency list
remove those that aren't used at all, and replace the ones that don't have
builders with simple getters rather than dynamically-generated methods.
2017-11-21 11:15:41 +00:00
Richard van der Hoff 7098b65cb8 Fix error on sqlite 3.7
Create the url_cache index on local_media_repository as a background update, so
that we can detect whether we are on sqlite or not and create a partial or
complete index accordingly.

To avoid running the cleanup job before we have built the index, add a bailout
which will defer the cleanup if the bg updates are still running.

Fixes https://github.com/matrix-org/synapse/issues/2572.
2017-11-21 11:14:17 +00:00
Matthew Hodgson 2145ee1976 don't double-invite in sync_room_to_group.pl 2017-11-19 00:48:47 +00:00
Richard van der Hoff 59a7275258 Merge pull request #2688 from matrix-org/rav/unlock_more_upsert
Avoid locking for upsert on pushers tables
2017-11-17 13:09:31 +00:00
Richard van der Hoff d8a05418f9 Merge branch 'master' into develop 2017-11-17 12:13:14 +00:00
Richard van der Hoff b102e93571 Merge branch 'release-v0.25.1' 2017-11-17 12:03:07 +00:00
Luke Barnard cdf6fc15b0 Merge pull request #2686 from matrix-org/luke/as-flair
Add automagical AS Publicised Group(s)
2017-11-17 10:13:46 +00:00
Richard van der Hoff 74bbeb4373 Bump version in __init__.py 2017-11-17 10:10:53 +00:00
Richard van der Hoff 2187724ad2 Prep changelog for v0.25.1 2017-11-17 10:09:16 +00:00
Jurek eded7084d2 Fix auth handler #2678 2017-11-17 10:07:27 +00:00
Matthew Hodgson 34c3d0a386 typo 2017-11-17 01:54:02 +00:00
Matthew Hodgson 9d50b6f0ea quick and dirty room membership<->group membership sync script 2017-11-17 01:54:02 +00:00
Luke Barnard ab1dc84779 Add extra space before inline comment 2017-11-16 18:22:40 +00:00
Luke Barnard 7fb0e98b03 Extract group_id from the dict for multiple use 2017-11-16 18:18:30 +00:00
Luke Barnard e836bdf734 Fix tests 2017-11-16 18:14:39 +00:00
Richard van der Hoff c46139a17e Avoid locking account_data tables for upserts 2017-11-16 18:08:01 +00:00
Luke Barnard d8391f0541 Remove unused GROUP_ID_REGEX 2017-11-16 18:05:57 +00:00
Luke Barnard 4e8374856d Document get_groups_for_user 2017-11-16 18:03:46 +00:00
Luke Barnard 270f9cd23a Flake8 2017-11-16 18:03:31 +00:00
Luke Barnard 9d83d52027 Use a generator instead of a list 2017-11-16 17:57:34 +00:00
Luke Barnard 5b48eec4a1 Make sure we check AS groups for lookup on bulk 2017-11-16 17:55:15 +00:00
Luke Barnard b1edf26051 Check group_id belongs to this domain 2017-11-16 17:54:27 +00:00
Richard van der Hoff 06e5bcfc83 Avoid locking for upsert on pushers tables
* replace the upsert into deleted_pushers with an insert
* no need to lock for upsert on pusher_throttle
2017-11-16 17:52:23 +00:00
Jurek 624a8bbd67 Fix auth handler #2678 2017-11-16 17:19:02 +00:00
Richard van der Hoff b26cbbb60e Revert "Merge pull request #2679 from jkolo/fix_auth_handler"
This PR was against master, not develop :(

This reverts commit 203058a027, reversing
changes made to 552f123bea.
2017-11-16 17:18:11 +00:00
Richard van der Hoff 203058a027 Merge pull request #2679 from jkolo/fix_auth_handler
Fix auth handler
2017-11-16 17:09:00 +00:00
Luke Barnard 97bd18af4e Add automagical AS Publicised Group(s)
via registration file "users" namespace:

```YAML
...
namespaces:
  users:
    - exclusive: true
      regex: '.*luke.*'
      group_id: '+all_the_lukes:hsdomain'
...
```

This is part of giving App Services their own groups for matching users. With this, ghost users will be given the appeareance that they are in a group and that they have publicised the fact, but _only_ from the perspective of the `get_publicised_groups_for_user` API.
2017-11-16 16:44:55 +00:00
Richard van der Hoff ba05f28ae7 Merge pull request #2684 from matrix-org/rav/unlock_upsert
Start work on avoiding table locks for upserts
2017-11-16 16:27:29 +00:00
Richard van der Hoff 77a1227870 Fix broken ref to IntegrityError 2017-11-16 16:03:38 +00:00
Richard van der Hoff 7ab2b69e18 Avoid locking pushers table on upsert
Now that _simple_upsert will retry on IntegrityError, we don't need to lock the
table.
2017-11-16 15:32:01 +00:00
Richard van der Hoff 10aaa1bc15 _simple_upsert: retry on IntegrityError
wrap the call to _simple_upsert_txn in a loop so that we retry on an
integrityerror: this means we can avoid locking the table provided there is an
unique index.
2017-11-16 15:30:15 +00:00
Richard van der Hoff cdc9e50a5d Cleanup in _simple_upsert_txn
Bail out early to reduce indentation
2017-11-16 15:29:10 +00:00
Richard von Seck 6f05de0e5e synapse/config/password_auth_providers: Fixed bracket typo
Signed-off-by: Richard von Seck <richard.von-seck@gmx.net>
2017-11-16 15:59:38 +01:00
Jurek 56e2a4333e Fix auth handler #2678 2017-11-15 22:49:43 +01:00
Richard van der Hoff f959c01600 Merge pull request #2661 from matrix-org/rav/statereadstore
Pull out bits of StateStore to a mixin
2017-11-15 17:23:01 +00:00
Richard van der Hoff 117a8c0d35 Merge pull request #2677 from matrix-org/rav/spec_r0.3.0
Declare support for r0.3.0
2017-11-15 17:10:45 +00:00
Richard van der Hoff 30d2730ee2 Declare support for r0.3.0 2017-11-15 16:24:22 +00:00
Erik Johnston aa812feb41 Merge branch 'master' of github.com:matrix-org/synapse into develop 2017-11-15 11:35:34 +00:00
Erik Johnston 552f123bea Merge branch 'release-v0.25.0' of github.com:matrix-org/synapse 2017-11-15 11:32:24 +00:00
Erik Johnston 5d0cbf763f Bump changelog 2017-11-15 11:29:32 +00:00
Richard van der Hoff 1b83c09c03 Merge pull request #2675 from matrix-org/rav/remove_broken_logcontext_funcs
Remove preserve_context_over_{fn, deferred}
2017-11-15 11:13:53 +00:00
David Baker 7190a550dc Merge pull request #2650 from matrix-org/dbkr/push_include_content_option
Rename redact_content option to include_content
2017-11-15 10:47:38 +00:00
Richard van der Hoff b2cd6accf5 Remove __PreservingContextDeferred too 2017-11-14 23:00:10 +00:00
Richard van der Hoff 038c994724 Merge pull request #2648 from krombel/update_prometheus
update prometheus-config to new format
2017-11-14 19:14:32 +00:00
Krombel c161472575 Make clear that the config has changed since prometheus v2
This restores the config that is usable for prometheus pre v2.0.0
The new config only works for Prometheus v2+
2017-11-14 19:59:26 +01:00
Richard van der Hoff 008aa2fc6d Merge pull request #2671 from matrix-org/rav/room_list_fixes
Reshuffle room list request code
2017-11-14 18:11:02 +00:00
Richard van der Hoff 6f30fd9235 Merge pull request #2673 from matrix-org/erikj/fix_port_script_groups
Add new boolean columns to port script
2017-11-14 16:03:41 +00:00
Erik Johnston 9ecf621404 Less s's 2017-11-14 15:55:15 +00:00
Erik Johnston 22db751d1e Add new boolean columns to port script 2017-11-14 15:48:50 +00:00
Erik Johnston 03feb7a34d Bump version and changelog 2017-11-14 14:51:25 +00:00
Richard van der Hoff 35a4b63240 Pull out bits of StateStore to a mixin
... so that we don't need to secretly gut-wrench it for use in the slaved
stores. I haven't done the other stores yet, but we should. I'm tired of the
workers breaking every time we tweak the stores because I forgot to gut-wrench
the right method.

fixes https://github.com/matrix-org/synapse/issues/2655.
2017-11-14 11:43:58 +00:00
Richard van der Hoff 4dd1bfa8c1 Revert "Revert "move _state_group_cache to statestore""
We're going to fix this properly on this branch, so that the _state_group_cache
can end up in StateGroupReadStore.

This reverts commit ab335edb02.
2017-11-14 11:43:58 +00:00
Richard van der Hoff 6caa379ba1 Merge pull request #2658 from matrix-org/rav/store_heirarchy_init
Make __init__ consistent across Store hierarchy
2017-11-14 11:43:12 +00:00
Richard van der Hoff 7e6fa29cb5 Remove preserve_context_over_{fn, deferred}
Both of these functions ae known to leak logcontexts. Replace the remaining
calls to them and kill them off.
2017-11-14 11:22:42 +00:00
Richard van der Hoff 44a1bfd6a6 Reshuffle room list request code
I'm not entirely sure if this will actually help anything, but it simplifies
the code and might give further clues about why room list search requests are
blowing out the get_current_state_ids caches.
2017-11-14 10:29:58 +00:00
Richard van der Hoff 1fc66c7460 Add a load of logging to the room_list handler
So we can see what it gets up to.
2017-11-14 10:23:47 +00:00
Richard van der Hoff 7bd6c87eca Merge pull request #2668 from turt2live/travis/whoami
Add a route for determining who you are
2017-11-14 09:54:21 +00:00
Travis Ralston 812c191939 Remove redundent call
Signed-off-by: Travis Ralston <travpc@gmail.com>
2017-11-13 12:44:21 -07:00
Richard van der Hoff c741ba59c9 Merge pull request #2669 from matrix-org/rav/cache_urlpreview_failure
Cache failures in url_preview handler
2017-11-13 18:36:24 +00:00
Richard van der Hoff 781c15a6a3 Merge pull request #2663 from matrix-org/rav/invalid_request_utf8
Fix 500 on invalid utf-8 in request
2017-11-13 18:35:24 +00:00
David Baker 45ab288e07 Print instead of logging
because we had to wait until the logger was set up
2017-11-13 18:32:08 +00:00
Richard van der Hoff 8b33ac8f6c Merge branch 'develop' into rav/invalid_request_utf8 2017-11-13 11:56:22 +00:00
Richard van der Hoff 63ef607f1f Fix tests for Store.__init__ update
Fix the test to pass the right number of args to the Store constructors
2017-11-13 10:46:08 +00:00
Richard van der Hoff 6cfee09be9 Make __init__ consitstent across Store heirarchy
Add db_conn parameters to the `__init__` methods of the *Store classes, so that
they are all consistent, which makes the multiple inheritance work correctly
(and so that we can later extract mixins which can be used in the slavedstores)
2017-11-13 10:46:07 +00:00
Erik Johnston ab335edb02 Revert "move _state_group_cache to statestore"
This reverts commit f5cf3638e9.
2017-11-13 10:05:33 +00:00
Erik Johnston bfbf1e1f1a Up cache size of get_global_account_data_by_type_for_user 2017-11-13 09:52:11 +00:00
Travis Ralston 2d314b771f Add a route for determining who you are
Useful for applications which may have an access token, but no idea as to who owns it.

Signed-off-by: Travis Ralston <travpc@gmail.com>
2017-11-12 23:39:38 -07:00
Richard van der Hoff 5d15abb120 Bit more logging 2017-11-10 16:58:04 +00:00
Richard van der Hoff 46790f50cf Cache failures in url_preview handler
Reshuffle the caching logic in the url_preview handler so that failures are
cached (and to generally simplify things and fix the logcontext leaks).
2017-11-10 16:50:50 +00:00
Richard van der Hoff 4d0414c714 Merge pull request #2662 from matrix-org/rav/fix_mxids_again
Downcase userid on registration
2017-11-10 13:57:34 +00:00
Richard van der Hoff e508145c9b Add some more comments appservice user registration
Explain why we don't validate userids registered via app services
2017-11-10 12:39:45 +00:00
Richard van der Hoff e0ebd1e4bd Downcase userids for shared-secret registration 2017-11-10 12:39:05 +00:00
Richard van der Hoff f90649eb2b Fix 500 on invalid utf-8 in request
If somebody sends us a request where the the body is invalid utf-8, we should
return a 400 rather than a 500. (json.loads throws a UnicodeError in this
situation)

We might as well catch all Exceptions here: it seems very unlikely that we
would get a request that *isn't caused by invalid json.
2017-11-10 09:15:39 +00:00
Richard van der Hoff 9b599bc18d Downcase userid on registration
Force username to lowercase before attempting to register

https://github.com/matrix-org/synapse/issues/2660
2017-11-09 22:20:01 +00:00
Richard van der Hoff 9b803ccc98 Revert "Allow upper-case characters in mxids"
This reverts commit b70b646903.
2017-11-09 21:57:24 +00:00
Richard van der Hoff 1282086f58 Merge pull request #2659 from matrix-org/rav/apparently_we_dont_follow_our_own_spec_now
Allow upper-case characters in mxids
2017-11-09 20:06:47 +00:00
Richard van der Hoff b70b646903 Allow upper-case characters in mxids
Because we're never going to be able to fix this :'(
2017-11-09 19:36:13 +00:00
Erik Johnston 2dce6b15c3 Fix typo 2017-11-09 15:56:16 +00:00
Erik Johnston 4e2b2508af Register group servlet 2017-11-09 15:49:42 +00:00
Luke Barnard 0ea5310290 Merge pull request #2657 from matrix-org/erikj/group_visibility_namespace
Namespace visibility options for groups
2017-11-09 15:41:51 +00:00
Erik Johnston 13735843c7 Namespace visibility options for groups 2017-11-09 15:27:18 +00:00
Richard van der Hoff 618c7b816a Merge pull request #2656 from matrix-org/rav/fix_deactivate
Fix 'NoneType' not iterable in /deactivate
2017-11-09 15:20:35 +00:00
Erik Johnston 0fcb5a8ce5 Merge pull request #2651 from matrix-org/erikj/update_group_room_settings
Change so that update group room visibility isn't an upsert
2017-11-09 15:20:24 +00:00
Richard van der Hoff 889102315e Fix 'NoneType' not iterable in /deactivate
make sure we actually return a value from user_delete_access_tokens
2017-11-09 15:15:33 +00:00
David Baker b2a788e902 Make the commented config have the default 2017-11-09 10:11:42 +00:00
Erik Johnston 82e4bfb53d Add brackets 2017-11-09 10:06:42 +00:00
Erik Johnston e8814410ef Have an explicit API to update room config 2017-11-08 16:13:27 +00:00
Erik Johnston 94ff2cda73 Revert "Modify group room association API to allow modification of is_public" 2017-11-08 15:43:34 +00:00
Erik Johnston d305987b40 Merge pull request #2631 from xyzz/fix_appservice_event_backlog
Fix appservices being backlogged and not receiving new events due to a bug in notify_interested_services
2017-11-08 11:54:10 +00:00
Erik Johnston 167eb01d83 Merge pull request #2637 from spantaleev/avoid-noop-media-deletes
Avoid no-op media deletes
2017-11-08 11:53:27 +00:00
David Baker ad408beb66 better comments 2017-11-08 11:50:08 +00:00
David Baker 1b870937ae Log if any of the old config flags are set 2017-11-08 11:46:24 +00:00
David Baker 2a98ba0ed3 Rename redact_content option to include_content
The redact_content option never worked because it read the wrong config
section. The PR introducing it
(https://github.com/matrix-org/synapse/pull/2301) had feedback suggesting the
name be changed to not re-use the term 'redact' but this wasn't
incorporated.

This reanmes the option to give it a less confusing name, and also
means that people who've set the redact_content option won't suddenly
see a behaviour change when upgrading synapse, but instead can set
include_content if they want to.

This PR also updates the wording of the config comment to clarify
that this has no effect on event_id_only push.

Includes https://github.com/matrix-org/synapse/pull/2422
2017-11-08 10:35:30 +00:00
Richard van der Hoff 02a9a93bde Merge pull request #2649 from matrix-org/rav/fix_delta_on_state_res
Fix bug in state group storage
2017-11-08 09:22:13 +00:00
Richard van der Hoff e148438e97 s/items/iteritems/ 2017-11-08 09:21:41 +00:00
Ilya Zhuravlev d46386d57e Remove useless assignment in notify_interested_services 2017-11-07 22:23:22 +03:00
Matthew Hodgson 228ccf1fe3 Merge pull request #2643 from matrix-org/matthew/user_dir_typos
Fix various embarrassing typos around user_directory and add some doc.
2017-11-07 17:31:11 +00:00
Richard van der Hoff 780dbb378f Update deltas when doing auth resolution
Fixes a bug where the persisted state groups were different to those actually
being used after auth resolution.
2017-11-07 16:43:00 +00:00
Richard van der Hoff 1ca4288135 factor out _update_context_for_auth_events
This is duplicated, so let's factor it out before fixing it
2017-11-07 16:43:00 +00:00
Richard van der Hoff f5cf3638e9 move _state_group_cache to statestore
this is internal to statestore, so let's keep it there.
2017-11-07 16:43:00 +00:00
Erik Johnston 5ef5e14ecc Merge pull request #2636 from farialima/me-master
Fix for #2635: correctly update rooms avatar/display name when modified by admin
2017-11-07 13:49:27 +00:00
Erik Johnston 76c9af193c Revert "Merge branch 'master' of github.com:matrix-org/synapse into develop"
This reverts commit f9b255cd62, reversing
changes made to 1bd654dabd.
2017-11-07 13:32:35 +00:00
Erik Johnston f9b255cd62 Merge branch 'master' of github.com:matrix-org/synapse into develop 2017-11-07 13:31:03 +00:00
Krombel 44ad6dd4bf update prometheus-config to new format 2017-11-07 13:35:35 +01:00
Luke Barnard 1bd654dabd Merge pull request #2647 from matrix-org/luke/get-group-users-is-privileged
Return whether a user is an admin within a group
2017-11-07 12:05:11 +00:00
Luke Barnard 38b265cb51 Remember to pick is_admin out of the db 2017-11-07 11:24:04 +00:00
Luke Barnard 5561c09091 Return whether a user is an admin within a group 2017-11-07 11:18:45 +00:00
Matthew Hodgson 3db5ff69b2 Merge pull request #2576 from maximevaillancourt/exclude-noscript-url-preview
Ignore <noscript> tags when generating URL preview descriptions
2017-11-07 11:09:22 +00:00
Richard van der Hoff ec12e7eada Merge pull request #2646 from matrix-org/rav/logging_for_limiter
Logging and logcontext fixes for Limiter
2017-11-07 10:47:00 +00:00
Matthew Hodgson 631fa4a1b7 create new indexes before dropping old ones to keep safetynet in place 2017-11-07 10:41:55 +00:00
Richard van der Hoff bf993db11c Logging and logcontext fixes for Limiter
Add some logging to the Limiter in a similar spirit to the Linearizer, to help
debug issues.

Also fix a logcontext leak.

Also refactor slightly to avoid throwing exceptions.
2017-11-07 00:48:57 +00:00
Matthew Hodgson 4ad883398f s/users_in_pubic_room/users_in_public_rooms/g 2017-11-04 19:39:40 +00:00
Matthew Hodgson d802e8ca6a s/users_in_pubic_room/users_in_public_rooms/g 2017-11-04 19:38:13 +00:00
Matthew Hodgson a100700630 fix copyright.... 2017-11-04 19:35:49 +00:00
Matthew Hodgson b6b075fd49 s/popualte/populate/ 2017-11-04 19:35:33 +00:00
Matthew Hodgson d1622e080f s/intial/initial/ 2017-11-04 19:35:14 +00:00
Matthew Hodgson 2ac6deafb7 simplify instructions for regenerating user_dir 2017-11-04 19:34:59 +00:00
Slavi Pantaleev 805196fbeb Avoid no-op media deletes
If there are no media entries to delete,
avoid creating transactions, prepared statements
and unnecessary log entries.

Signed-off-by: Slavi Pantaleev <slavi@devture.com>
2017-11-04 09:50:15 +02:00
Francois Granade f103b91ffa removed unused import flagged by flake8a 2017-11-03 18:45:49 +01:00
Francois Granade fa4f337b49 Fix for issue 2635: correctly update rooms avatar/display name when modified by admin 2017-11-03 18:25:04 +01:00
Ilya Zhuravlev 8a4a0ddea6 Fix appservice tests to account for new behavior of notify_interested_services 2017-11-02 23:19:57 +03:00
Ilya Zhuravlev 45fbe4ff67 Fix appservices being backlogged and not receiving new events due to a bug in notify_interested_services 2017-11-02 22:49:43 +03:00
David Baker f851bc8182 Merge pull request #2630 from matrix-org/luke/fix-rooms-in-group
Make the get_rooms_in_group API more sane
2017-11-02 17:23:17 +00:00
David Baker 9e09a1800b Merge pull request #2629 from matrix-org/rav/register_inhibit_login
support inhibit_login in /register
2017-11-02 16:51:35 +00:00
Luke Barnard a34c586a89 Make the get_rooms_in_group API more sane
Return entries with is_public = True when they're public and is_public = False otherwise.
2017-11-02 16:42:30 +00:00
Richard van der Hoff 6c3a02072b support inhibit_login in /register
Allow things to pass inhibit_login when registering to ... inhibit logins.
2017-11-02 16:31:07 +00:00
David Baker 4a6754baf2 Merge pull request #2628 from matrix-org/rav/module_api_hooks
Add more hooks to ModuleApi
2017-11-02 15:37:14 +00:00
David Baker 4b36897cd9 Merge remote-tracking branch 'origin/develop' into rav/module_api_hooks 2017-11-02 15:19:17 +00:00
David Baker d4553818a0 Merge pull request #2627 from matrix-org/rav/custom_rest_endpoints
Add a hook for custom rest endpoints
2017-11-02 15:18:37 +00:00
David Baker 6b6f03ae05 Merge pull request #2626 from matrix-org/rav/refactor_module_api
Factor _AccountHandler proxy out to ModuleApi
2017-11-02 15:15:30 +00:00
David Baker 77e3757fa9 Merge pull request #2625 from matrix-org/rav/named_resource_refactor
Factor out _configure_named_resource
2017-11-02 15:11:30 +00:00
Richard van der Hoff 6b60f7dca0 Add more hooks to ModuleApi
add `get_user_by_req` and `invalidate_access_token`
2017-11-02 14:37:39 +00:00
Richard van der Hoff fcdfc911ee Add a hook for custom rest endpoints
Let the user specify custom modules which can be used for implementing extra
endpoints.
2017-11-02 14:36:55 +00:00
Richard van der Hoff 1189be43a2 Factor _AccountHandler proxy out to ModuleApi
We're going to need to use this from places that aren't password auth, so let's
move it to a proper class.
2017-11-02 14:36:11 +00:00
Richard van der Hoff 6650a07ede Factor out _configure_named_resource
This was a bit of a code vomit, so let's factor it out to preserve some sanity
2017-11-02 14:33:37 +00:00
David Baker b19d9e2174 Merge pull request #2624 from matrix-org/rav/password_provider_notify_logout
Notify auth providers on logout
2017-11-02 10:55:17 +00:00
David Baker 1f080a6c97 Merge pull request #2623 from matrix-org/rav/callbacks_for_auth_providers
Allow password_auth_providers to return a callback
2017-11-02 10:49:03 +00:00
David Baker 04897c9dc1 Merge pull request #2622 from matrix-org/rav/db_access_for_auth_providers
Let auth providers get to the database
2017-11-02 10:41:25 +00:00
Richard van der Hoff 979eed4362 Fix user-interactive password auth
this got broken in the previous commit
2017-11-01 17:03:20 +00:00
Richard van der Hoff bc8a5c0330 Notify auth providers on logout
Provide a hook by which auth providers can be notified of logouts.
2017-11-01 16:51:51 +00:00
Richard van der Hoff 4c8f94ac94 Allow password_auth_providers to return a callback
... so that they have a way to record access tokens.
2017-11-01 16:51:03 +00:00
Richard van der Hoff 846a94fbc9 Merge pull request #2620 from matrix-org/rav/auth_non_password
Let password auth providers handle arbitrary login types
2017-11-01 16:45:33 +00:00
Richard van der Hoff 3cd6b22c7b Let password auth providers handle arbitrary login types
Provide a hook where password auth providers can say they know about other
login types, and get passed the relevant parameters
2017-11-01 16:43:57 +00:00
David Baker c9b9ef575b Merge pull request #2621 from matrix-org/rav/refactor_accesstoken_delete
Move access token deletion into auth handler
2017-11-01 16:26:06 +00:00
Matthew Hodgson 275826f234 Merge pull request #2617 from matrix-org/matthew/auto-displayname
automatically set default displayname on register
2017-11-01 16:21:16 +00:00
David Baker 4f0488b307 Merge remote-tracking branch 'origin/develop' into rav/refactor_accesstoken_delete 2017-11-01 16:20:19 +00:00
David Baker e5e930aec3 Merge pull request #2615 from matrix-org/rav/break_auth_device_dep
Break dependency of auth_handler on device_handler
2017-11-01 16:06:31 +00:00
David Baker fbbacb284e Merge pull request #2613 from matrix-org/rav/kill_refresh_tokens
Remove the last vestiges of refresh_tokens
2017-11-01 15:57:35 +00:00
Matthew Hodgson 9f7a555b4e switch to setting default displayname in the storage layer
to avoid clobbering guest user displaynames on registration
2017-11-01 15:51:30 +00:00
Richard van der Hoff dd13310fb8 Move access token deletion into auth handler
Also move duplicated deactivation code into the auth handler.

I want to add some hooks when we deactivate an access token, so let's bring it
all in here so that there's somewhere to put it.
2017-11-01 15:46:22 +00:00
David Baker 691cc4e036 Merge pull request #2618 from matrix-org/dbkr/log_login_requests
Log login requests
2017-11-01 14:11:28 +00:00
David Baker 0bb253f37b Apparently this is python 2017-11-01 14:02:52 +00:00
David Baker 59e7e62c4b Log login requests
Carefully though, to avoid logging passwords
2017-11-01 13:58:01 +00:00
Matthew Hodgson f8420d6279 automatically set default displayname on register
to avoid leaking ugly MXIDs and cluttering up the timeline with
displayname changes as well as membership joins for autojoin rooms
(e.g. the status autojoin rooms), automatically set the displayname
to match the localpart of the mxid upon registration.
2017-11-01 13:15:41 +00:00
Luke Barnard 99354b430e Merge pull request #2612 from matrix-org/luke/groups-room-relationship-is-public
Modify group room association API to allow modification of is_public
2017-11-01 11:08:36 +00:00
Richard van der Hoff 74c56f794c Break dependency of auth_handler on device_handler
I'm going to need to make the device_handler depend on the auth_handler, so I
need to break this dependency to avoid a cycle.

It turns out that the auth_handler was only using the device_handler in one
place which was an edge case which we can more elegantly handle by throwing an
error rather than fixing it up.
2017-11-01 10:27:06 +00:00
Richard van der Hoff 02237ce725 Fix tests for refresh_token removal 2017-11-01 10:19:42 +00:00
Luke Barnard 318a249c8b Leave is_public as required argument of update_room_group_association 2017-11-01 09:36:01 +00:00
Luke Barnard 207fabbc6a Update docs for updating room group association 2017-11-01 09:35:15 +00:00
Richard van der Hoff 356bcafc44 Remove the last vestiges of refresh_tokens 2017-10-31 20:35:58 +00:00
Richard van der Hoff 3e0aaad190 Let auth providers get to the database
Somewhat open to abuse, but also somewhat unavoidable :/
2017-10-31 17:22:29 +00:00
Richard van der Hoff a72e4e3e28 Merge pull request #2610 from matrix-org/rav/schema_for_pw_providers
DB schema interface for password auth providers
2017-10-31 17:20:47 +00:00
Luke Barnard 13b3d7b4a0 Flake8 2017-10-31 17:20:11 +00:00
Richard van der Hoff e025aec028 Merge pull request #2611 from matrix-org/dbkr/port_script_drop_nuls
Make the port script drop NUL values in all tables
2017-10-31 17:16:08 +00:00
Luke Barnard 20fe347906 Modify group room association API to allow modification of is_public
also includes renamings to make things more consistent.
2017-10-31 17:04:28 +00:00
David Baker 9d419f48e6 Make the port script drop NUL values in all tables
Postgres doesn't support NULs in strings so it makes the script
throw an exception and stop if any values contain \0. Drop them
with appropriate warning.
2017-10-31 16:58:49 +00:00
Richard van der Hoff 9ded00f221 fix tests 2017-10-31 14:21:13 +00:00
Richard van der Hoff 1650eb5847 DB schema interface for password auth providers
Provide an interface by which password auth providers can register db schema
files to be run at startup
2017-10-31 14:01:53 +00:00
David Baker c31a7c3ff6 Merge pull request #2609 from matrix-org/rav/refactor_login
Refactor some logic from LoginRestServlet into AuthHandler
2017-10-31 13:51:36 +00:00
Richard van der Hoff b8e54fbc08 Merge pull request #2607 from matrix-org/rav/cleanup_ldap_hacks
Clean up backwards-compat hacks for ldap
2017-10-31 13:41:12 +00:00
David Baker a1f8b0fd64 Merge pull request #2608 from matrix-org/rav/password_provider_doc
Start some documentation on password providers
2017-10-31 13:40:10 +00:00
Richard van der Hoff 1b65ae00ac Refactor some logic from LoginRestServlet into AuthHandler
I'm going to need some more flexibility in handling login types in password
auth providers, so as a first step, move some stuff from LoginRestServlet into
AuthHandler.

In particular, we pass everything other than SAML, JWT and token logins down to
the AuthHandler, which now has responsibility for checking the login type and
fishing the password out of the login dictionary, as well as qualifying the
user_id if need be. Ideally SAML, JWT and token would go that way too, but
there's no real need for it right now and I'm trying to minimise impact.

This commit *should* be non-functional.
2017-10-31 10:48:41 +00:00
Richard van der Hoff ebda45de4c Start some documentation on password providers
Document the existing interface, before I start adding new stuff.
2017-10-31 10:47:52 +00:00
Richard van der Hoff ffc574a6f9 Clean up backwards-compat hacks for ldap
try to make the backwards-compat flows follow the same code paths as the modern
impl.

This commit should be non-functional.
2017-10-31 10:47:02 +00:00
Richard van der Hoff e2f4190209 Merge pull request #2605 from matrix-org/luke/fix-group-creation-error-wording
Fix wording on group creation error
2017-10-30 16:06:42 +00:00
Luke Barnard 9bc17fc5fb Fix wording on group creation error 2017-10-30 15:17:23 +00:00
Matthew Hodgson 208a6647f1 fix typo 2017-10-29 20:54:20 +00:00
Matthew Hodgson e51c2bcaef move url_previews to MD as RST does my head in 2017-10-29 20:47:17 +00:00
Erik Johnston 71a1bd53b2 Merge pull request #2599 from matrix-org/erikj/groups_invite
Fix typo when checking if user is invited to group
2017-10-27 17:34:54 +01:00
Erik Johnston d0abb4e8e6 Fix typo when checking if user is invited to group 2017-10-27 16:57:19 +01:00
Erik Johnston 977078f06d Fix bad merge 2017-10-27 15:10:50 +01:00
Erik Johnston 6980c4557e Merge branch 'erikj/attestation_jitter' of github.com:matrix-org/synapse into develop 2017-10-27 15:09:05 +01:00
Erik Johnston 632baf799e Merge pull request #2598 from matrix-org/revert-2596-erikj/attestation_jitter
Revert "Add jitter to validity period of attestations"
2017-10-27 15:08:29 +01:00
Erik Johnston af92f5b00f Revert "Add jitter to validity period of attestations" 2017-10-27 15:07:21 +01:00
Erik Johnston 4ab8abbc2b Merge branch 'erikj/attestation_local_fix' of github.com:matrix-org/synapse into develop 2017-10-27 15:07:08 +01:00
Erik Johnston b1e62d4a57 Merge pull request #2596 from matrix-org/erikj/attestation_jitter
Add jitter to validity period of attestations
2017-10-27 14:20:25 +01:00
Erik Johnston 6af3656deb Merge pull request #2595 from matrix-org/erikj/attestation_commnet
Add comment about attestations
2017-10-27 14:20:19 +01:00
Richard van der Hoff 4d83632009 Merge pull request #2591 from matrix-org/rav/device_delete_auth
Device deletion: check UI auth matches access token
2017-10-27 12:30:10 +01:00
Richard van der Hoff 110b373e9c Merge pull request #2589 from matrix-org/rav/as_deactivate_account
Allow ASes to deactivate their own users
2017-10-27 12:29:32 +01:00
Erik Johnston ca571b0ec3 Add jitter to validity period of attestations
This helps ensure that the renewals of attestations are spread out more
evenly.
2017-10-27 11:57:27 +01:00
Luke Barnard d8c26162a1 Merge pull request #2582 from matrix-org/luke/group-is-public
Add is_public to groups table to allow for private groups
2017-10-27 11:41:13 +01:00
Erik Johnston c067088747 Add comment about attestations 2017-10-27 11:35:41 +01:00
Luke Barnard 5451cc7792 Request is_public from database 2017-10-27 11:27:43 +01:00
Luke Barnard 124314672f group is dict 2017-10-27 11:08:19 +01:00
Luke Barnard 6362298fa5 Create groups with is_public = True 2017-10-27 11:04:20 +01:00
Richard van der Hoff 8b56977b6f Merge pull request #2586 from matrix-org/rav/frontend_proxy_auth_header
Front-end proxy: pass through auth header
2017-10-27 11:01:50 +01:00
Richard van der Hoff 173567a7f2 Docstring for post_urlencoded_get_json 2017-10-27 10:59:50 +01:00
Luke Barnard c7d9f25d22 Fix create_group to pass requester_user_id 2017-10-27 10:57:20 +01:00
Erik Johnston e27b76d117 Import logger 2017-10-27 10:54:02 +01:00
Richard van der Hoff 8854c039f2 Merge pull request #2585 from matrix-org/rav/unstable_to_r0
Support /keys/upload on /r0 as well as /unstable
2017-10-27 10:53:48 +01:00
Richard van der Hoff 14f581abc2 Merge pull request #2584 from matrix-org/rav/fix_httpclient_logcontexts
Fix logcontext leaks in httpclient
2017-10-27 10:53:29 +01:00
Luke Barnard 2ca46c7afc Correct logic for checking private group membership 2017-10-27 10:48:01 +01:00
Erik Johnston 82d8c1bacb Fixup 2017-10-27 10:30:21 +01:00
Erik Johnston 2fd9831f7c Merge pull request #2574 from matrix-org/erikj/room_list_fixes
Add logging and fix log contexts for publicRooms
2017-10-27 10:01:23 +01:00
Erik Johnston 195abfe7a5 Remove incorrect attestations 2017-10-27 09:58:13 +01:00
Erik Johnston d8dde19f04 Log if we try to do attestations for our own user and group 2017-10-27 09:55:01 +01:00
Erik Johnston 585972b51a Don't generate group attestations for local users 2017-10-27 09:46:56 +01:00
Richard van der Hoff 7a6546228b Device deletion: check UI auth matches access token
(otherwise there's no point in the UI auth)
2017-10-27 00:04:31 +01:00
Matthew Hodgson 92f680889d spell out need for libxml2 for lxml to work 2017-10-27 00:02:29 +01:00
Richard van der Hoff 785bd7fd75 Allow ASes to deactivate their own users 2017-10-27 00:01:00 +01:00
Richard van der Hoff c89e6aadff Merge pull request #2581 from matrix-org/rav/fix_init_with_no_logfile
Fix error when running synapse with no logfile
2017-10-26 22:16:57 +01:00
Richard van der Hoff 54a2525133 Front-end proxy: pass through auth header
So that access-token-in-an-auth-header works.
2017-10-26 18:19:01 +01:00
Richard van der Hoff 0a5866bec9 Support /keys/upload on /r0 as well as /unstable
(So that we can stop riot relying on it in /unstable)
2017-10-26 18:18:23 +01:00
Richard van der Hoff 0d8e3ad48b Fix logcontext leaks in httpclient
`preserve_context_over_fn` is borked
2017-10-26 18:17:10 +01:00
Richard van der Hoff 12ef02dc3d SimpleHTTPClient: add support for headers
Sometimes we need to pass headers into these methods
2017-10-26 17:59:50 +01:00
Luke Barnard 69e8a05f35 Make it work 2017-10-26 17:55:58 +01:00
Luke Barnard 007cd48af6 Recreate groups table instead of adding column
Adding a column with non-constant default not possible in sqlite3
2017-10-26 17:55:22 +01:00
Luke Barnard 713e60b9b6 Awful hack to get default true 2017-10-26 17:38:14 +01:00
Luke Barnard e86cefcb6f Add groups table to BOOLEAN_COLUMNS in synapse_port_db 2017-10-26 17:24:54 +01:00
Luke Barnard cfa4e658e0 Bump schema version to 46 2017-10-26 17:23:49 +01:00
Luke Barnard 595fe67f01 delint 2017-10-26 17:20:24 +01:00
Luke Barnard 9b2feef9eb Add is_public to groups table to allow for private groups
Prevent group API access to non-members for private groups

Also make all the group code paths consistent with `requester_user_id` always being the User ID of the requesting user.
2017-10-26 16:51:32 +01:00
Richard van der Hoff f7f90e0c8d Fix error when running synapse with no logfile
Fixes 'UnboundLocalError: local variable 'sighup' referenced before assignment'
2017-10-26 16:45:20 +01:00
Richard van der Hoff 1dd0f53b21 Merge pull request #2579 from krombel/move_unstable_to_r0
register some /unstable endpoints in /r0 as well
2017-10-26 16:10:43 +01:00
Krombel 8299b323ee add release endpoints for /thirdparty 2017-10-26 16:58:20 +02:00
Krombel 9b436c8b4c register some /unstable endpoints in /r0 as well 2017-10-26 15:22:50 +02:00
Richard van der Hoff 5b38fdab31 Merge pull request #2578 from matrix-org/rav/code_style_imports
Code_style updates
2017-10-26 11:55:59 +01:00
Richard van der Hoff 1eb300e1fc Document import rules 2017-10-26 11:55:41 +01:00
Richard van der Hoff f7f6bfaae4 code_style: more formatting 2017-10-26 11:55:41 +01:00
Erik Johnston 4ea882ede4 Merge pull request #2577 from matrix-org/erikj/fix_port
Fix port script
2017-10-26 11:54:12 +01:00
Erik Johnston 566e21eac8 Update room_list.py 2017-10-26 11:39:54 +01:00
Richard van der Hoff 351cc35342 code_style.rst: a couple of tidyups 2017-10-26 10:29:26 +01:00
Erik Johnston 37d766aedd Fix port script
We changed _simple_update_one_txn to use _simple_update_txn but didn't
yank it out in the port script.

Fixes #2565
2017-10-26 10:01:03 +01:00
Maxime Vaillancourt 5287e57c86 Ignore noscript tags when generating URL previews 2017-10-25 20:44:34 -04:00
Erik Johnston 2a7e9faeec Do logcontexts outside ResponseCache 2017-10-25 15:21:08 +01:00
Erik Johnston 1ad1ba9e6a Merge branch 'master' of github.com:matrix-org/synapse into develop 2017-10-25 10:27:23 +01:00
Erik Johnston 33a9026cdf Add logging and fix log contexts for publicRooms 2017-10-25 10:26:06 +01:00
Matthew Hodgson efd0f5a3c5 tip for generating tls_fingerprints 2017-10-24 18:49:49 +01:00
Erik Johnston f009df23ec Merge branch 'release-v0.24.1' of github.com:matrix-org/synapse 2017-10-24 15:02:25 +01:00
Erik Johnston 6ba4fabdb9 Bump version and changelog 2017-10-24 14:15:27 +01:00
Erik Johnston 9e2c22c97f Merge pull request #2567 from matrix-org/erikj/group_fed_update_profile
Correctly wire in update group profile over federation
2017-10-24 09:21:47 +01:00
Erik Johnston 39dc52157d Merge branch 'develop' of github.com:matrix-org/synapse into erikj/group_fed_update_profile 2017-10-24 09:16:20 +01:00
Richard van der Hoff 0d437698b2 Merge pull request #2568 from matrix-org/rav/pep8
PEP8 fixes
2017-10-23 17:39:31 +01:00
Richard van der Hoff 0be99858f3 fix vars named l
E741 says "do not use variables named ‘l’, ‘O’, or ‘I’".
2017-10-23 15:56:38 +01:00
Richard van der Hoff eaaabc6c4f replace 'except:' with 'except Exception:'
what could possibly go wrong
2017-10-23 15:52:32 +01:00
Erik Johnston ce6d4914f4 Correctly wire in update group profile over federation 2017-10-23 15:21:24 +01:00
Richard van der Hoff ecf198aab8 Merge pull request #2566 from matrix-org/rav/media_logcontext_leak
Fix a logcontext leak in the media repo
2017-10-23 14:47:49 +01:00
Richard van der Hoff 3267b81b81 Merge pull request #2561 from matrix-org/rav/id_checking
Updates to ID checking
2017-10-23 14:39:20 +01:00
Richard van der Hoff d03cfc4258 Fix a logcontext leak in the media repo 2017-10-23 14:34:27 +01:00
Erik Johnston 1de557975f Merge branch 'master' of github.com:matrix-org/synapse into develop 2017-10-23 13:18:12 +01:00
Erik Johnston ffba978077 Merge branch 'release-v0.24.0' of github.com:matrix-org/synapse 2017-10-23 13:13:53 +01:00
Erik Johnston 13e16cf302 Bump version and changelog 2017-10-23 13:13:31 +01:00
Richard van der Hoff bd0d84bf92 Merge pull request #2560 from matrix-org/rav/kill_pointless_method
Remove pointless create() method
2017-10-23 11:06:57 +01:00
Richard van der Hoff 1135193dfd Validate group ids when parsing
May as well do it whenever we parse a Group ID. We check the sigil and basic
structure here so it makes sense to check the grammar in the same place.
2017-10-21 00:30:39 +01:00
Richard van der Hoff 29812c628b Allow = in mxids and groupids
... because the spec says we should.
2017-10-20 23:42:53 +01:00
Richard van der Hoff 58fbbe0f1d Disallow capital letters in userids
Factor out a common function for checking user ids and group ids, which forbids
capitals.
2017-10-20 23:37:22 +01:00
Richard van der Hoff 631d7b87b5 Remove pointless create() method
It just calls the constructor, so we may as well kill it rather than having
random codepaths.
2017-10-20 22:14:55 +01:00
Erik Johnston 6070647774 Correctly bump version 2017-10-19 16:40:20 +01:00
Erik Johnston d6237859f6 Update changelog 2017-10-19 14:19:52 +01:00
Erik Johnston 0ef0aeceac Bump version and changelog 2017-10-19 13:48:33 +01:00
Erik Johnston b4a6b7f720 Merge pull request #2559 from matrix-org/erikj/group_id_validation
Add config to enable group creation
2017-10-19 13:45:09 +01:00
Erik Johnston c7d46510d7 Flake8 2017-10-19 13:36:06 +01:00
Erik Johnston ffd3f1a783 Add missing file... 2017-10-19 12:17:30 +01:00
Erik Johnston 29bafe2f7e Add config to enable group creation 2017-10-19 12:13:44 +01:00
Erik Johnston 287dd1ee2c Merge pull request #2558 from matrix-org/erikj/group_id_validation
Enforce sensible group IDs
2017-10-19 12:13:31 +01:00
Erik Johnston 513c23bfd9 Enforce sensible group IDs 2017-10-19 12:01:01 +01:00
Erik Johnston 011d03a0f6 Fix typo 2017-10-19 11:22:48 +01:00
Erik Johnston 9ab859f27b Fix typo in group attestation handling 2017-10-19 10:55:52 +01:00
Erik Johnston f4f65ef93e Merge pull request #2557 from matrix-org/erikj/media_tumbnails
Fix typo in thumbnail generation
2017-10-19 10:38:52 +01:00
Erik Johnston bd5718d0ad Fix typo in thumbnail generation 2017-10-19 10:27:18 +01:00
Erik Johnston 161a862ffb Fix typo 2017-10-19 10:17:43 +01:00
Richard van der Hoff 69994c385a Merge pull request #2553 from matrix-org/rav/fix_500_on_event_send
Fix 500 error when we get an error handling a PDU
2017-10-18 11:25:44 +01:00
Richard van der Hoff b5dbbac308 Merge pull request #2552 from matrix-org/rav/fix_500_on_dodgy_powerlevels
Fix 500 error when fields missing from power_levels event
2017-10-17 20:53:30 +01:00
Richard van der Hoff 582bd19ee9 Fix 500 error when we get an error handling a PDU
FederationServer doesn't have a send_failure (and nor does its subclass,
ReplicationLayer), so this was failing.

I'm not really sure what the idea behind send_failure is, given (a) we don't do
anything at the other end with it except log it, and (b) we also send back the
failure via the transaction response. I suspect there's a whole lot of dead
code around it, but for now I'm just removing the broken bit.
2017-10-17 20:52:40 +01:00
Richard van der Hoff 74f99f227c Doc some more dynamic Homeserver methods 2017-10-17 20:51:29 +01:00
Richard van der Hoff c2bd177ea0 Fix 500 error when fields missing from power_levels event
If the users or events keys were missing from a power_levels event, then
we would throw 500s when trying to auth them.
2017-10-17 17:05:42 +01:00
Erik Johnston fe6e9f580b Merge pull request #2550 from krombel/fix_thumbnail_2548
fix thumbnailing (#2548)
2017-10-17 15:35:18 +01:00
Richard van der Hoff 7216c76654 Improve error handling for missing files (#2551)
`os.path.exists` doesn't allow us to distinguish between permissions errors and
the path actually not existing, which repeatedly confuses people. It also means
that we try to overwrite existing key files, which is super-confusing. (cf
issues #2455, #2379). Use os.stat instead.

Also, don't recomemnd the the use of --generate-config, which screws everything
up if you're using debian (cf #2455).
2017-10-17 14:46:17 +01:00
Richard van der Hoff dbdfd8967d Merge pull request #2546 from matrix-org/rav/remove_dead_event_injector
Remove dead class
2017-10-17 13:21:22 +01:00
Richard van der Hoff b8e40d146f Merge pull request #2547 from matrix-org/rav/test_make_deferred_yieldable
Add some tests for make_deferred_yieldable
2017-10-17 13:21:09 +01:00
Richard van der Hoff 4cc8bb0767 Merge pull request #2549 from matrix-org/rav/event_persist_logcontexts
Fix logcontext handling for persist_events
2017-10-17 13:20:35 +01:00
David Baker 4e242b3e20 Merge pull request #2545 from matrix-org/dbkr/auto_join_rooms
Add config option to auto-join new users to rooms
2017-10-17 11:45:49 +01:00
Krombel a6245478c8 fix thumbnailing (#2548)
in commit 0e28281a the code for thumbnailing got refactored and the
renaming of this variables was not done correctly.

Signed-Off-by: Matthias Kesler <krombel@krombel.de>
2017-10-17 12:45:33 +02:00
Richard van der Hoff 2e9f5ea31a Fix logcontext handling for persist_events
* don't use preserve_context_over_deferred, which is known broken.

* remove a redundant preserve_fn.

* add/improve some comments
2017-10-17 10:59:30 +01:00
Richard van der Hoff a6ad8148b9 Fix name of test_logcontext
The file under test is logcontext.py, not log_context.py
2017-10-17 10:53:34 +01:00
Richard van der Hoff 5b5f35ccc0 Add some tests for make_deferred_yieldable 2017-10-17 10:52:31 +01:00
Richard van der Hoff 9b714abf35 Remove dead class
This isn't used anywhere.
2017-10-17 10:43:36 +01:00
David Baker 33122c5a1b Fix test 2017-10-17 10:39:50 +01:00
David Baker a9c2e930ac pep8 2017-10-17 10:13:13 +01:00
David Baker c05e6015cc Add config option to auto-join new users to rooms
New users who register on the server will be dumped into all rooms in
auto_join_rooms in the config.
2017-10-16 17:57:27 +01:00
Luke Barnard e0a75e0c25 Merge pull request #2544 from matrix-org/luke/groups-invited-users
Implement GET /groups/$groupId/invited_users
2017-10-16 17:33:33 +02:00
Luke Barnard 85f5674e44 Delint 2017-10-16 15:52:17 +01:00
Luke Barnard c43e8a9736 Make it work. Warn about lack of user profile 2017-10-16 15:50:39 +01:00
Luke Barnard a3ac4f6b0a _create_rererouter for get_invited_users_in_group 2017-10-16 15:41:03 +01:00
Luke Barnard 5dfd0350c7 Merge branch 'develop' into luke/groups-invited-users 2017-10-16 15:34:36 +01:00
Luke Barnard ca96d609e4 Merge pull request #2543 from matrix-org/luke/fix-on-group-invite-no-profile
Log a warning when no profile for invited member
2017-10-16 16:33:53 +02:00
Luke Barnard 2c5972f87f Implement GET /groups/$groupId/invited_users 2017-10-16 15:31:11 +01:00
Luke Barnard 6079d0027a Log a warning when no profile for invited member
And return empty profile
2017-10-16 14:20:45 +01:00
David Baker 99a6c9dbf2 Merge pull request #2542 from matrix-org/dbkr/room_notif_no_glob
Omit the *s for @room notifications
2017-10-16 13:46:17 +01:00
David Baker 9342bcfce0 Omit the *s for @room notifications
They're just redundant
2017-10-16 13:38:10 +01:00
Richard van der Hoff e504816977 Merge pull request #2540 from 4nd3r/patch-1
make it absolutely clear that purge doesn't remove everything
2017-10-16 10:43:22 +01:00
Ander Punnar b2e02084b8 make it absolutely clear that Purge History API does not remove all traces of events and message contents
because this topic pops up too often

#890 #1621 #1730 #2260 #2315 and so on
2017-10-14 13:25:42 +03:00
Erik Johnston db3d84f46c Merge pull request #2538 from matrix-org/erikj/media_backup
Basic implementation of backup media store
2017-10-13 16:11:31 +01:00
Erik Johnston 1b6b0b1e66 Add try/finally block to close t_byte_source 2017-10-13 15:34:08 +01:00
Erik Johnston 6b725cf56a Remove old comment 2017-10-13 15:23:41 +01:00
Matthew Hodgson 64665b57d0 oops 2017-10-13 14:26:07 +01:00
Erik Johnston 2b24416e90 Don't reuse source but instead copy from primary media store to backup 2017-10-13 14:11:34 +01:00
Erik Johnston b92a8e6e4a PEP8 2017-10-13 13:58:57 +01:00
Matthew Hodgson 931fc43cc8 fix copyright to companies which actually exist(ed) 2017-10-13 13:54:31 +01:00
Erik Johnston 31aa7bd8d1 Move type into key 2017-10-13 13:47:38 +01:00
Erik Johnston ad1911bbf4 Comment 2017-10-13 13:47:05 +01:00
Erik Johnston c021c39cbd Remove spurious addition 2017-10-13 13:46:53 +01:00
Erik Johnston 1f43d22397 Don't needlessly rename variable 2017-10-13 11:42:07 +01:00
Erik Johnston a675bd08bd Add paths back in... 2017-10-13 11:41:06 +01:00
Erik Johnston 4d7e1dde70 Remove unnecessary diff 2017-10-13 11:36:32 +01:00
Erik Johnston ae5d18617a Make things be absolute paths again 2017-10-13 11:35:44 +01:00
Erik Johnston 9732ec6797 s/write_to_file/write_to_file_and_backup/ 2017-10-13 11:34:41 +01:00
Erik Johnston 0e28281a02 Fix up 2017-10-13 11:33:49 +01:00
Erik Johnston 505371414f Fix up thumbnailing function 2017-10-13 11:23:53 +01:00
Erik Johnston e3428d26ca Fix typo 2017-10-13 10:39:59 +01:00
Erik Johnston 35332298ef Fix up comments 2017-10-13 10:39:32 +01:00
Erik Johnston 64db043a71 Move makedirs to thread 2017-10-13 10:25:01 +01:00
Erik Johnston b60859d6cc Use make_deferred_yieldable 2017-10-13 10:24:19 +01:00
Erik Johnston d76621a47b Fix comments 2017-10-12 18:16:25 +01:00
Erik Johnston 4ae85ae121 Don't close prematurely.. 2017-10-12 17:57:31 +01:00
Erik Johnston cc505b4b5e getvalue closes buffer 2017-10-12 17:52:30 +01:00
Erik Johnston 1259a76047 Get len before close 2017-10-12 17:39:23 +01:00
Erik Johnston 802ca12d05 Don't close file prematurely 2017-10-12 17:37:21 +01:00
Erik Johnston e283b555b1 Copy everything to backup 2017-10-12 17:31:24 +01:00
Erik Johnston b77a13812c Typo 2017-10-12 15:32:32 +01:00
Erik Johnston 6dfde6d485 Remove dead code 2017-10-12 15:30:26 +01:00
Erik Johnston c8eeef6947 Fix typos 2017-10-12 15:28:24 +01:00
Erik Johnston 67cb89fbdf Fix typo 2017-10-12 15:23:41 +01:00
Erik Johnston bf4fb1fb40 Basic implementation of backup media store 2017-10-12 15:20:59 +01:00
hera f807f7f804 log when we get an exception handling replication updates 2017-10-12 11:51:24 +01:00
David Baker b8d8ed1ba9 Merge pull request #2531 from matrix-org/dbkr/spamcheck_error_messages
Allow error strings from spam checker
2017-10-12 10:31:03 +01:00
Richard van der Hoff cc794d60e7 Merge pull request #2532 from matrix-org/rav/fix_linearizer
Fix stackoverflow and logcontexts from linearizer
2017-10-11 17:29:32 +01:00
Richard van der Hoff 8dd0c85ac5 Merge pull request #2529 from matrix-org/rav/fix_transaction_failure_handling
log pdu_failures from incoming transactions
2017-10-11 17:29:14 +01:00
Richard van der Hoff 76fa695241 Merge pull request #2515 from matrix-org/rav/fix_receipt_logcontext
A logformatter which includes the stack when the exception was caught when
logging exceptions.
2017-10-11 17:28:01 +01:00
Richard van der Hoff f30c4ed2bc logformatter: fix AttributeError
make sure we have the relevant fields before we try to log them.
2017-10-11 17:26:17 +01:00
Erik Johnston b752507b48 Fix fetching remote summaries 2017-10-11 16:59:18 +01:00
Erik Johnston af94ba9d02 Merge pull request #2533 from matrix-org/erikj/fix_group_repl
Fix group stream replication
2017-10-11 16:01:57 +01:00
Erik Johnston 818b08d0e4 peeeeeeeeep8888888888888888888888888888 2017-10-11 15:54:00 +01:00
Erik Johnston ea18996f54 Fix group stream replication
The stream update functions expect the storage function to return a list
of tuples.
2017-10-11 15:44:39 +01:00
Richard van der Hoff 68fd82e840 Merge pull request #2530 from matrix-org/rav/fix_receipt_logcontext
fix a logcontext leak in read receipt handling
2017-10-11 15:08:53 +01:00
Richard van der Hoff 4fad8efbfb Fix stackoverflow and logcontexts from linearizer
1. make it not blow out the stack when there are more than 50 things waiting
   for a lock. Fixes https://github.com/matrix-org/synapse/issues/2505.

2. Make it not mess up the log contexts.
2017-10-11 15:05:05 +01:00
David Baker b78bae2d51 fix isinstance 2017-10-11 14:49:09 +01:00
Erik Johnston 271f5601f3 Fix typo in invite to group 2017-10-11 14:45:33 +01:00
David Baker c3b7a45e84 Allow error strings from spam checker 2017-10-11 14:39:22 +01:00
Richard van der Hoff c3e190ce67 fix a logcontext leak in read receipt handling 2017-10-11 14:37:20 +01:00
Richard van der Hoff b75d443caf log pdu_failures from incoming transactions
... even if we have no EDUs.

This appears to have been introduced in
476899295f.
2017-10-11 14:36:13 +01:00
Erik Johnston 27e727a146 Fix typo 2017-10-11 14:32:40 +01:00
Erik Johnston 4ce4379235 Fix attestations to check correct server name 2017-10-11 14:11:43 +01:00
Erik Johnston c2c47550f9 Fix schema delta versions 2017-10-11 13:23:15 +01:00
Erik Johnston 535cc49f27 Merge pull request #2466 from matrix-org/erikj/groups_merged
Initial Group Implementation
2017-10-11 13:20:07 +01:00
Erik Johnston dfbf73408c Merge pull request #2501 from matrix-org/dbkr/channel_notifications
Support for channel notifications
2017-10-11 13:19:29 +01:00
Erik Johnston bc7f3eb32f Merge pull request #2483 from jeremycline/unfreeze-ujson-dump
Unfreeze event before serializing with ujson
2017-10-11 13:18:52 +01:00
Erik Johnston ec954f47fb Validate room ids 2017-10-11 13:15:44 +01:00
David Baker 81a5e0073c pep8 2017-10-10 15:53:34 +01:00
David Baker ab1bc9bf5f Don't KeyError if no power_levels event 2017-10-10 15:34:05 +01:00
David Baker 0f1eb3e914 Use notification levels in power_levels
Rather than making the condition directly require a specific power
level. This way the level require to notify a room can be configured
per room.
2017-10-10 15:23:00 +01:00
Erik Johnston 84e27a592d Merge pull request #2490 from matrix-org/erikj/drop_left_room_events
Ignore incoming events for rooms that we have left
2017-10-10 11:58:32 +01:00
David Baker c9f034b4ac There was already a constant for this
also update copyright
2017-10-10 11:47:10 +01:00
David Baker a9f9d68631 More optimisation 2017-10-10 11:38:31 +01:00
David Baker 707374d5dc What year is it!? Who's the president!? 2017-10-10 11:21:41 +01:00
David Baker 89fa00ddff Merge branch 'develop' into dbkr/channel_notifications 2017-10-10 11:20:17 +01:00
Richard van der Hoff 79bea15830 Merge pull request #2520 from matrix-org/rav/process_incoming_rooms_in_parallel
fed server: process PDUs for different rooms in parallel
2017-10-10 10:26:21 +01:00
Richard van der Hoff 426f8b0f66 Merge pull request #2518 from matrix-org/rav/linearize_incoming_transactions
Fed server: use a linearizer for ongoing transactions
2017-10-10 10:20:51 +01:00
Richard van der Hoff 6a6cc27aee fed server: process PDUs for different rooms in parallel
With luck, this will give a real-time improvement when there are many rooms and
the server ends up calling out to fetch missing events.
2017-10-09 18:30:31 +01:00
Richard van der Hoff 4c7c4d4061 Fed server: use a linearizer for ongoing transactions
We don't want to process the same transaction multiple times concurrently, so
use a linearizer.
2017-10-09 18:30:10 +01:00
Richard van der Hoff 4d24becf7f Merge pull request #2517 from matrix-org/rav/fed_server_refactor
fed server: refactor on_incoming_transaction
2017-10-09 18:19:23 +01:00
Richard van der Hoff ba5b9b80a5 fed server: refactor on_incoming_transaction
Move as much as possible to after the have_responded check, and reduce the
number of times we iterate over the pdu list.
2017-10-09 18:10:53 +01:00
Richard van der Hoff c7b0678356 Merge pull request #2516 from matrix-org/rav/fix_fed_server_origin_check
Fed server: Move origin-check code to _handle_received_pdu
2017-10-09 18:09:43 +01:00
Richard van der Hoff a6e3222fe5 Fed server: Move origin-check code to _handle_received_pdu
The response-building code expects there to be an entry in the `results` list
for each entry in the pdu_list, so the early `continue` was messing this
up. That doesn't really matter, because all that the federation client does is
log any errors, but it's pretty poor form.
2017-10-09 17:53:32 +01:00
Richard van der Hoff 3cc852d339 Fancy logformatter to format exceptions better
This is a bit of an experimental change at this point; the idea is to see if it
helps us track down where our stack overflows are coming from by logging the
stack when the exception was caught and turned into a Failure. (We'll also need
https://github.com/richvdh/twisted/commit/edf27044200e74680ea67c525768e36dc9d9af2b).

If we deploy this, we'll be able to enable it via the log config yaml.
2017-10-09 17:44:42 +01:00
Richard van der Hoff 0eeaa25694 Merge pull request #2508 from matrix-org/rav/federation_queue_logcontexts
Fix up logcontext handling in (federation) TransactionQueue
2017-10-09 17:43:48 +01:00
Richard van der Hoff aa3fac8057 Merge pull request #2507 from matrix-org/rav/execute_concurrently_log_contexts
Fix logcontext handling for concurrently_execute
2017-10-09 17:43:32 +01:00
Richard van der Hoff c1c81ee2a4 Merge pull request #2506 from matrix-org/rav/unhandled_failure
Fix up deferred handling in federation.py
2017-10-09 17:43:10 +01:00
Erik Johnston e8496efe84 Fix up comment 2017-10-09 15:17:34 +01:00
Richard van der Hoff 01bbacf3c4 Fix up logcontext handling in (federation) TransactionQueue
Avoid using preserve_context_over_function, which has problems with respect to
logcontexts.
2017-10-06 22:39:25 +01:00
Richard van der Hoff 148428ce76 Fix logcontext handling for concurrently_execute
Avoid preserve_context_over_deferred, which is broken.
2017-10-06 22:24:28 +01:00
Richard van der Hoff c8f568ddf9 Fix up deferred handling in federation.py
* Avoid preserve_context_over_deferred, which is broken

* set consumeErrors=True on defer.gatherResults, to avoid spurious "unhandled
  failure" erros
2017-10-06 22:14:24 +01:00
Richard van der Hoff 3ddda939d3 some comments in the state res code 2017-10-05 14:58:17 +01:00
David Baker 5de926d66f Merge pull request #2502 from matrix-org/dbkr/allow_dms_to_admins
Spam checking: add the invitee to user_may_invite
2017-10-05 14:14:13 +01:00
David Baker f878e6f8af Spam checking: add the invitee to user_may_invite 2017-10-05 14:02:28 +01:00
David Baker 269af961e9 Make be faster 2017-10-05 13:27:12 +01:00
David Baker ed80c6b6cc Add fastpath optimisation 2017-10-05 13:20:22 +01:00
David Baker e433393c4f pep8 2017-10-05 13:08:02 +01:00
David Baker 985ce80375 They're called rooms 2017-10-05 13:03:44 +01:00
David Baker b9b9714fd5 Get rule type right 2017-10-05 13:02:19 +01:00
David Baker fa969cfdde Support for channel notifications
Add condition type to check the sender's power level and add a base
rule using it for @channel notifications.
2017-10-05 12:39:18 +01:00
David Baker 44f8e383f3 Merge pull request #2500 from matrix-org/dbkr/fix_word_boundary_mentions
Fix notif kws that start/end with non-word chars
2017-10-05 12:27:59 +01:00
David Baker 0c8da8b519 Use better method for word boundary searching
From https://github.com/matrix-org/matrix-js-sdk/commit/ebc95667b8a5777d13e5d3c679972bedae022fd5
2017-10-05 11:57:43 +01:00
Erik Johnston eaaa837e00 Don't corrupt cache 2017-10-05 11:43:22 +01:00
David Baker cbe3c3fdd4 pep8 2017-10-05 11:43:10 +01:00
David Baker 6748f0a579 Fix notif kws that start/end with non-word chars
Only prepend / append word bounary characters if the search
expression starts or ends with a word character, otherwise they
don't work because there's no word bounary between whitespace and
a non-word char.
2017-10-05 11:33:30 +01:00
David Baker 93b0cf7a99 Merge pull request #2495 from matrix-org/dbkr/spam_check_room_creation
Add room creation checks to spam checker
2017-10-04 14:49:20 +01:00
David Baker d8ce68b09b spam check room publishing 2017-10-04 14:29:33 +01:00
David Baker 78d4ced829 un-double indent 2017-10-04 12:44:27 +01:00
David Baker 197c14dbcf Add room creation checks to spam checker
Lets the spam checker deny attempts to create rooms and add aliases
to them.
2017-10-04 10:47:54 +01:00
David Baker 5f20a91fa1 Merge pull request #2492 from matrix-org/dbkr/spam_check_invites
Allow spam checker to reject invites too
2017-10-03 18:10:23 +01:00
David Baker 1e2ac54351 s/roomid/room_id/ 2017-10-03 17:41:38 +01:00
David Baker 1e375468de pass room id too 2017-10-03 17:13:14 +01:00
David Baker c2c188b699 Federation was passing strings anyway
so pass string everywhere
2017-10-03 15:46:19 +01:00
David Baker c46a0d7eb4 this shouldn't be debug 2017-10-03 15:20:14 +01:00
David Baker bd769a81e1 better logging 2017-10-03 15:16:40 +01:00
David Baker 537088e7dc Actually write warpper function 2017-10-03 14:28:12 +01:00
David Baker 41fd9989a2 Skip spam check for admin users 2017-10-03 14:17:44 +01:00
Erik Johnston 11d62f43c9 Invalidate cache 2017-10-03 14:12:28 +01:00
Erik Johnston e4ab96021e Update comments 2017-10-03 14:10:41 +01:00
David Baker 2a7ed700d5 Fix param name & lint 2017-10-03 14:04:10 +01:00
David Baker 84716d267c Allow spam checker to reject invites too 2017-10-03 13:56:43 +01:00
Richard van der Hoff e4779be97a Merge pull request #2491 from matrix-org/rav/port_db_fixes
Drop search values with nul characters
2017-10-03 13:49:05 +01:00
Erik Johnston f2da6df568 Remove spurious line feed 2017-10-03 11:31:06 +01:00
Erik Johnston 30848c0fcd Ignore incoming events for rooms that we have left
When synapse receives an event for a room its not in over federation, it
double checks with the remote server to see if it is in fact in the
room. This is done so that if the server has forgotten about the room
(usually as a result of the database being dropped) it can recover from
it.

However, in the presence of state resets in large rooms, this can cause
a lot of work for servers that have legitimately left. As a hacky
solution that supports both cases we drop incoming events for rooms that
we have explicitly left.

This means that we no longer support the case of servers having
forgotten that they've rejoined a room, but that is sufficiently rare
that we're not going to support it for now.
2017-10-03 11:18:21 +01:00
Erik Johnston e585c83209 Merge branch 'master' of github.com:matrix-org/synapse into develop 2017-10-02 18:11:24 +01:00
Erik Johnston 6c1bb1601e Bump version and changelog 2017-10-02 18:05:17 +01:00
Erik Johnston ea87cb1ba5 Make 'affinity' package optional 2017-10-02 18:03:59 +01:00
Erik Johnston 3fed5bb25f Move quit_with_error 2017-10-02 17:59:34 +01:00
David Baker 27955056e0 Merge branch 'develop' into erikj/groups_merged 2017-10-02 16:20:41 +01:00
Erik Johnston 90d70af269 Merge branch 'master' of github.com:matrix-org/synapse into develop 2017-10-02 16:20:23 +01:00
Erik Johnston b23cb8fba8 Merge branch 'release-v0.23.0' of github.com:matrix-org/synapse 2017-10-02 13:52:03 +01:00
Erik Johnston e4a709eda3 Bump version and change log 2017-10-02 13:51:38 +01:00
Richard van der Hoff 7fc1aad195 Drop search values with nul characters
https://github.com/matrix-org/synapse/issues/2187 contains a report of a port
failing due to nul characters somewhere in the search table. Let's try dropping
the offending rows.
2017-10-02 00:53:32 +01:00
Jeremy Cline cafb8de132 Unfreeze event before serializing with ujson
In newer versions of https://github.com/esnme/ultrajson, ujson does not
serialize frozendicts (introduced in esnme/ultrajson@53f85b1). Although
the PyPI version is still 1.35, Fedora ships with a build from commit
esnme/ultrajson@2f1d487. This causes the serialization to fail if the
distribution-provided package is used.

This runs the event through the unfreeze utility before serializing it.

Thanks to @ignatenkobrain for tracking down the root cause.

fixes #2351

Signed-off-by: Jeremy Cline <jeremy@jcline.org>
2017-09-30 11:22:37 -04:00
Richard van der Hoff d5325d7ef1 Merge pull request #2480 from matrix-org/rav/federation_client_logging
Improve logging of failures in matrixfederationclient
2017-09-29 17:32:53 +01:00
Erik Johnston d5694ac5fa Only log if we've removed media 2017-09-28 16:08:08 +01:00
Richard van der Hoff e43de3ae4b Improve logging of failures in matrixfederationclient
* don't log exception types twice
* not all exceptions have a meaningful 'message'. Use the repr rather than
  attempting to build a string ourselves.
2017-09-28 15:38:09 +01:00
Richard van der Hoff 75e67b9ee4 Handle SERVFAILs when doing AAAA lookups for federation (#2477)
... to cope with people with broken dnssec setups, mostly
2017-09-28 15:24:00 +01:00
Erik Johnston 768f00dedb Up the limits on number of url cache entries to delete at one time 2017-09-28 14:27:27 +01:00
Erik Johnston 4dc07e93a8 Add old indices 2017-09-28 14:10:33 +01:00
Erik Johnston 7cc483aa0e Clear up expired url cache every 10s 2017-09-28 13:56:53 +01:00
Erik Johnston e1e7d76cf1 Actually assign result to variable 2017-09-28 13:55:29 +01:00
Erik Johnston 93247a424a Only pull out local media that were for url cache 2017-09-28 13:48:14 +01:00
Erik Johnston 5f501ec7e2 Fix typo in url cache expiry timer 2017-09-28 12:59:01 +01:00
Erik Johnston 761d255fdf Merge pull request #2479 from matrix-org/erikj/expire_url_cache_thumbnails
Support new and old style media id formats
2017-09-28 12:58:13 +01:00
Erik Johnston ace8079086 Support new and old style media id formats 2017-09-28 12:52:51 +01:00
Erik Johnston 7a44c01d89 Fix typo 2017-09-28 12:46:04 +01:00
Erik Johnston c9bc4b7031 Merge pull request #2478 from matrix-org/erikj/expire_url_cache_thumbnails
Delete expired url cache data
2017-09-28 12:42:33 +01:00
Erik Johnston ae79764fe5 Change expires column to expires_ts 2017-09-28 12:37:53 +01:00
Erik Johnston 77f1d24de3 More brackets 2017-09-28 12:23:15 +01:00
Erik Johnston 9ccb4226ba Delete expired url cache data 2017-09-28 12:18:06 +01:00
Erik Johnston bf86a41ef1 Merge pull request #2476 from matrix-org/erikj/joined_members_auth
Fix /joined_members to work with AS users
2017-09-28 10:44:44 +01:00
Erik Johnston 8090fd4664 Fix /joined_members to work with AS users 2017-09-28 10:09:32 +01:00
Erik Johnston 3a743f649c Merge pull request #2475 from matrix-org/erikj/joined_members_auth
Fix bug where /joined_members didn't check user was in room
2017-09-27 16:42:23 +01:00
Erik Johnston adec03395d Fix bug where /joined_members didn't check user was in room 2017-09-27 15:14:39 +01:00
David Baker 74e494b010 Merge pull request #2474 from matrix-org/dbkr/spam_check_module
Make the spam checker a module
2017-09-27 11:31:00 +01:00
David Baker ef3a5ae787 Don't test is spam_checker not None
Sometimes it's a Mock object which is not none but is still not
what we're after
2017-09-27 11:24:19 +01:00
David Baker 8c06dd6071 Remove unintentional debugging 2017-09-27 10:31:14 +01:00
David Baker 60c78666ab pep8 2017-09-27 10:26:13 +01:00
David Baker 1786b0e768 Forgot the new file again :( 2017-09-27 10:22:54 +01:00
David Baker 8ad5f34908 pep8 2017-09-26 19:21:41 +01:00
David Baker 6cd5fcd536 Make the spam checker a module 2017-09-26 19:20:23 +01:00
David Baker ccc67d445b Merge pull request #2473 from matrix-org/dbkr/factor_out_module_loading
Factor out module loading to a separate place
2017-09-26 18:12:17 +01:00
David Baker 9fd086e506 unnecessary parens 2017-09-26 17:59:46 +01:00
David Baker 0b03a97708 Add module_loader.py 2017-09-26 17:56:41 +01:00
David Baker 4824a33c31 Factor out module loading to a separate place
So it can be reused
2017-09-26 17:51:26 +01:00
Erik Johnston 1e5fcfd14a Merge pull request #2472 from matrix-org/erikj/groups_rooms
Add remove room from group API
2017-09-26 16:05:46 +01:00
Erik Johnston 17b8e2bd02 Add remove room API 2017-09-26 15:52:41 +01:00
Erik Johnston a8e2a3df32 Add unique index to group_rooms table 2017-09-26 15:39:21 +01:00
Erik Johnston 0d7c7fd907 Merge pull request #2471 from matrix-org/erikj/group_summary_publicised
Add is_publicised to group summary
2017-09-26 11:33:21 +01:00
Erik Johnston 95298783bb Add is_publicised to group summary 2017-09-26 11:04:37 +01:00
Erik Johnston 1a398b19fd Merge branch 'develop' of github.com:matrix-org/synapse into release-v0.23.0 2017-09-26 10:08:59 +01:00
Erik Johnston f4c8cd5e85 Bump changelog and version 2017-09-26 10:02:48 +01:00
Erik Johnston b8d832a08c Merge pull request #2470 from matrix-org/erikj/sync_speed_fix
Refactor to speed up incremental syncs
2017-09-25 17:43:14 +01:00
Erik Johnston e3edca3b5d Refactor to speed up incremental syncs 2017-09-25 17:35:39 +01:00
Richard van der Hoff cacfa04cb6 Merge pull request #2468 from maxidor/develop
Clarify recommended network setup
2017-09-25 16:37:33 +01:00
Max Dor e591f7b3f0 Include review feedback 2017-09-25 16:42:26 +02:00
Max Dor 7141f1a5cc Clarify recommended network setup 2017-09-25 16:20:23 +02:00
Erik Johnston 44edac0497 Merge branch 'release-v0.23.0' of github.com:matrix-org/synapse into develop 2017-09-25 14:52:46 +01:00
Richard van der Hoff 29e1c717c3 Merge pull request #2390 from r3dey3/develop
Fix iteration of requests_missing_keys; list doesn't have .values()
2017-09-25 11:56:01 +01:00
Richard van der Hoff 94133d7ce8 Merge branch 'develop' into develop 2017-09-25 11:50:11 +01:00
Erik Johnston b15c2b7971 Update CHANGES 2017-09-25 11:34:12 +01:00
Erik Johnston ba8fdc925c Bump version and changes 2017-09-25 11:01:31 +01:00
Richard van der Hoff 79b3cf3e02 Fix logcontxt leak in keyclient (#2465)
preserve_context_over_function doesn't do what you want it to do.
2017-09-25 09:51:39 +01:00
Richard van der Hoff b4fd710e1a Merge pull request #2464 from rnbdsh/patch-4
Remove non-existing files, add stop, use synctl
2017-09-25 09:33:22 +01:00
rnbdsh b68b0ede7a Start traditionally, stop synctl
Starting with synctl lead to "no config file found"
Stopping also leads to some (code=exited, status=1/FAILURE), but at least now we can stop the service.
2017-09-24 04:55:19 +02:00
rnbdsh 68f737702b Remove non-existing files, add stop, use synctl
Non-existing files, when running the suggested from https://github.com/matrix-org/synapse#configuring-synapse
/etc/synapse/log_config.yaml so the --log-config leads to an error
/etc/sysconfig/synapse The environment-file or even the /etc/sysconfig does not exist in arch linux

Also instead of calling python2 we use synctl, as this seems to be the proper way to start it, and it gives us a more useful error in the systemctl status. And we now allow stop (and therefore restart).
2017-09-24 04:26:23 +02:00
Richard van der Hoff f65e31d22f Do an AAAA lookup on SRV record targets (#2462)
Support SRV records which point at AAAA records, as well as A records.

Fixes https://github.com/matrix-org/synapse/issues/2405
2017-09-22 20:26:47 +01:00
Matthew Hodgson f496399ac4 fix thinko'd docstring 2017-09-22 15:34:14 +01:00
Erik Johnston 3166ed55b2 Fix device list when rejoining room (#2461) 2017-09-22 14:44:17 +01:00
Erik Johnston e1dec2f1a7 Remove user from group summary when the leave the group 2017-09-21 16:09:57 +01:00
Erik Johnston bb746a9de1 Revert: Keep room_id's in group summary 2017-09-21 15:57:22 +01:00
Erik Johnston ae8d4bb0f0 Keep room_id's in group summary 2017-09-21 15:55:18 +01:00
Richard van der Hoff c94ab5976a Merge pull request #2459 from matrix-org/rav/keyring_cleanups
Clean up Keyring code
2017-09-20 11:29:39 +01:00
Erik Johnston 197d82dc07 Correctly return next token 2017-09-20 11:12:11 +01:00
Erik Johnston 069ae2df12 Fix initial sync 2017-09-20 10:52:12 +01:00
Richard van der Hoff 6de74ea6d7 Fix logcontexts in _check_sigs_and_hashes 2017-09-20 01:32:42 +01:00
Richard van der Hoff 72472456d8 Add some more tests for Keyring 2017-09-20 01:32:42 +01:00
Richard van der Hoff c5c24c239b Fix logcontext handling in verify_json_objects_for_server
preserve_context_over_fn is essentially broken, because (a) it pointlessly
drops the current logcontext before calling its wrapped function, which means
we don't get any useful logcontexts for _handle_key_deferred; (b) it wraps the
resulting deferred in a _PreservingContextDeferred, which is very dangerous
because you then can't yield on it without leaking context back into the
reactor.

Instead, let's specify that the resultant deferreds call their callbacks with
no logcontext.
2017-09-20 01:32:42 +01:00
Richard van der Hoff c5b0e9f485 Turn _start_key_lookups into an inlineCallbacks function
... which means that logcontexts can be correctly preserved for the stuff it
does.

get_server_verify_keys is now called with the logcontext, so needs to
preserve_fn when it fires off its nested inlineCallbacks function.

Also renames get_server_verify_keys to reflect the fact it's meant to be
private.
2017-09-20 01:32:42 +01:00
Richard van der Hoff abdefb8a01 Fix potential race in _start_key_lookups
If the verify_request.deferred has already completed, then `remove_deferreds`
will be called immediately. It therefore might resolve the server_to_deferred
deferred while there are still other requests for that server in flight.

To avoid that, we should build the complete list of requests, and *then* add the
callbacks.
2017-09-20 01:32:42 +01:00
Richard van der Hoff afbd773dc6 Add some comments to _start_key_lookups 2017-09-20 01:32:42 +01:00
Richard van der Hoff 2a4b9ea233 Consistency for how verify_request.deferred is called
Define that it is run with no log context, and make sure that happens.

If we aren't careful to reset the logcontext, we can't bung the deferreds into
defer.gatherResults etc. We don't actually do that directly, but we *do*
resolve other deferreds from affected callbacks (notably the server_to_deferred
map in _start_key_lookups), and those *do* get passed into
defer.gatherResults. It turns out that this way ends up being least confusing.
2017-09-20 01:32:42 +01:00
Richard van der Hoff 3b98439eca Factor out _start_key_lookups
... to make it easier to see what's going on.
2017-09-20 01:32:42 +01:00
Richard van der Hoff fde63b880d Replace server_and_json with verify_requests
This is a precursor to factoring some of this code out.
2017-09-20 01:32:42 +01:00
Richard van der Hoff 2d511defd9 pull out handle_key_deferred to top level
There's no need for this to be a nested definition; pulling it out not only
makes it more efficient, but makes it easier to check that it's not accessing
any local variables it shouldn't be.
2017-09-20 01:32:42 +01:00
Richard van der Hoff dd1ea9763a Fix incorrect key_ids in error message 2017-09-20 01:32:42 +01:00
Richard van der Hoff e76d1135dd Invalidate signing key cache when we gat an update
This might make the cache slightly more efficient.
2017-09-20 01:32:42 +01:00
Richard van der Hoff fcf2c0fd1a Remove redundant preserve_fn
preserve_fn is a no-op unless the wrapped function returns a
Deferred. verify_json_objects_for_server returns a list, so this is doing
nothing.
2017-09-20 01:32:42 +01:00
Richard van der Hoff 9864efa532 Fix concurrent server_key requests (#2458)
Fix a bug where we could end up firing off multiple requests for server_keys
for the same server at the same time.
2017-09-19 23:25:44 +01:00
Richard van der Hoff aa620d09a0 Add a config option to block all room invites (#2457)
- allows sysadmins the ability to lock down their servers so that people can't
send their users room invites.
2017-09-19 16:08:14 +01:00
Richard van der Hoff 2eabdf3f98 add some comments to on_exchange_third_party_invite_request 2017-09-19 12:20:36 +01:00
Richard van der Hoff 5ed109d59f PoC for filtering spammy events (#2456)
Demonstration of how you might add some hooks to filter out spammy events.
2017-09-19 12:20:11 +01:00
Erik Johnston 47d9848dc4 Merge pull request #2454 from matrix-org/erikj/groups_sync_creator
Ensure that creator of group sees group down /sync
2017-09-19 11:21:26 +01:00
Erik Johnston 93e504d04e Ensure that creator of group sees group down /sync 2017-09-19 11:08:16 +01:00
Erik Johnston b5feaa5a49 Merge branch 'develop' of github.com:matrix-org/synapse into erikj/groups_merged 2017-09-19 11:07:45 +01:00
Richard van der Hoff 3f405b34e9 Fix overzealous kicking of guest users (#2453)
We should only kick guest users if the guest access event is authorised.
2017-09-19 08:52:52 +01:00
Richard van der Hoff 290777b3d9 Clean up and document handling of logcontexts in Keyring (#2452)
I'm still unclear on what the intended behaviour for
`verify_json_objects_for_server` is, but at least I now understand the
behaviour of most of the things it calls...
2017-09-18 18:31:01 +01:00
Erik Johnston 77c81ca6ea Merge pull request #2451 from matrix-org/erikj/add_state_to_timeline
Don't filter out current state events from timeline
2017-09-18 17:22:33 +01:00
Erik Johnston 2d1b7955ae Don't filter out current state events from timeline 2017-09-18 17:13:03 +01:00
David Baker 862c8da560 Merge pull request #2450 from matrix-org/dbkr/push_event_id_only
Add support for event_id_only push format
2017-09-18 16:41:29 +01:00
Erik Johnston 2d9f341c3e Merge pull request #2449 from matrix-org/erikj/rejoin_device_lists
Correctly handle leaving room in /key/changes
2017-09-18 15:59:13 +01:00
David Baker 436ee0a2ea Also include the room_id
as really it's part of the event ID
2017-09-18 15:58:38 +01:00
David Baker b393f5db51 Use .get - it's much shorter 2017-09-18 15:50:26 +01:00
David Baker a2562f9d74 Add support for event_id_only push format
Param in the data dict of a pusher that tells an HTTP pusher to
send just the event_id of the event it's notifying about and the
notification counts. For clients that want to go & fetch the body
of the event themselves anyway.
2017-09-18 15:39:39 +01:00
Erik Johnston d6dadd95ac Correctly handle leaving room in /key/changes 2017-09-18 15:38:22 +01:00
Erik Johnston 993d3f710b Merge pull request #2443 from matrix-org/erikj/rejoin_device_lists
Send down device list change notif when member leaves/rejoins room
2017-09-18 13:17:53 +01:00
Erik Johnston 4a94eb3ea4 Fix typo 2017-09-15 09:56:54 +01:00
Erik Johnston 3a0cee28d6 Actually hook leave notifs up 2017-09-14 11:49:37 +01:00
Erik Johnston 4f845a0713 Handle joining/leaving rooms in /keys/changes 2017-09-13 16:28:08 +01:00
Erik Johnston 473700f016 Get left rooms 2017-09-13 15:13:41 +01:00
Erik Johnston 9ce866ed4f In sync handle device lists for newly joined/left rooms 2017-09-12 16:44:26 +01:00
Erik Johnston 69ef4987a6 Add left section to /keys/changes 2017-09-08 14:44:36 +01:00
Erik Johnston 53cc8ad35a Send down device list change notif when member leaves/rejoins room 2017-09-07 15:08:39 +01:00
Richard van der Hoff e2fcba038c Merge pull request #2439 from matrix-org/rav/tox_tweaks
do tox install with pip -e
2017-09-06 17:08:19 +01:00
Richard van der Hoff 5f59f20636 Merge remote-tracking branch 'origin/master' into develop 2017-09-05 21:58:19 +01:00
Richard van der Hoff 59de2c7afa Exclude the github issue template from our sdist (#2440)
PR #2413 added an issue template, but just adding files to the project
directory upsets the packaging scripts: we need to explicitly include or
exclude them.

Move the template into a .github directory to make that easy, and to de-clutter
the root a bit.
2017-09-05 21:57:19 +01:00
Richard van der Hoff 4b616c8cf2 Merge branch 'master' into develop 2017-09-05 17:51:13 +01:00
Richard van der Hoff 4dd61df6f8 do tox install with pip -e
- this ensures we end up with a working virtualenv which we can use for other
things.
2017-09-05 16:35:23 +01:00
Erik Johnston c0c31656ff Merge pull request #2433 from ptman/patch-1
Document known to work postgres version
2017-09-01 15:28:42 +01:00
Paul Tötterman 8b16b43b7f Document known to work postgres version 2017-09-01 16:52:45 +03:00
Richard van der Hoff dff396de0f Set --python when running sytest
.. because I want to make the 'install_and_run' script useful for non-synapse
jobs, which do not accept --python. In any case we set up the path here, so
sytest shouldn't be guessing it.
2017-09-01 11:20:37 +01:00
Richard van der Hoff f06ffdb6fa fix python path in jenkins scripts 2017-09-01 10:31:45 +01:00
Richard van der Hoff 6e67aaa7f2 Set --python when running sytest
.. because I want to make the 'install_and_run' script useful for non-synapse
jobs, which do not accept --python. In any case we set up the path here, so
sytest shouldn't be guessing it.
2017-09-01 10:06:21 +01:00
Erik Johnston 7f0d0ba3bc Merge pull request #2430 from matrix-org/erikj/groups_profile_cache
Add user profiles to summary from group server
2017-08-25 16:48:58 +01:00
Erik Johnston 4a9b1cf253 Add user profiles to summary from group server 2017-08-25 16:23:58 +01:00
Erik Johnston 6d8799af1a Merge pull request #2429 from matrix-org/erikj/groups_profile_cache
Add a remote user profile cache
2017-08-25 15:52:42 +01:00
Erik Johnston 258409ef61 Fix typos and reinherit 2017-08-25 14:45:20 +01:00
Erik Johnston bf81f3cf2c Split out profile handler to fix tests 2017-08-25 14:34:56 +01:00
Erik Johnston 27ebc5c8f2 Add remote profile cache 2017-08-25 11:25:47 +01:00
Erik Johnston 97c544f91f Add _simple_update 2017-08-25 11:11:37 +01:00
Richard van der Hoff 934ab76835 Merge pull request #2428 from matrix-org/rav/update_upgrade
Tweaks to the upgrade instructions
2017-08-24 11:42:02 +01:00
Richard van der Hoff fc9878f6a4 Tweaks to the upgrade instructions 2017-08-23 15:27:02 +01:00
Richard van der Hoff a4d3bfe3d6 Merge pull request #2417 from matrix-org/rav/federation_client
Improvements to the federation test client
2017-08-23 14:50:26 +01:00
Richard van der Hoff a7effa8400 Merge pull request #2288 from kyrias/bcrypt
python_dependencies: Use bcrypt module instead of py-bcrypt
2017-08-23 14:14:56 +01:00
Richard van der Hoff a04c6bbf8f test federation client: Allow server-name and key-file as options
so that you don't necessarily need a config file.
2017-08-22 11:19:30 +01:00
Richard van der Hoff 77ea8cbdd7 Merge pull request #2416 from matrix-org/rav/prometheus_config
Add prometheus config
2017-08-22 10:34:40 +01:00
Erik Johnston 2800983f3e Merge pull request #2410 from matrix-org/erikj/groups_publicise
Add ability to publicise group membership
2017-08-21 16:51:56 +01:00
Erik Johnston 8b50fe5330 Use BOOLEAN rather than TEXT type 2017-08-21 16:37:29 +01:00
Erik Johnston 73b4e18c62 Merge pull request #2426 from matrix-org/erikj/groups_fix_sync
Groups: Fix mising json.load in initial sync
2017-08-21 16:37:14 +01:00
Tom Lant 20b3660495 Merge pull request #2413 from matrix-org/toml-issue-template
Issue template for Synapse
2017-08-21 16:07:35 +01:00
Erik Johnston 175a01f56c Groups: Fix mising json.load in initial sync 2017-08-21 14:45:56 +01:00
Richard van der Hoff 046b659ce2 Improvements to the federation test client
Make it read the config file, primarily.
2017-08-17 16:59:11 +01:00
Tom Lant 413c270723 Update ISSUE_TEMPLATE.md
Added instructions for checking server version.
2017-08-17 11:14:35 +01:00
Tom Lant ec3a2dc773 Update ISSUE_TEMPLATE.md
Responding to review comments.
2017-08-17 11:00:51 +01:00
Richard van der Hoff 012875258c Add prometheus config
... from https://github.com/matrix-org/synapse-prometheus-config.
2017-08-16 15:31:44 +01:00
Richard van der Hoff 692250c6be Fix user_dir startup
Add missing parameter to _base.start_worker_reactor
2017-08-16 15:11:29 +01:00
Richard van der Hoff d2352347cf Fix process startup
escape the % that got added in 92168cb so that the process starts up ok.
2017-08-16 14:57:35 +01:00
Matthew Hodgson 92168cbbc5 explain why CPU affinity is a good idea 2017-08-15 18:27:42 +01:00
Richard van der Hoff 963015005e Merge pull request #2415 from matrix-org/rav/synctl_cpu_affinity
Allow configuration of CPU affinity
2017-08-15 17:42:05 +01:00
Richard van der Hoff 10d8b701a1 Allow configuration of CPU affinity
Make it possible to set the CPU affinity in the config file, so that we don't
need to remember to do it manually every time.
2017-08-15 17:08:28 +01:00
Richard van der Hoff 543c794a76 Factor out common application start
We have 10 copies of this code, and I don't really want to update each one
separately.
2017-08-15 17:04:40 +01:00
Tom Lant 57cd0c3dea Update ISSUE_TEMPLATE.md
Removed the sentence encouraging people not to file a bug - if people are in doubt we'd rather they filed a bug than gave up entirely.
2017-08-14 14:40:32 +01:00
Tom Lant b524dd4c35 Update ISSUE_TEMPLATE.md
Oops capital L.
2017-08-14 14:36:49 +01:00
Tom Lant 09703609fc Create ISSUE_TEMPLATE.md
A new issue template proposed to try and steer people towards #matrix:matrix.org for support queries relating to running their own homeserver.
2017-08-14 14:35:25 +01:00
Erik Johnston ba3ff7918b Fixup 2017-08-11 13:42:42 +01:00
Erik Johnston ef8e578677 Add bulk group publicised lookup API 2017-08-09 13:36:22 +01:00
Erik Johnston b880ff190a Allow update group publicity 2017-08-08 14:19:41 +01:00
Erik Johnston 05e21285aa Store whether the user wants to publicise their membership of a group 2017-08-08 13:01:46 +01:00
hera eae04f1952 fix english 2017-08-04 23:56:42 +01:00
hera 5699b05072 typo 2017-08-04 23:44:37 +01:00
Erik Johnston a1e67bcb97 Remove stale TODO comments 2017-08-04 10:07:10 +01:00
Erik Johnston 09552f9d9c Reduce spammy log line in synchrotrons 2017-08-02 17:29:51 +01:00
Kenny Keslar f18373dc5d Fix iteration of requests_missing_keys; list doesn't have .values()
Signed-off-by: Kenny Keslar <r3dey3@r3dey3.com>
2017-07-26 22:44:19 -05:00
Erik Johnston ebbaae5526 Merge pull request #2382 from matrix-org/erikj/group_privilege
Include users membership in group in summary API
2017-07-24 18:09:12 +01:00
Erik Johnston 966a70f1fa Update comment 2017-07-24 17:49:39 +01:00
Erik Johnston 629cdfb124 Use join rather than joined, etc. 2017-07-24 14:54:05 +01:00
Erik Johnston ed666d3969 Fix all the typos 2017-07-24 14:05:09 +01:00
Erik Johnston b76ef6ccb8 Include users membership in group in summary API 2017-07-24 13:55:39 +01:00
Erik Johnston 851aeae7c7 Check users/rooms are in group before adding to summary 2017-07-24 13:40:56 +01:00
Erik Johnston d5e32c843f Correctly add joins to correct segment 2017-07-24 13:31:26 +01:00
Erik Johnston 96917d5552 Merge pull request #2378 from matrix-org/erikj/group_sync_support
Add groups to sync stream
2017-07-21 11:05:39 +01:00
Erik Johnston 0401604222 Merge pull request #2377 from matrix-org/erikj/group_profile_update
Add update group profile API
2017-07-20 17:53:39 +01:00
Erik Johnston b238cf7f6b Remove spurious content param 2017-07-20 17:49:55 +01:00
Erik Johnston 960dae3340 Add notifier 2017-07-20 17:14:44 +01:00
Erik Johnston 2cc998fed8 Fix replication. And notify 2017-07-20 17:13:18 +01:00
Erik Johnston 139fe30f47 Remember to cast to bool 2017-07-20 16:47:35 +01:00
Erik Johnston 4d793626ff Fix bug in generating current token 2017-07-20 16:42:44 +01:00
Erik Johnston c544188ee3 Add groups to sync stream 2017-07-20 16:36:42 +01:00
Erik Johnston 0ab153d201 Check values are strings 2017-07-20 16:24:18 +01:00
Erik Johnston 8209b5f033 Fix a storage desc 2017-07-20 16:22:22 +01:00
Erik Johnston b27429729d Merge pull request #2375 from matrix-org/erikj/port_script
Fix port script for user directory tables
2017-07-20 16:16:43 +01:00
Erik Johnston 60a9a49f83 Extend comment 2017-07-20 16:16:29 +01:00
Erik Johnston b3bf6a1218 Merge pull request #2374 from matrix-org/erikj/group_server_local
Add local group server support
2017-07-20 13:15:42 +01:00
Erik Johnston 57826d645b Fix typo 2017-07-20 13:15:22 +01:00
Erik Johnston d7d24750be Fix port script for user directory tables 2017-07-20 10:47:01 +01:00
Erik Johnston 6f443a74cf Add update group profile API 2017-07-20 09:46:33 +01:00
Erik Johnston 14a34f12d7 Comments 2017-07-18 17:28:42 +01:00
Erik Johnston 3431ec55dc Comments 2017-07-18 17:23:50 +01:00
Erik Johnston 6027b1992f Fix permissions 2017-07-18 16:51:25 +01:00
Erik Johnston e884ff31d8 Add DELETE 2017-07-18 16:41:44 +01:00
Erik Johnston 05c13f6c22 Add 'args' param to post_json 2017-07-18 16:40:21 +01:00
Erik Johnston 94ecd871a0 Fix typos 2017-07-18 16:38:54 +01:00
Erik Johnston 12ed4ee48e Correctly parse query params 2017-07-18 15:33:09 +01:00
Erik Johnston 332839f6ea Update federation client pokes 2017-07-18 14:45:37 +01:00
Erik Johnston e5ea6dd021 Add client apis 2017-07-18 14:37:06 +01:00
Erik Johnston cccfcfa7b9 Comments 2017-07-18 10:35:18 +01:00
Erik Johnston 68f34e85ce Use transport client directly 2017-07-18 10:29:57 +01:00
Erik Johnston 3e703eb04e Comment 2017-07-18 10:17:25 +01:00
Erik Johnston 508460f240 Remove sync stuff 2017-07-18 09:55:46 +01:00
Erik Johnston 6e9f147faa Add GroupID type 2017-07-18 09:47:25 +01:00
Erik Johnston 4540730111 Remove unused tables 2017-07-18 09:38:15 +01:00
Erik Johnston e96ee95a7e Remove sync stuff 2017-07-18 09:38:08 +01:00
Erik Johnston 2f9eafdd36 Add local group server support 2017-07-17 12:03:49 +01:00
Erik Johnston b3de67234e Merge pull request #2363 from matrix-org/erikj/group_server_summary
Add group summary APIs
2017-07-17 11:54:14 +01:00
Erik Johnston 514c2d3c4d Merge pull request #2371 from matrix-org/erikj/push_cache_hit
Increase cache hit ratio for push
2017-07-17 09:42:27 +01:00
Erik Johnston bfde076022 Increase cache hit ratio for push
We don't update the cache in all code paths, which causes subsequent
calls to miss the cache
2017-07-14 16:11:26 +01:00
Erik Johnston cb3aee8219 Ensure category and role ids are non-null 2017-07-14 14:06:55 +01:00
Erik Johnston 85fda57208 Add DEFAULT_ROLE_ID 2017-07-14 14:03:54 +01:00
Erik Johnston 4b203bdba5 Correctly increment orders 2017-07-14 14:02:31 +01:00
Erik Johnston d3862812ff Merge pull request #2366 from matrix-org/erikj/push_metrics
Add more metrics to push rule evaluation
2017-07-14 11:04:03 +01:00
Erik Johnston 8d26385d76 Add more metrics to push rule evaluation 2017-07-13 14:37:30 +01:00
Erik Johnston 3b0470dba5 Remove unused functions 2017-07-13 13:54:01 +01:00
Erik Johnston 8575e3160f Comments 2017-07-13 13:52:41 +01:00
Erik Johnston 67b7b904ba Merge pull request #2365 from matrix-org/erikj/push_skip_lock
Push: Don't acquire lock unless necessary
2017-07-13 11:44:48 +01:00
Erik Johnston f60218ec41 Push: Don't acquire lock unless necessary 2017-07-13 11:23:53 +01:00
Erik Johnston a78cda4baf Remove TODO 2017-07-13 11:17:07 +01:00
Erik Johnston 7a39da8cc6 Add summary APIs to federation 2017-07-13 11:13:19 +01:00
Erik Johnston 5bbb53580a raise NotImplementedError 2017-07-13 10:25:29 +01:00
Erik Johnston 26451a09eb Comments 2017-07-12 14:47:18 +01:00
Erik Johnston 8d55877c9e Simplify checking if admin 2017-07-12 11:43:39 +01:00
Erik Johnston a62406aaa5 Add group summary APIs 2017-07-12 11:36:15 +01:00
Erik Johnston 91818723a1 Merge pull request #2362 from matrix-org/erikj/sync_user_users_who_share
Use less DB for device list handling in sync
2017-07-12 10:45:30 +01:00
Erik Johnston e9aec001f4 Use less DB for device list handling in sync 2017-07-12 10:30:10 +01:00
Erik Johnston 28e8c46f29 Merge pull request #2352 from matrix-org/erikj/group_server_split
Initial Group Server
2017-07-12 10:26:29 +01:00
Erik Johnston 6d586dc05c Comment 2017-07-12 09:58:37 +01:00
Erik Johnston 410b4e14a1 Move comment 2017-07-11 15:44:18 +01:00
Erik Johnston fe4e885f54 Add federation API for adding room to group 2017-07-11 14:35:07 +01:00
Erik Johnston bbb739d24a Comment 2017-07-11 14:31:36 +01:00
Erik Johnston 26752df503 Typo 2017-07-11 14:29:03 +01:00
Erik Johnston e52c391cd4 Rename column to attestation_json 2017-07-11 14:25:46 +01:00
Erik Johnston 0aac30d53b Comments 2017-07-11 14:23:50 +01:00
Erik Johnston 0184a97dbd Merge pull request #2354 from krombel/reduce_static_sync_reply
encode sync-response statically
2017-07-11 14:19:56 +01:00
Krombel 85b9f76f1d split out reducing stuff; just make encode_* static 2017-07-11 13:14:35 +02:00
Erik Johnston 6322fbbd41 Comment 2017-07-11 11:52:03 +01:00
Erik Johnston 8ba89f1050 Remove u/ requirement 2017-07-11 11:45:32 +01:00
Erik Johnston 429925a5e9 Lift out visibility parsing 2017-07-11 11:44:08 +01:00
Erik Johnston 83936293eb Comments 2017-07-11 11:42:25 +01:00
Erik Johnston e2cb760dcc Merge pull request #2357 from matrix-org/erikj/push
Don't compute push actions for backfilled events
2017-07-11 10:53:22 +01:00
Erik Johnston 925b3638ff Reduce log levels in tcp replication 2017-07-11 10:04:21 +01:00
Erik Johnston 9a6fd3ef29 Don't compute push actions for backfilled events 2017-07-11 10:02:21 +01:00
Krombel 2f82de18ee fix test 2017-07-10 17:34:58 +02:00
Erik Johnston b8ca494ee9 Initial group server implementation 2017-07-10 15:44:15 +01:00
Krombel 6e16aca8b0 encode sync-response statically; omit empty objects from sync-response 2017-07-10 16:42:17 +02:00
Erik Johnston d4d12daed9 Include registration and as stores in frontend proxy 2017-07-07 18:36:45 +01:00
Erik Johnston f467a8f66d Merge branch 'master' of github.com:matrix-org/synapse into develop 2017-07-07 18:26:28 +01:00
Erik Johnston c9184ed87e Merge pull request #2344 from matrix-org/erikj/frontend_proxy
Add a frontend proxy
2017-07-07 18:25:46 +01:00
Erik Johnston 1fc4a962e4 Add a frontend proxy 2017-07-07 18:19:46 +01:00
Erik Johnston 08284c86ed Merge pull request #2343 from matrix-org/erikj/fastpush
Perf: Don't filter events for push
2017-07-07 14:25:42 +01:00
Erik Johnston f502b0dea1 Perf: Don't filter events for push
We know the users are joined and we can explicitly check for if they are
ignoring the user, so lets do that.
2017-07-07 14:04:40 +01:00
Erik Johnston 1200f28d66 Merge branch 'hotfixes-v0.22.1' of github.com:matrix-org/synapse 2017-07-06 18:11:49 +01:00
Erik Johnston 76ed3476d3 Bump version and changelog 2017-07-06 18:11:22 +01:00
Erik Johnston 58dc1f2c78 Merge pull request #2342 from matrix-org/erikj/pusher_pool_instantiate
Fix bug where pusherpool didn't start and broke some rooms
2017-07-06 18:08:43 +01:00
Erik Johnston 5a7f561a9b Fix bug where pusherpool didn't start and broke some rooms
Since we didn't instansiate the PusherPool at start time it could fail
at run time, which it did for some users.

This may or may not fix things for those users, but it should happen at
start time and stop the server from starting.
2017-07-06 17:55:51 +01:00
Erik Johnston ed9a7f5436 Merge pull request #2309 from matrix-org/erikj/user_ip_repl
Fix up user_ip replication commands
2017-07-06 14:33:14 +01:00
Erik Johnston 1f64207f26 Merge branch 'master' of github.com:matrix-org/synapse into develop 2017-07-06 13:57:45 +01:00
Erik Johnston 42b50483be Merge branch 'release-v0.22.0' of github.com:matrix-org/synapse 2017-07-06 10:36:25 +01:00
Erik Johnston 6264cf9666 Bump version and changelog 2017-07-06 10:35:56 +01:00
Erik Johnston f386632800 Merge pull request #2334 from matrix-org/erikj/refactor_transport_server
Separate federation servlet into different lists
2017-07-05 17:09:07 +01:00
Erik Johnston 5e49a57ecc Separate federation servlet into different lists 2017-07-05 14:32:24 +01:00
Richard van der Hoff 3d31b39297 Merge pull request #2332 from matrix-org/rav/fix_pushes
Fix caching error in the push evaluator
2017-07-05 11:10:53 +01:00
Richard van der Hoff 73cfe48031 Fix caching error in the push evaluator
Initialising `result` to `{}` in the parameters meant that every call to
_flatten_dict used the *same* target dictionary.

I'm hopeful this will fix https://github.com/matrix-org/synapse/issues/2270,
but I suspect it won't. (This code seems to have been here since forever,
unlike the bug, and I don't really think it explains the observed
behaviour). Still, it makes it hard to investigate the problem.
2017-07-05 00:28:43 +01:00
Erik Johnston 05538587ef Bump version and changelog 2017-07-04 14:02:21 +01:00
Erik Johnston f92d7416d7 Merge pull request #2330 from matrix-org/erikj/cache_size_factor
Increase default cache size
2017-07-04 10:51:21 +01:00
Mark Haines 1f12d808e7 Merge pull request #2323 from matrix-org/markjh/invite_checks
Improve the error handling for bad invites received over federation
2017-07-04 10:50:43 +01:00
Erik Johnston 29a4066a4d Update test 2017-07-04 10:21:25 +01:00
Erik Johnston 7afb4e3f54 Update README 2017-07-04 10:00:52 +01:00
Erik Johnston 495f075b41 Increase default cache factor size. 2017-07-04 09:58:32 +01:00
Erik Johnston b5e8d529e6 Define CACHE_SIZE_FACTOR once 2017-07-04 09:56:44 +01:00
Mark Haines 3e279411fe Improve the error handling for bad invites received over federation 2017-06-30 16:20:30 +01:00
Erik Johnston 47574c9cba Merge pull request #2321 from matrix-org/erikj/prefill_forward
Prefill forward extrems and event to state groups
2017-06-30 11:03:04 +01:00
Erik Johnston 6ff14ddd2e Make into list 2017-06-29 15:47:37 +01:00
Erik Johnston 5946aa0877 Prefill forward extrems and event to state groups 2017-06-29 15:38:48 +01:00
Erik Johnston d800ab2847 Merge pull request #2320 from matrix-org/erikj/cache_macaroon_parse
Cache macaroon parse and validation
2017-06-29 15:06:43 +01:00
Erik Johnston 2c365f4723 Cache macaroon parse and validation
Turns out this can be quite expensive for requests, and is easily
cachable. We don't cache the lookup to the DB so invalidation still
works.
2017-06-29 14:50:18 +01:00
Erik Johnston a1a253ea50 Merge pull request #2319 from matrix-org/erikj/prune_sessions
Use an ExpiringCache for storing registration sessions
2017-06-29 14:20:24 +01:00
Erik Johnston c72058bcc6 Use an ExpiringCache for storing registration sessions
This is because pruning them was a significant performance drain on
matrix.org
2017-06-29 14:08:37 +01:00
Erik Johnston 27f26e48b7 Serialize user ip command as json 2017-06-27 16:25:38 +01:00
Erik Johnston 8c23221666 Fix up 2017-06-27 15:53:45 +01:00
Erik Johnston 731f3c37a0 Merge branch 'release-v0.22.0' of github.com:matrix-org/synapse into develop 2017-06-27 15:41:34 +01:00
Erik Johnston 4b444723f0 Merge pull request #2308 from matrix-org/erikj/user_ip_repl
Make workers report to master for user ip updates
2017-06-27 15:36:47 +01:00
Erik Johnston 816605a137 Merge pull request #2307 from matrix-org/erikj/user_ip_batch
Batch upsert user ips
2017-06-27 15:08:32 +01:00
Erik Johnston 78cefd78d6 Make workers report to master for user ip updates 2017-06-27 14:58:10 +01:00
Erik Johnston a0a561ae85 Fix up client ips to read from pending data 2017-06-27 14:46:12 +01:00
Erik Johnston ed3d0170d9 Batch upsert user ips 2017-06-27 13:37:04 +01:00
Erik Johnston 976128f368 Update version and changelog 2017-06-26 16:14:56 +01:00
Erik Johnston d04d672a80 Merge pull request #2290 from matrix-org/erikj/ensure_round_trip
Reject local events that don't round trip the DB
2017-06-26 15:12:02 +01:00
Erik Johnston 036f439f53 Merge pull request #2304 from matrix-org/erikj/users_share_fix
Fix up indices for users_who_share_rooms
2017-06-26 15:11:39 +01:00
Erik Johnston 1bce3e6b35 Remove unused variables 2017-06-26 14:03:27 +01:00
Erik Johnston e3cbec10c1 Merge branch 'develop' of github.com:matrix-org/synapse into erikj/ensure_round_trip 2017-06-26 14:02:44 +01:00
Erik Johnston 8abdd7b553 Fix up indices for users_who_share_rooms 2017-06-26 14:01:30 +01:00
Erik Johnston ff13c5e7af Merge pull request #2301 from xwiki-labs/push-redact-content
Add configuration parameter to allow redaction of content from push m…
2017-06-24 13:13:51 +01:00
Caleb James DeLisle 27bd0b9a91 Change the config file generator to more descriptive explanation of push.redact_content 2017-06-24 10:32:12 +02:00
Caleb James DeLisle bce144595c Fix TravisCI tests for PR #2301 - Fat finger mistake 2017-06-23 15:26:09 +02:00
Caleb James DeLisle 75eba3b07d Fix TravisCI tests for PR #2301 2017-06-23 15:15:18 +02:00
Caleb James DeLisle 1591eddaea Add configuration parameter to allow redaction of content from push messages for google/apple devices 2017-06-23 13:01:04 +02:00
Erik Johnston 4fec80ba6f Merge pull request #2299 from matrix-org/erikj/segregate_url_cache_downloads
Store URL cache preview downloads separately
2017-06-23 12:00:45 +01:00
Erik Johnston 7fe8ed1787 Store URL cache preview downloads seperately
This makes it easier to clear old media out at a later date
2017-06-23 11:14:11 +01:00
Erik Johnston e204062310 Merge pull request #2297 from matrix-org/erikj/user_dir_fix
Fix thinko in initial public room user spam
2017-06-22 15:14:47 +01:00
Erik Johnston 44c722931b Make some more params configurable 2017-06-22 14:59:52 +01:00
Erik Johnston 2d520a9826 Typo. ARGH. 2017-06-22 14:42:39 +01:00
Erik Johnston 24d894e2e2 Fix thinko in unhandled user spam 2017-06-22 14:39:05 +01:00
Matthew Hodgson ccfcef6b59 Merge branch 'master' into develop 2017-06-22 13:03:44 +01:00
Erik Johnston e0004aa28a Add desc 2017-06-22 10:03:48 +01:00
Erik Johnston b668112320 Merge pull request #2296 from matrix-org/erikj/dont_appserver_shar
Don't work out users who share room with appservice users
2017-06-21 14:50:24 +01:00
Erik Johnston dae9a00a28 Initialise exclusive_user_regex 2017-06-21 14:19:33 +01:00
Erik Johnston 71995e1397 Merge pull request #2219 from krombel/avoid_duplicate_filters
only add new filter when not existent prevoisly
2017-06-21 14:11:26 +01:00
Erik Johnston 8177563ebe Fix for workers 2017-06-21 13:57:49 +01:00
Krombel 4202fba82a Merge branch 'develop' into avoid_duplicate_filters 2017-06-21 14:48:21 +02:00
Krombel 812c030e87 replaced json.dumps with encode_canonical_json 2017-06-21 14:48:12 +02:00
Erik Johnston 1217c7da91 Don't work out users who share room with appservice users 2017-06-21 12:00:41 +01:00
Erik Johnston 7d69f2d956 Merge pull request #2292 from matrix-org/erikj/quarantine_media
Add API to quarantine media
2017-06-19 18:15:00 +01:00
Erik Johnston 385dcb7c60 Handle thumbnail urls 2017-06-19 17:48:28 +01:00
Erik Johnston b8b936a6ea Add API to quarantine media 2017-06-19 17:39:21 +01:00
Erik Johnston b5f665de32 Merge pull request #2291 from matrix-org/erikj/shutdown_room
Add shutdown room API
2017-06-19 16:20:55 +01:00
Erik Johnston e5ae386ea4 Handle all cases of sending membership events 2017-06-19 16:07:54 +01:00
Erik Johnston 36e51aad3c Remove unused import 2017-06-19 14:42:21 +01:00
Erik Johnston b490299a3b Change to create new room and join other users 2017-06-19 14:10:13 +01:00
Erik Johnston 5db7070dd1 Forget room 2017-06-19 12:40:29 +01:00
Erik Johnston d7fe6b356c Add shutdown room API 2017-06-19 12:37:27 +01:00
Erik Johnston fcf01dd88e Reject local events that don't round trip the DB 2017-06-19 11:33:40 +01:00
Johannes Löthberg 4f66312df8 python_dependencies: Use bcrypt module instead of py-bcrypt
py-bcrypt has been unmaintained for a long while, while bcrypt is
actively maintained. And since ff8b87118d
we're compatible with the bcrypt anyway.

Signed-off-by: Johannes Löthberg <johannes@kyriasis.com>
2017-06-17 17:39:35 +02:00
Matthew 3fafb7b189 add missing boolean to synapse_port_db 2017-06-16 20:51:19 +01:00
Matthew 776a070421 fix synapse_port script 2017-06-16 20:24:14 +01:00
Erik Johnston dfeca6cf40 Merge pull request #2286 from matrix-org/erikj/split_out_user_dir
Split out user directory to a separate process
2017-06-16 13:01:19 +01:00
Erik Johnston 6aa5bc8635 Initial worker impl 2017-06-16 11:47:11 +01:00
Erik Johnston d8f47d2efa Merge pull request #2280 from matrix-org/erikj/share_room_user_dir
Include users who you share a room with in user directory
2017-06-16 11:06:00 +01:00
Erik Johnston 0a9315bbc7 Merge pull request #2285 from krombel/allow_authorization_header
allow Authorization header
2017-06-16 10:41:58 +01:00
Krombel 1ff419d343 allow Authorization header which handling got implemented in #1098
Signed-off-by: Matthias Kesler <krombel@krombel.de>
2017-06-16 11:21:14 +02:00
Erik Johnston 24df576795 Merge pull request #2282 from matrix-org/release-v0.21.1
Release v0.21.1
2017-06-15 13:24:16 +01:00
Erik Johnston fdf1ca30f0 Bump version and changelog 2017-06-15 12:59:06 +01:00
Erik Johnston 052c5d19d5 Merge pull request #2281 from matrix-org/erikj/phone_home_stats
Fix phone home stats
2017-06-15 12:46:23 +01:00
Erik Johnston 5ddd199870 Typo 2017-06-15 10:49:10 +01:00
Erik Johnston a9d6fa8b2b Include users who share room with requester in user directory 2017-06-15 10:17:21 +01:00
Erik Johnston 4564b05483 Implement updating users who share rooms on the fly 2017-06-15 10:17:17 +01:00
Erik Johnston 72613bc379 Implement initial population of users who share rooms table 2017-06-15 09:59:04 +01:00
Erik Johnston ebcd55d641 Add DB schema for tracking users who share rooms 2017-06-15 09:45:48 +01:00
Erik Johnston 4b461a6931 Add some more stats 2017-06-15 09:39:39 +01:00
Erik Johnston 93e7a38370 Remove unhelpful test 2017-06-15 09:30:54 +01:00
Erik Johnston 617304b2cf Fix phone home stats 2017-06-14 19:47:15 +01:00
Matthew Hodgson ba502fb89a add notes on running out of FDs 2017-06-14 02:23:14 +01:00
Erik Johnston 6c6b9689bb Merge pull request #2279 from matrix-org/erikj/fix_user_dir
Fix user directory insertion due to missing room_id
2017-06-13 12:48:50 +01:00
Erik Johnston d9fd937e39 Fix user directory insertion due to missing room_id 2017-06-13 11:50:24 +01:00
Erik Johnston fe9dc522d4 Merge pull request #2278 from matrix-org/erikj/fix_user_dir
Fix user dir to not assume existence of user
2017-06-13 11:38:44 +01:00
Erik Johnston 505e7e8b9d Fix up sql 2017-06-13 11:19:18 +01:00
Erik Johnston 6fd7e6db3d Fix user dir to not assume existence of user 2017-06-13 11:11:26 +01:00
Erik Johnston fdca6e36ee Merge pull request #2274 from matrix-org/erikj/cache_is_host_joined
Add cache for is_host_joined
2017-06-13 10:58:53 +01:00
Erik Johnston 90ae0cffec Merge pull request #2275 from matrix-org/erikj/tweark_user_directory_search
Tweak the ranking of PG user dir search
2017-06-13 10:58:43 +01:00
Erik Johnston de4cb50ca6 Merge pull request #2276 from matrix-org/erikj/fix_user_di
Don't assume existence of events when updating user directory
2017-06-13 10:55:41 +01:00
Erik Johnston a09e09ce76 Merge pull request #2277 from matrix-org/erikj/media
Throw exception when not retrying when downloading media
2017-06-13 10:52:45 +01:00
Erik Johnston 48d2949416 Throw exception when not retrying when downloading media 2017-06-13 10:23:14 +01:00
Erik Johnston 6ae8373d40 Don't assume existance of events when updating user directory 2017-06-13 10:19:26 +01:00
Erik Johnston b58e24cc3c Tweak the ranking of PG user dir search 2017-06-13 10:16:31 +01:00
Erik Johnston d53fe399eb Add cache for is_host_joined 2017-06-13 09:56:18 +01:00
Erik Johnston a837765e8c Merge pull request #2266 from matrix-org/erikj/host_in_room
Change is_host_joined to use current_state table
2017-06-12 09:49:51 +01:00
Erik Johnston f540b494a4 Merge pull request #2269 from matrix-org/erikj/cache_state_delta
Cache state deltas
2017-06-09 18:32:04 +01:00
Erik Johnston 8060974344 Fix replication 2017-06-09 16:40:52 +01:00
Erik Johnston b0d975e216 Comments 2017-06-09 16:25:42 +01:00
Erik Johnston e54d7d536e Cache state deltas 2017-06-09 16:24:00 +01:00
Erik Johnston 1e9b4d5a95 Merge pull request #2268 from matrix-org/erikj/entity_has_changed
Fix has_any_entity_changed
2017-06-09 15:30:55 +01:00
Erik Johnston efc2b7db95 Rewrite conditional 2017-06-09 13:35:15 +01:00
Erik Johnston bfd68019c2 Merge pull request #2267 from matrix-org/erikj/missing_notifier
Fix removing of pushers when using workers
2017-06-09 13:07:29 +01:00
Erik Johnston 1946867bc2 Merge pull request #2265 from matrix-org/erikj/remote_leave_outlier
Mark remote invite rejections as outliers
2017-06-09 13:05:15 +01:00
Erik Johnston 1664948e41 Comment 2017-06-09 13:05:05 +01:00
Erik Johnston 935e588799 Tweak SQL 2017-06-09 13:01:23 +01:00
Erik Johnston eed59dcc1e Fix has_any_entity_changed
Occaisonally has_any_entity_changed would throw the error: "Set changed
size during iteration" when taking the max of the `sorteddict`. While
its uncertain how that happens, its quite inefficient to iterate over
the entire dict anyway so we change to using the more traditional
`bisect_*` functions.
2017-06-09 11:44:01 +01:00
Erik Johnston 2cac7623a5 Add missing notifier 2017-06-09 11:24:41 +01:00
Erik Johnston 298d83b340 Fix replication 2017-06-09 11:01:28 +01:00
Erik Johnston 0185b75381 Change is_host_joined to use current_state table
This bypasses a bug where using the state groups to figure out if a host
is in a room sometimes errors if the servers isn't in the room. (For
example when the server rejected an invite to a remote room)
2017-06-09 10:52:26 +01:00
Erik Johnston 7132e5cdff Mark remote invite rejections as outliers 2017-06-09 10:08:18 +01:00
Erik Johnston 98bdb4468b Merge pull request #2263 from matrix-org/erikj/fix_state_woes
Ensure we don't use unpersisted state group as prev group
2017-06-08 12:56:18 +01:00
Erik Johnston ea11ee09f3 Ensure we don't use unpersisted state group as prev group 2017-06-08 11:59:57 +01:00
Erik Johnston c62c480dc6 Merge pull request #2259 from matrix-org/erikj/fix_state_woes
Fix bug where state_group tables got corrupted
2017-06-07 17:51:25 +01:00
Erik Johnston 197bd126f0 Fix bug where state_group tables got corrupted
This is due to the fact that we prefilled caches using txn.call_after,
which always gets called including on error.

We fix this by making txn.call_after only fire when a transaction
completes successfully, which is what we want most of the time anyway.
2017-06-07 17:39:36 +01:00
Erik Johnston f45f07ab86 Merge pull request #2258 from matrix-org/erikj/user_dir
Don't start user_directory handling on workers
2017-06-07 14:04:50 +01:00
Erik Johnston a053ff3979 Merge pull request #2248 from matrix-org/erikj/state_fixup
Faster cache for get_joined_hosts
2017-06-07 14:01:06 +01:00
Erik Johnston ecdd2a3658 Don't start user_directory handling on workers 2017-06-07 12:02:53 +01:00
Erik Johnston 2f34ad31ac Add some logging to user directory 2017-06-07 11:50:44 +01:00
Erik Johnston 671f0afa1d Merge pull request #2256 from matrix-org/erikj/faster_device_updates
Split up device_lists_outbound_pokes table for faster updates.
2017-06-07 11:48:00 +01:00
Erik Johnston 64ed74c01e When pruning, delete from device_lists_outbound_last_success 2017-06-07 11:20:47 +01:00
Erik Johnston 1a81a1898e Keep pruning background task 2017-06-07 11:16:56 +01:00
Erik Johnston 6ba21bf2b8 Comments 2017-06-07 11:08:36 +01:00
Erik Johnston 09e4bc0501 Merge branch 'develop' of github.com:matrix-org/synapse into erikj/state_fixup 2017-06-07 11:05:23 +01:00
Erik Johnston 6e2a7ee1bc Remove spurious log lines 2017-06-07 11:05:17 +01:00
Erik Johnston 65f0513a33 Split up device_lists_outbound_pokes table for faster updates. 2017-06-07 11:02:38 +01:00
Erik Johnston 6f83c4537c Increase size of IP cache 2017-06-07 10:18:44 +01:00
Erik Johnston cca94272fa Fix typo when getting app name 2017-06-06 11:50:07 +01:00
Erik Johnston 66b121b2fc Fix wrong number of arguments 2017-06-06 11:46:38 +01:00
Erik Johnston 8d34120a53 Merge pull request #2253 from matrix-org/erikj/user_dir
Handle profile updates in user directory
2017-06-01 17:33:20 +01:00
Erik Johnston 1a01af079e Handle profile updates in user directory 2017-06-01 15:39:51 +01:00
Erik Johnston 87e5e05aea Merge pull request #2252 from matrix-org/erikj/user_dir
Add a user directory
2017-06-01 15:39:32 +01:00
Erik Johnston 4d039aa2ca Fix sqlite 2017-06-01 14:58:48 +01:00
Erik Johnston 21e255a8f1 Split the table in two 2017-06-01 14:50:46 +01:00
Erik Johnston d5477c7afd Tweak search query 2017-06-01 13:28:01 +01:00
Erik Johnston 02a6108235 Tweak search query 2017-06-01 13:16:40 +01:00
Erik Johnston 7233341eac Comments 2017-06-01 13:11:38 +01:00
Erik Johnston 8be6fd95a3 Check if host is still in room 2017-06-01 13:05:39 +01:00
Erik Johnston 59dbb47065 Remove spurious inlineCallbacks 2017-06-01 11:41:29 +01:00
Erik Johnston 9c7db2491b Fix removing users 2017-06-01 11:36:50 +01:00
Erik Johnston 0fe6f3c521 Bug fixes and logging
- Check if room is public when a user joins before adding to user dir
- Fix typo of field name "content.join_rules" -> "content.join_rule"
2017-06-01 11:09:49 +01:00
Erik Johnston 036362ede6 Order by if they have profile info 2017-06-01 09:41:08 +01:00
Erik Johnston a757dd4863 Use prefix matching 2017-06-01 09:40:37 +01:00
Erik Johnston f5cc22bdc6 Comment on why arbitrary comments 2017-05-31 17:30:26 +01:00
Erik Johnston 5dd1b2c525 Use unique indices 2017-05-31 17:29:12 +01:00
Erik Johnston cc7609aa9f Comment briefly on how we keep user_directory up to date 2017-05-31 17:11:18 +01:00
Erik Johnston f1378aef91 Convert to int 2017-05-31 17:03:08 +01:00
Erik Johnston b2d8d07109 Lifts things into separate function 2017-05-31 17:00:24 +01:00
Erik Johnston f9791498ae Typos 2017-05-31 16:50:57 +01:00
Erik Johnston f091061711 Fix tests 2017-05-31 16:34:40 +01:00
Erik Johnston 4abcff0177 Fix typo 2017-05-31 16:22:36 +01:00
Erik Johnston 63c58c2a3f Limit number of things we fetch out of the db 2017-05-31 16:17:58 +01:00
Erik Johnston 304880d185 Add stream change cache 2017-05-31 15:46:36 +01:00
Erik Johnston 5d79d728f5 Split out directory and search tables 2017-05-31 15:23:49 +01:00
Erik Johnston dc51af3d03 Pull max id from correct table 2017-05-31 15:13:49 +01:00
Erik Johnston 350622a107 Handle the server leaving a public room 2017-05-31 15:11:36 +01:00
Erik Johnston 63fda37e20 Add comments 2017-05-31 15:00:29 +01:00
Erik Johnston 293ef29655 Weight differently 2017-05-31 14:29:32 +01:00
Erik Johnston 535c99f157 Use POST 2017-05-31 14:15:45 +01:00
Erik Johnston 45a5df5914 Add REST API 2017-05-31 14:11:55 +01:00
Erik Johnston 3b5f22ca40 Add search 2017-05-31 14:00:01 +01:00
Erik Johnston b5db4ed5f6 Update room column when room becomes unpublic 2017-05-31 13:40:28 +01:00
Erik Johnston 168524543f Add call later 2017-05-31 11:59:36 +01:00
Erik Johnston 3e123b8497 Start later 2017-05-31 11:56:27 +01:00
Erik Johnston 42137efde7 Don't go round in circles 2017-05-31 11:55:13 +01:00
Erik Johnston eeb2f9e546 Add user_directory to database 2017-05-31 11:51:01 +01:00
Erik Johnston 5dbaa520a5 Merge pull request #2251 from matrix-org/erikj/current_state_delta_stream
Add current_state_delta_stream table
2017-05-30 15:06:17 +01:00
Erik Johnston dd48f7204c Add comment 2017-05-30 15:01:22 +01:00
Erik Johnston 04095f7581 Add clobbered event_id 2017-05-30 14:53:01 +01:00
Erik Johnston a584a81b3e Add current_state_delta_stream table 2017-05-30 14:44:09 +01:00
Erik Johnston 619e8ecd0c Handle None state group correctly 2017-05-26 10:46:03 +01:00
Erik Johnston 23da638360 Fix typing tests 2017-05-26 10:02:04 +01:00
Erik Johnston dfbda5e025 Faster cache for get_joined_hosts 2017-05-25 17:24:44 +01:00
Erik Johnston 2b03751c3c Don't return weird prev_group 2017-05-25 14:47:39 +01:00
Erik Johnston dbc0dfd2d5 Remove unused options 2017-05-25 14:28:34 +01:00
Erik Johnston 11f139a647 Merge pull request #2247 from matrix-org/erikj/auth_event
Only store event_auth for state events
2017-05-24 16:46:34 +01:00
Erik Johnston 6e614e9e10 Add background task to clear out old event_auth 2017-05-24 15:23:34 +01:00
Erik Johnston c049472b8a Only store event_auth for state events 2017-05-24 15:23:31 +01:00
Erik Johnston 9a804b2812 Merge pull request #2243 from matrix-org/matthew/fix-url-preview-length-again
actually trim oversize og:description meta
2017-05-23 13:26:28 +01:00
Erik Johnston fbbc40f385 Merge pull request #2237 from matrix-org/erikj/sync_key_count
Add count of one time keys to sync stream
2017-05-23 11:18:13 +01:00
Erik Johnston 8cf9f0a3e7 Remove redundant invalidation 2017-05-23 09:46:59 +01:00
Erik Johnston e6618ece2d Missed an invalidation 2017-05-23 09:36:52 +01:00
Erik Johnston 58c4720293 Merge pull request #2242 from matrix-org/erikj/email_refactor
Only load jinja2 templates once
2017-05-23 09:34:08 +01:00
Matthew Hodgson 836d5c44b6 actually trim oversize og:description meta 2017-05-22 21:14:20 +01:00
Erik Johnston 11c2a3655f Only load jinja2 templates once
Instead of every time a new email pusher is created, as loading jinja2
templates is slow.
2017-05-22 17:48:58 +01:00
Erik Johnston f85a415279 Add missing storage function to slave store 2017-05-22 16:31:24 +01:00
Erik Johnston 74bf4ee7bf Stream count_e2e_one_time_keys cache invalidation 2017-05-22 16:19:22 +01:00
Erik Johnston b4f59c7e27 Add count of one time keys to sync stream 2017-05-19 15:47:55 +01:00
Krombel 55af207321 Merge branch 'develop' into avoid_duplicate_filters 2017-05-16 15:29:59 +02:00
Krombel 64953c8ed2 avoid access-error if no filter_id matches 2017-05-15 18:36:37 +02:00
Krombel eb7cbf27bc insert whitespace to fix travis build 2017-05-12 12:09:42 +02:00
Krombel 6b95e35e96 add check to only add a new filter if the same filter does not exist previously
Signed-off-by: Matthias Kesler <krombel@krombel.de>
2017-05-11 16:05:30 +02:00
252 changed files with 16020 additions and 3418 deletions
+47
View File
@@ -0,0 +1,47 @@
<!--
**IF YOU HAVE SUPPORT QUESTIONS ABOUT RUNNING OR CONFIGURING YOUR OWN HOME SERVER**:
You will likely get better support more quickly if you ask in ** #matrix:matrix.org ** ;)
This is a bug report template. By following the instructions below and
filling out the sections with your information, you will help the us to get all
the necessary data to fix your issue.
You can also preview your report before submitting it. You may remove sections
that aren't relevant to your particular case.
Text between <!-- and --> marks will be invisible in the report.
-->
### Description
Describe here the problem that you are experiencing, or the feature you are requesting.
### Steps to reproduce
- For bugs, list the steps
- that reproduce the bug
- using hyphens as bullet points
Describe how what happens differs from what you expected.
If you can identify any relevant log snippets from _homeserver.log_, please include
those here (please be careful to remove any personal or private data):
### Version information
<!-- IMPORTANT: please answer the following questions, to help us narrow down the problem -->
- **Homeserver**: Was this issue identified on matrix.org or another homeserver?
If not matrix.org:
- **Version**: What version of Synapse is running? <!--
You can find the Synapse version by inspecting the server headers (replace matrix.org with
your own homeserver domain):
$ curl -v https://matrix.org/_matrix/client/versions 2>&1 | grep "Server:"
-->
- **Install method**: package manager/git clone/pip
- **Platform**: Tell us about the environment in which your homeserver is operating
- distro, hardware, if it's running in a vm/container, etc.
+243
View File
@@ -1,3 +1,246 @@
Changes in synapse v0.25.1 (2017-11-17)
=======================================
Bug fixes:
* Fix login with LDAP and other password provider modules (PR #2678). Thanks to
@jkolo!
Changes in synapse v0.25.0 (2017-11-15)
=======================================
Bug fixes:
* Fix port script (PR #2673)
Changes in synapse v0.25.0-rc1 (2017-11-14)
===========================================
Features:
* Add is_public to groups table to allow for private groups (PR #2582)
* Add a route for determining who you are (PR #2668) Thanks to @turt2live!
* Add more features to the password providers (PR #2608, #2610, #2620, #2622,
#2623, #2624, #2626, #2628, #2629)
* Add a hook for custom rest endpoints (PR #2627)
* Add API to update group room visibility (PR #2651)
Changes:
* Ignore <noscript> tags when generating URL preview descriptions (PR #2576)
Thanks to @maximevaillancourt!
* Register some /unstable endpoints in /r0 as well (PR #2579) Thanks to
@krombel!
* Support /keys/upload on /r0 as well as /unstable (PR #2585)
* Front-end proxy: pass through auth header (PR #2586)
* Allow ASes to deactivate their own users (PR #2589)
* Remove refresh tokens (PR #2613)
* Automatically set default displayname on register (PR #2617)
* Log login requests (PR #2618)
* Always return `is_public` in the `/groups/:group_id/rooms` API (PR #2630)
* Avoid no-op media deletes (PR #2637) Thanks to @spantaleev!
* Fix various embarrassing typos around user_directory and add some doc. (PR
#2643)
* Return whether a user is an admin within a group (PR #2647)
* Namespace visibility options for groups (PR #2657)
* Downcase UserIDs on registration (PR #2662)
* Cache failures when fetching URL previews (PR #2669)
Bug fixes:
* Fix port script (PR #2577)
* Fix error when running synapse with no logfile (PR #2581)
* Fix UI auth when deleting devices (PR #2591)
* Fix typo when checking if user is invited to group (PR #2599)
* Fix the port script to drop NUL values in all tables (PR #2611)
* Fix appservices being backlogged and not receiving new events due to a bug in
notify_interested_services (PR #2631) Thanks to @xyzz!
* Fix updating rooms avatar/display name when modified by admin (PR #2636)
Thanks to @farialima!
* Fix bug in state group storage (PR #2649)
* Fix 500 on invalid utf-8 in request (PR #2663)
Changes in synapse v0.24.1 (2017-10-24)
=======================================
Bug fixes:
* Fix updating group profiles over federation (PR #2567)
Changes in synapse v0.24.0 (2017-10-23)
=======================================
No changes since v0.24.0-rc1
Changes in synapse v0.24.0-rc1 (2017-10-19)
===========================================
Features:
* Add Group Server (PR #2352, #2363, #2374, #2377, #2378, #2382, #2410, #2426,
#2430, #2454, #2471, #2472, #2544)
* Add support for channel notifications (PR #2501)
* Add basic implementation of backup media store (PR #2538)
* Add config option to auto-join new users to rooms (PR #2545)
Changes:
* Make the spam checker a module (PR #2474)
* Delete expired url cache data (PR #2478)
* Ignore incoming events for rooms that we have left (PR #2490)
* Allow spam checker to reject invites too (PR #2492)
* Add room creation checks to spam checker (PR #2495)
* Spam checking: add the invitee to user_may_invite (PR #2502)
* Process events from federation for different rooms in parallel (PR #2520)
* Allow error strings from spam checker (PR #2531)
* Improve error handling for missing files in config (PR #2551)
Bug fixes:
* Fix handling SERVFAILs when doing AAAA lookups for federation (PR #2477)
* Fix incompatibility with newer versions of ujson (PR #2483) Thanks to
@jeremycline!
* Fix notification keywords that start/end with non-word chars (PR #2500)
* Fix stack overflow and logcontexts from linearizer (PR #2532)
* Fix 500 error when fields missing from power_levels event (PR #2552)
* Fix 500 error when we get an error handling a PDU (PR #2553)
Changes in synapse v0.23.1 (2017-10-02)
=======================================
Changes:
* Make 'affinity' package optional, as it is not supported on some platforms
Changes in synapse v0.23.0 (2017-10-02)
=======================================
No changes since v0.23.0-rc2
Changes in synapse v0.23.0-rc2 (2017-09-26)
===========================================
Bug fixes:
* Fix regression in performance of syncs (PR #2470)
Changes in synapse v0.23.0-rc1 (2017-09-25)
===========================================
Features:
* Add a frontend proxy worker (PR #2344)
* Add support for event_id_only push format (PR #2450)
* Add a PoC for filtering spammy events (PR #2456)
* Add a config option to block all room invites (PR #2457)
Changes:
* Use bcrypt module instead of py-bcrypt (PR #2288) Thanks to @kyrias!
* Improve performance of generating push notifications (PR #2343, #2357, #2365,
#2366, #2371)
* Improve DB performance for device list handling in sync (PR #2362)
* Include a sample prometheus config (PR #2416)
* Document known to work postgres version (PR #2433) Thanks to @ptman!
Bug fixes:
* Fix caching error in the push evaluator (PR #2332)
* Fix bug where pusherpool didn't start and broke some rooms (PR #2342)
* Fix port script for user directory tables (PR #2375)
* Fix device lists notifications when user rejoins a room (PR #2443, #2449)
* Fix sync to always send down current state events in timeline (PR #2451)
* Fix bug where guest users were incorrectly kicked (PR #2453)
* Fix bug talking to IPv6 only servers using SRV records (PR #2462)
Changes in synapse v0.22.1 (2017-07-06)
=======================================
Bug fixes:
* Fix bug where pusher pool didn't start and caused issues when
interacting with some rooms (PR #2342)
Changes in synapse v0.22.0 (2017-07-06)
=======================================
No changes since v0.22.0-rc2
Changes in synapse v0.22.0-rc2 (2017-07-04)
===========================================
Changes:
* Improve performance of storing user IPs (PR #2307, #2308)
* Slightly improve performance of verifying access tokens (PR #2320)
* Slightly improve performance of event persistence (PR #2321)
* Increase default cache factor size from 0.1 to 0.5 (PR #2330)
Bug fixes:
* Fix bug with storing registration sessions that caused frequent CPU churn
(PR #2319)
Changes in synapse v0.22.0-rc1 (2017-06-26)
===========================================
Features:
* Add a user directory API (PR #2252, and many more)
* Add shutdown room API to remove room from local server (PR #2291)
* Add API to quarantine media (PR #2292)
* Add new config option to not send event contents to push servers (PR #2301)
Thanks to @cjdelisle!
Changes:
* Various performance fixes (PR #2177, #2233, #2230, #2238, #2248, #2256,
#2274)
* Deduplicate sync filters (PR #2219) Thanks to @krombel!
* Correct a typo in UPGRADE.rst (PR #2231) Thanks to @aaronraimist!
* Add count of one time keys to sync stream (PR #2237)
* Only store event_auth for state events (PR #2247)
* Store URL cache preview downloads separately (PR #2299)
Bug fixes:
* Fix users not getting notifications when AS listened to that user_id (PR
#2216) Thanks to @slipeer!
* Fix users without push set up not getting notifications after joining rooms
(PR #2236)
* Fix preview url API to trim long descriptions (PR #2243)
* Fix bug where we used cached but unpersisted state group as prev group,
resulting in broken state of restart (PR #2263)
* Fix removing of pushers when using workers (PR #2267)
* Fix CORS headers to allow Authorization header (PR #2285) Thanks to @krombel!
Changes in synapse v0.21.1 (2017-06-15)
=======================================
Bug fixes:
* Fix bug in anonymous usage statistic reporting (PR #2281)
Changes in synapse v0.21.0 (2017-05-18)
=======================================
+1
View File
@@ -27,4 +27,5 @@ exclude jenkins*.sh
exclude jenkins*
recursive-exclude jenkins *.sh
prune .github
prune demo/etc
+51 -21
View File
@@ -200,11 +200,11 @@ different. See `the spec`__ for more information on key management.)
.. __: `key_management`_
The default configuration exposes two HTTP ports: 8008 and 8448. Port 8008 is
configured without TLS; it is not recommended this be exposed outside your
local network. Port 8448 is configured to use TLS with a self-signed
certificate. This is fine for testing with but, to avoid your clients
complaining about the certificate, you will almost certainly want to use
another certificate for production purposes. (Note that a self-signed
configured without TLS; it should be behind a reverse proxy for TLS/SSL
termination on port 443 which in turn should be used for clients. Port 8448
is configured to use TLS with a self-signed certificate. If you would like
to do initial test with a client without having to setup a reverse proxy,
you can temporarly use another certificate. (Note that a self-signed
certificate is fine for `Federation`_). You can do so by changing
``tls_certificate_path``, ``tls_private_key_path`` and ``tls_dh_params_path``
in ``homeserver.yaml``; alternatively, you can use a reverse-proxy, but be sure
@@ -283,10 +283,16 @@ Connecting to Synapse from a client
The easiest way to try out your new Synapse installation is by connecting to it
from a web client. The easiest option is probably the one at
http://riot.im/app. You will need to specify a "Custom server" when you log on
or register: set this to ``https://localhost:8448`` - remember to specify the
port (``:8448``) unless you changed the configuration. (Leave the identity
or register: set this to ``https://domain.tld`` if you setup a reverse proxy
following the recommended setup, or ``https://localhost:8448`` - remember to specify the
port (``:8448``) if not ``:443`` unless you changed the configuration. (Leave the identity
server as the default - see `Identity servers`_.)
If using port 8448 you will run into errors until you accept the self-signed
certificate. You can easily do this by going to ``https://localhost:8448``
directly with your browser and accept the presented certificate. You can then
go back in your web client and proceed further.
If all goes well you should at least be able to log in, create a room, and
start sending messages.
@@ -359,7 +365,7 @@ https://www.archlinux.org/packages/community/any/matrix-synapse/, which should p
the necessary dependencies. If the default web client is to be served (enabled by default in
the generated config),
https://www.archlinux.org/packages/community/any/python2-matrix-angular-sdk/ will also need to
be installed.
be installed.
Alternatively, to install using pip a few changes may be needed as ArchLinux
defaults to python 3, but synapse currently assumes python 2.7 by default:
@@ -528,6 +534,30 @@ fix try re-installing from PyPI or directly from
# Install from github
pip install --user https://github.com/pyca/pynacl/tarball/master
Running out of File Handles
~~~~~~~~~~~~~~~~~~~~~~~~~~~
If synapse runs out of filehandles, it typically fails badly - live-locking
at 100% CPU, and/or failing to accept new TCP connections (blocking the
connecting client). Matrix currently can legitimately use a lot of file handles,
thanks to busy rooms like #matrix:matrix.org containing hundreds of participating
servers. The first time a server talks in a room it will try to connect
simultaneously to all participating servers, which could exhaust the available
file descriptors between DNS queries & HTTPS sockets, especially if DNS is slow
to respond. (We need to improve the routing algorithm used to be better than
full mesh, but as of June 2017 this hasn't happened yet).
If you hit this failure mode, we recommend increasing the maximum number of
open file handles to be at least 4096 (assuming a default of 1024 or 256).
This is typically done by editing ``/etc/security/limits.conf``
Separately, Synapse may leak file handles if inbound HTTP requests get stuck
during processing - e.g. blocked behind a lock or talking to a remote server etc.
This is best diagnosed by matching up the 'Received request' and 'Processed request'
log lines and looking for any 'Processed request' lines which take more than
a few seconds to execute. Please let us know at #matrix-dev:matrix.org if
you see this failure mode so we can help debug it, however.
ArchLinux
~~~~~~~~~
@@ -569,8 +599,9 @@ you to run your server on a machine that might not have the same name as your
domain name. For example, you might want to run your server at
``synapse.example.com``, but have your Matrix user-ids look like
``@user:example.com``. (A SRV record also allows you to change the port from
the default 8448. However, if you are thinking of using a reverse-proxy, be
sure to read `Reverse-proxying the federation port`_ first.)
the default 8448. However, if you are thinking of using a reverse-proxy on the
federation port, which is not recommended, be sure to read
`Reverse-proxying the federation port`_ first.)
To use a SRV record, first create your SRV record and publish it in DNS. This
should have the format ``_matrix._tcp.<yourdomain.com> <ttl> IN SRV 10 0 <port>
@@ -650,7 +681,7 @@ For information on how to install and use PostgreSQL, please see
Using a reverse proxy with Synapse
==================================
It is possible to put a reverse proxy such as
It is recommended to put a reverse proxy such as
`nginx <https://nginx.org/en/docs/http/ngx_http_proxy_module.html>`_,
`Apache <https://httpd.apache.org/docs/current/mod/mod_proxy_http.html>`_ or
`HAProxy <http://www.haproxy.org/>`_ in front of Synapse. One advantage of
@@ -668,9 +699,9 @@ federation port has a number of pitfalls. It is possible, but be sure to read
`Reverse-proxying the federation port`_.
The recommended setup is therefore to configure your reverse-proxy on port 443
for client connections, but to also expose port 8448 for server-server
connections. All the Matrix endpoints begin ``/_matrix``, so an example nginx
configuration might look like::
to port 8008 of synapse for client connections, but to also directly expose port
8448 for server-server connections. All the Matrix endpoints begin ``/_matrix``,
so an example nginx configuration might look like::
server {
listen 443 ssl;
@@ -792,7 +823,9 @@ spidering 'internal' URLs on your network. At the very least we recommend that
your loopback and RFC1918 IP addresses are blacklisted.
This also requires the optional lxml and netaddr python dependencies to be
installed.
installed. This in turn requires the libxml2 library to be available - on
Debian/Ubuntu this means ``apt-get install libxml2-dev``, or equivalent for
your OS.
Password reset
@@ -875,12 +908,9 @@ cache a lot of recent room data and metadata in RAM in order to speed up
common requests. We'll improve this in future, but for now the easiest
way to either reduce the RAM usage (at the risk of slowing things down)
is to set the almost-undocumented ``SYNAPSE_CACHE_FACTOR`` environment
variable. Roughly speaking, a SYNAPSE_CACHE_FACTOR of 1.0 will max out
at around 3-4GB of resident memory - this is what we currently run the
matrix.org on. The default setting is currently 0.1, which is probably
around a ~700MB footprint. You can dial it down further to 0.02 if
desired, which targets roughly ~512MB. Conversely you can dial it up if
you need performance for lots of users and have a box with a lot of RAM.
variable. The default is 0.5, which can be decreased to reduce RAM usage
in memory constrained enviroments, or increased if performance starts to
degrade.
.. _`key_management`: https://matrix.org/docs/spec/server_server/unstable.html#retrieving-server-keys
+51 -42
View File
@@ -5,39 +5,48 @@ Before upgrading check if any special steps are required to upgrade from the
what you currently have installed to current version of synapse. The extra
instructions that may be required are listed later in this document.
If synapse was installed in a virtualenv then active that virtualenv before
upgrading. If synapse is installed in a virtualenv in ``~/.synapse/`` then run:
1. If synapse was installed in a virtualenv then active that virtualenv before
upgrading. If synapse is installed in a virtualenv in ``~/.synapse/`` then
run:
.. code:: bash
source ~/.synapse/bin/activate
2. If synapse was installed using pip then upgrade to the latest version by
running:
.. code:: bash
pip install --upgrade --process-dependency-links https://github.com/matrix-org/synapse/tarball/master
# restart synapse
synctl restart
If synapse was installed using git then upgrade to the latest version by
running:
.. code:: bash
# Pull the latest version of the master branch.
git pull
# Update the versions of synapse's python dependencies.
python synapse/python_dependencies.py | xargs pip install --upgrade
# restart synapse
./synctl restart
To check whether your update was sucessful, you can check the Server header
returned by the Client-Server API:
.. code:: bash
source ~/.synapse/bin/activate
If synapse was installed using pip then upgrade to the latest version by
running:
.. code:: bash
pip install --upgrade --process-dependency-links https://github.com/matrix-org/synapse/tarball/master
If synapse was installed using git then upgrade to the latest version by
running:
.. code:: bash
# Pull the latest version of the master branch.
git pull
# Update the versions of synapse's python dependencies.
python synapse/python_dependencies.py | xargs -n1 pip install --upgrade
To check whether your update was sucessfull, run:
.. code:: bash
# replace your.server.domain with ther domain of your synapse homeserver
curl https://<your.server.domain>/_matrix/federation/v1/version
So for the Matrix.org HS server the URL would be: https://matrix.org/_matrix/federation/v1/version.
# replace <host.name> with the hostname of your synapse homeserver.
# You may need to specify a port (eg, :8448) if your server is not
# configured on port 443.
curl -kv https://<host.name>/_matrix/client/versions 2>&1 | grep "Server:"
Upgrading to v0.15.0
====================
@@ -77,7 +86,7 @@ It has been replaced by specifying a list of application service registrations i
``homeserver.yaml``::
app_service_config_files: ["registration-01.yaml", "registration-02.yaml"]
Where ``registration-01.yaml`` looks like::
url: <String> # e.g. "https://my.application.service.com"
@@ -166,7 +175,7 @@ This release completely changes the database schema and so requires upgrading
it before starting the new version of the homeserver.
The script "database-prepare-for-0.5.0.sh" should be used to upgrade the
database. This will save all user information, such as logins and profiles,
database. This will save all user information, such as logins and profiles,
but will otherwise purge the database. This includes messages, which
rooms the home server was a member of and room alias mappings.
@@ -175,18 +184,18 @@ file and ask for help in #matrix:matrix.org. The upgrade process is,
unfortunately, non trivial and requires human intervention to resolve any
resulting conflicts during the upgrade process.
Before running the command the homeserver should be first completely
Before running the command the homeserver should be first completely
shutdown. To run it, simply specify the location of the database, e.g.:
./scripts/database-prepare-for-0.5.0.sh "homeserver.db"
Once this has successfully completed it will be safe to restart the
homeserver. You may notice that the homeserver takes a few seconds longer to
Once this has successfully completed it will be safe to restart the
homeserver. You may notice that the homeserver takes a few seconds longer to
restart than usual as it reinitializes the database.
On startup of the new version, users can either rejoin remote rooms using room
aliases or by being reinvited. Alternatively, if any other homeserver sends a
message to a room that the homeserver was previously in the local HS will
message to a room that the homeserver was previously in the local HS will
automatically rejoin the room.
Upgrading to v0.4.0
@@ -245,7 +254,7 @@ automatically generate default config use::
--config-path homeserver.config \
--generate-config
This config can be edited if desired, for example to specify a different SSL
This config can be edited if desired, for example to specify a different SSL
certificate to use. Once done you can run the home server using::
$ python synapse/app/homeserver.py --config-path homeserver.config
@@ -266,20 +275,20 @@ This release completely changes the database schema and so requires upgrading
it before starting the new version of the homeserver.
The script "database-prepare-for-0.0.1.sh" should be used to upgrade the
database. This will save all user information, such as logins and profiles,
database. This will save all user information, such as logins and profiles,
but will otherwise purge the database. This includes messages, which
rooms the home server was a member of and room alias mappings.
Before running the command the homeserver should be first completely
Before running the command the homeserver should be first completely
shutdown. To run it, simply specify the location of the database, e.g.:
./scripts/database-prepare-for-0.0.1.sh "homeserver.db"
Once this has successfully completed it will be safe to restart the
homeserver. You may notice that the homeserver takes a few seconds longer to
Once this has successfully completed it will be safe to restart the
homeserver. You may notice that the homeserver takes a few seconds longer to
restart than usual as it reinitializes the database.
On startup of the new version, users can either rejoin remote rooms using room
aliases or by being reinvited. Alternatively, if any other homeserver sends a
message to a room that the homeserver was previously in the local HS will
message to a room that the homeserver was previously in the local HS will
automatically rejoin the room.
+37
View File
@@ -0,0 +1,37 @@
This directory contains some sample monitoring config for using the
'Prometheus' monitoring server against synapse.
To use it, first install prometheus by following the instructions at
http://prometheus.io/
### for Prometheus v1
Add a new job to the main prometheus.conf file:
job: {
name: "synapse"
target_group: {
target: "http://SERVER.LOCATION.HERE:PORT/_synapse/metrics"
}
}
### for Prometheus v2
Add a new job to the main prometheus.yml file:
- job_name: "synapse"
metrics_path: "/_synapse/metrics"
# when endpoint uses https:
scheme: "https"
static_configs:
- targets: ['SERVER.LOCATION:PORT']
To use `synapse.rules` add
rule_files:
- "/PATH/TO/synapse-v2.rules"
Metrics are disabled by default when running synapse; they must be enabled
with the 'enable-metrics' option, either in the synapse config file or as a
command-line option.
+395
View File
@@ -0,0 +1,395 @@
{{ template "head" . }}
{{ template "prom_content_head" . }}
<h1>System Resources</h1>
<h3>CPU</h3>
<div id="process_resource_utime"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#process_resource_utime"),
expr: "rate(process_cpu_seconds_total[2m]) * 100",
name: "[[job]]",
min: 0,
max: 100,
renderer: "line",
height: 150,
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "%",
yTitle: "CPU Usage"
})
</script>
<h3>Memory</h3>
<div id="process_resource_maxrss"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#process_resource_maxrss"),
expr: "process_psutil_rss:max",
name: "Maxrss",
min: 0,
renderer: "line",
height: 150,
yAxisFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yHoverFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yUnits: "bytes",
yTitle: "Usage"
})
</script>
<h3>File descriptors</h3>
<div id="process_fds"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#process_fds"),
expr: "process_open_fds{job='synapse'}",
name: "FDs",
min: 0,
renderer: "line",
height: 150,
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "",
yTitle: "Descriptors"
})
</script>
<h1>Reactor</h1>
<h3>Total reactor time</h3>
<div id="reactor_total_time"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#reactor_total_time"),
expr: "rate(python_twisted_reactor_tick_time:total[2m]) / 1000",
name: "time",
max: 1,
min: 0,
renderer: "area",
height: 150,
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "s/s",
yTitle: "Usage"
})
</script>
<h3>Average reactor tick time</h3>
<div id="reactor_average_time"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#reactor_average_time"),
expr: "rate(python_twisted_reactor_tick_time:total[2m]) / rate(python_twisted_reactor_tick_time:count[2m]) / 1000",
name: "time",
min: 0,
renderer: "line",
height: 150,
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "s",
yTitle: "Time"
})
</script>
<h3>Pending calls per tick</h3>
<div id="reactor_pending_calls"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#reactor_pending_calls"),
expr: "rate(python_twisted_reactor_pending_calls:total[30s])/rate(python_twisted_reactor_pending_calls:count[30s])",
name: "calls",
min: 0,
renderer: "line",
height: 150,
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yTitle: "Pending Cals"
})
</script>
<h1>Storage</h1>
<h3>Queries</h3>
<div id="synapse_storage_query_time"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_storage_query_time"),
expr: "rate(synapse_storage_query_time:count[2m])",
name: "[[verb]]",
yAxisFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yHoverFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yUnits: "queries/s",
yTitle: "Queries"
})
</script>
<h3>Transactions</h3>
<div id="synapse_storage_transaction_time"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_storage_transaction_time"),
expr: "rate(synapse_storage_transaction_time:count[2m])",
name: "[[desc]]",
min: 0,
yAxisFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yHoverFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yUnits: "txn/s",
yTitle: "Transactions"
})
</script>
<h3>Transaction execution time</h3>
<div id="synapse_storage_transactions_time_msec"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_storage_transactions_time_msec"),
expr: "rate(synapse_storage_transaction_time:total[2m]) / 1000",
name: "[[desc]]",
min: 0,
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "s/s",
yTitle: "Usage"
})
</script>
<h3>Database scheduling latency</h3>
<div id="synapse_storage_schedule_time"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_storage_schedule_time"),
expr: "rate(synapse_storage_schedule_time:total[2m]) / 1000",
name: "Total latency",
min: 0,
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "s/s",
yTitle: "Usage"
})
</script>
<h3>Cache hit ratio</h3>
<div id="synapse_cache_ratio"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_cache_ratio"),
expr: "rate(synapse_util_caches_cache:total[2m]) * 100",
name: "[[name]]",
min: 0,
max: 100,
yAxisFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yHoverFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yUnits: "%",
yTitle: "Percentage"
})
</script>
<h3>Cache size</h3>
<div id="synapse_cache_size"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_cache_size"),
expr: "synapse_util_caches_cache:size",
name: "[[name]]",
yAxisFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yHoverFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yUnits: "",
yTitle: "Items"
})
</script>
<h1>Requests</h1>
<h3>Requests by Servlet</h3>
<div id="synapse_http_server_requests_servlet"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_http_server_requests_servlet"),
expr: "rate(synapse_http_server_requests:servlet[2m])",
name: "[[servlet]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "req/s",
yTitle: "Requests"
})
</script>
<h4>&nbsp;(without <tt>EventStreamRestServlet</tt> or <tt>SyncRestServlet</tt>)</h4>
<div id="synapse_http_server_requests_servlet_minus_events"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_http_server_requests_servlet_minus_events"),
expr: "rate(synapse_http_server_requests:servlet{servlet!=\"EventStreamRestServlet\", servlet!=\"SyncRestServlet\"}[2m])",
name: "[[servlet]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "req/s",
yTitle: "Requests"
})
</script>
<h3>Average response times</h3>
<div id="synapse_http_server_response_time_avg"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_http_server_response_time_avg"),
expr: "rate(synapse_http_server_response_time:total[2m]) / rate(synapse_http_server_response_time:count[2m]) / 1000",
name: "[[servlet]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "s/req",
yTitle: "Response time"
})
</script>
<h3>All responses by code</h3>
<div id="synapse_http_server_responses"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_http_server_responses"),
expr: "rate(synapse_http_server_responses[2m])",
name: "[[method]] / [[code]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "req/s",
yTitle: "Requests"
})
</script>
<h3>Error responses by code</h3>
<div id="synapse_http_server_responses_err"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_http_server_responses_err"),
expr: "rate(synapse_http_server_responses{code=~\"[45]..\"}[2m])",
name: "[[method]] / [[code]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "req/s",
yTitle: "Requests"
})
</script>
<h3>CPU Usage</h3>
<div id="synapse_http_server_response_ru_utime"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_http_server_response_ru_utime"),
expr: "rate(synapse_http_server_response_ru_utime:total[2m])",
name: "[[servlet]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "s/s",
yTitle: "CPU Usage"
})
</script>
<h3>DB Usage</h3>
<div id="synapse_http_server_response_db_txn_duration"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_http_server_response_db_txn_duration"),
expr: "rate(synapse_http_server_response_db_txn_duration:total[2m])",
name: "[[servlet]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "s/s",
yTitle: "DB Usage"
})
</script>
<h3>Average event send times</h3>
<div id="synapse_http_server_send_time_avg"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_http_server_send_time_avg"),
expr: "rate(synapse_http_server_response_time:total{servlet='RoomSendEventRestServlet'}[2m]) / rate(synapse_http_server_response_time:count{servlet='RoomSendEventRestServlet'}[2m]) / 1000",
name: "[[servlet]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "s/req",
yTitle: "Response time"
})
</script>
<h1>Federation</h1>
<h3>Sent Messages</h3>
<div id="synapse_federation_client_sent"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_federation_client_sent"),
expr: "rate(synapse_federation_client_sent[2m])",
name: "[[type]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "req/s",
yTitle: "Requests"
})
</script>
<h3>Received Messages</h3>
<div id="synapse_federation_server_received"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_federation_server_received"),
expr: "rate(synapse_federation_server_received[2m])",
name: "[[type]]",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "req/s",
yTitle: "Requests"
})
</script>
<h3>Pending</h3>
<div id="synapse_federation_transaction_queue_pending"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_federation_transaction_queue_pending"),
expr: "synapse_federation_transaction_queue_pending",
name: "[[type]]",
yAxisFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yHoverFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yUnits: "",
yTitle: "Units"
})
</script>
<h1>Clients</h1>
<h3>Notifiers</h3>
<div id="synapse_notifier_listeners"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_notifier_listeners"),
expr: "synapse_notifier_listeners",
name: "listeners",
min: 0,
yAxisFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yHoverFormatter: PromConsole.NumberFormatter.humanizeNoSmallPrefix,
yUnits: "",
yTitle: "Listeners"
})
</script>
<h3>Notified Events</h3>
<div id="synapse_notifier_notified_events"></div>
<script>
new PromConsole.Graph({
node: document.querySelector("#synapse_notifier_notified_events"),
expr: "rate(synapse_notifier_notified_events[2m])",
name: "events",
yAxisFormatter: PromConsole.NumberFormatter.humanize,
yHoverFormatter: PromConsole.NumberFormatter.humanize,
yUnits: "events/s",
yTitle: "Event rate"
})
</script>
{{ template "prom_content_tail" . }}
{{ template "tail" }}
+21
View File
@@ -0,0 +1,21 @@
synapse_federation_transaction_queue_pendingEdus:total = sum(synapse_federation_transaction_queue_pendingEdus or absent(synapse_federation_transaction_queue_pendingEdus)*0)
synapse_federation_transaction_queue_pendingPdus:total = sum(synapse_federation_transaction_queue_pendingPdus or absent(synapse_federation_transaction_queue_pendingPdus)*0)
synapse_http_server_requests:method{servlet=""} = sum(synapse_http_server_requests) by (method)
synapse_http_server_requests:servlet{method=""} = sum(synapse_http_server_requests) by (servlet)
synapse_http_server_requests:total{servlet=""} = sum(synapse_http_server_requests:by_method) by (servlet)
synapse_cache:hit_ratio_5m = rate(synapse_util_caches_cache:hits[5m]) / rate(synapse_util_caches_cache:total[5m])
synapse_cache:hit_ratio_30s = rate(synapse_util_caches_cache:hits[30s]) / rate(synapse_util_caches_cache:total[30s])
synapse_federation_client_sent{type="EDU"} = synapse_federation_client_sent_edus + 0
synapse_federation_client_sent{type="PDU"} = synapse_federation_client_sent_pdu_destinations:count + 0
synapse_federation_client_sent{type="Query"} = sum(synapse_federation_client_sent_queries) by (job)
synapse_federation_server_received{type="EDU"} = synapse_federation_server_received_edus + 0
synapse_federation_server_received{type="PDU"} = synapse_federation_server_received_pdus + 0
synapse_federation_server_received{type="Query"} = sum(synapse_federation_server_received_queries) by (job)
synapse_federation_transaction_queue_pending{type="EDU"} = synapse_federation_transaction_queue_pending_edus + 0
synapse_federation_transaction_queue_pending{type="PDU"} = synapse_federation_transaction_queue_pending_pdus + 0
+60
View File
@@ -0,0 +1,60 @@
groups:
- name: synapse
rules:
- record: "synapse_federation_transaction_queue_pendingEdus:total"
expr: "sum(synapse_federation_transaction_queue_pendingEdus or absent(synapse_federation_transaction_queue_pendingEdus)*0)"
- record: "synapse_federation_transaction_queue_pendingPdus:total"
expr: "sum(synapse_federation_transaction_queue_pendingPdus or absent(synapse_federation_transaction_queue_pendingPdus)*0)"
- record: 'synapse_http_server_requests:method'
labels:
servlet: ""
expr: "sum(synapse_http_server_requests) by (method)"
- record: 'synapse_http_server_requests:servlet'
labels:
method: ""
expr: 'sum(synapse_http_server_requests) by (servlet)'
- record: 'synapse_http_server_requests:total'
labels:
servlet: ""
expr: 'sum(synapse_http_server_requests:by_method) by (servlet)'
- record: 'synapse_cache:hit_ratio_5m'
expr: 'rate(synapse_util_caches_cache:hits[5m]) / rate(synapse_util_caches_cache:total[5m])'
- record: 'synapse_cache:hit_ratio_30s'
expr: 'rate(synapse_util_caches_cache:hits[30s]) / rate(synapse_util_caches_cache:total[30s])'
- record: 'synapse_federation_client_sent'
labels:
type: "EDU"
expr: 'synapse_federation_client_sent_edus + 0'
- record: 'synapse_federation_client_sent'
labels:
type: "PDU"
expr: 'synapse_federation_client_sent_pdu_destinations:count + 0'
- record: 'synapse_federation_client_sent'
labels:
type: "Query"
expr: 'sum(synapse_federation_client_sent_queries) by (job)'
- record: 'synapse_federation_server_received'
labels:
type: "EDU"
expr: 'synapse_federation_server_received_edus + 0'
- record: 'synapse_federation_server_received'
labels:
type: "PDU"
expr: 'synapse_federation_server_received_pdus + 0'
- record: 'synapse_federation_server_received'
labels:
type: "Query"
expr: 'sum(synapse_federation_server_received_queries) by (job)'
- record: 'synapse_federation_transaction_queue_pending'
labels:
type: "EDU"
expr: 'synapse_federation_transaction_queue_pending_edus + 0'
- record: 'synapse_federation_transaction_queue_pending'
labels:
type: "PDU"
expr: 'synapse_federation_transaction_queue_pending_pdus + 0'
+3 -2
View File
@@ -9,9 +9,10 @@ Description=Synapse Matrix homeserver
Type=simple
User=synapse
Group=synapse
EnvironmentFile=-/etc/sysconfig/synapse
WorkingDirectory=/var/lib/synapse
ExecStart=/usr/bin/python2.7 -m synapse.app.homeserver --config-path=/etc/synapse/homeserver.yaml --log-config=/etc/synapse/log_config.yaml
ExecStart=/usr/bin/python2.7 -m synapse.app.homeserver --config-path=/etc/synapse/homeserver.yaml
ExecStop=/usr/bin/synctl stop /etc/synapse/homeserver.yaml
[Install]
WantedBy=multi-user.target
+2
View File
@@ -4,6 +4,8 @@ Purge History API
The purge history API allows server admins to purge historic events from their
database, reclaiming disk space.
**NB!** This will not delete local events (locally sent messages content etc) from the database, but will remove lots of the metadata about them and does dramatically reduce the on disk space usage
Depending on the amount of history being purged a call to the API may take
several minutes or longer. During this period users will not be able to
paginate further back in the room from the point being purged from.
+105 -38
View File
@@ -1,52 +1,119 @@
Basically, PEP8
- Everything should comply with PEP8. Code should pass
``pep8 --max-line-length=100`` without any warnings.
- NEVER tabs. 4 spaces to indent.
- Max line width: 79 chars (with flexibility to overflow by a "few chars" if
- **Indenting**:
- NEVER tabs. 4 spaces to indent.
- follow PEP8; either hanging indent or multiline-visual indent depending
on the size and shape of the arguments and what makes more sense to the
author. In other words, both this::
print("I am a fish %s" % "moo")
and this::
print("I am a fish %s" %
"moo")
and this::
print(
"I am a fish %s" %
"moo",
)
...are valid, although given each one takes up 2x more vertical space than
the previous, it's up to the author's discretion as to which layout makes
most sense for their function invocation. (e.g. if they want to add
comments per-argument, or put expressions in the arguments, or group
related arguments together, or want to deliberately extend or preserve
vertical/horizontal space)
- **Line length**:
Max line length is 79 chars (with flexibility to overflow by a "few chars" if
the overflowing content is not semantically significant and avoids an
explosion of vertical whitespace).
- Use camel case for class and type names
- Use underscores for functions and variables.
- Use double quotes.
- Use parentheses instead of '\\' for line continuation where ever possible
(which is pretty much everywhere)
- There should be max a single new line between:
Use parentheses instead of ``\`` for line continuation where ever possible
(which is pretty much everywhere).
- **Naming**:
- Use camel case for class and type names
- Use underscores for functions and variables.
- Use double quotes ``"foo"`` rather than single quotes ``'foo'``.
- **Blank lines**:
- There should be max a single new line between:
- statements
- functions in a class
- There should be two new lines between:
- There should be two new lines between:
- definitions in a module (e.g., between different classes)
- There should be spaces where spaces should be and not where there shouldn't be:
- a single space after a comma
- a single space before and after for '=' when used as assignment
- no spaces before and after for '=' for default values and keyword arguments.
- Indenting must follow PEP8; either hanging indent or multiline-visual indent
depending on the size and shape of the arguments and what makes more sense to
the author. In other words, both this::
print("I am a fish %s" % "moo")
- **Whitespace**:
and this::
There should be spaces where spaces should be and not where there shouldn't
be:
print("I am a fish %s" %
"moo")
- a single space after a comma
- a single space before and after for '=' when used as assignment
- no spaces before and after for '=' for default values and keyword arguments.
and this::
- **Comments**: should follow the `google code style
<http://google.github.io/styleguide/pyguide.html?showone=Comments#Comments>`_.
This is so that we can generate documentation with `sphinx
<http://sphinxcontrib-napoleon.readthedocs.org/en/latest/>`_. See the
`examples
<http://sphinxcontrib-napoleon.readthedocs.io/en/latest/example_google.html>`_
in the sphinx documentation.
print(
"I am a fish %s" %
"moo"
)
- **Imports**:
...are valid, although given each one takes up 2x more vertical space than
the previous, it's up to the author's discretion as to which layout makes most
sense for their function invocation. (e.g. if they want to add comments
per-argument, or put expressions in the arguments, or group related arguments
together, or want to deliberately extend or preserve vertical/horizontal
space)
- Prefer to import classes and functions than packages or modules.
Comments should follow the `google code style <http://google.github.io/styleguide/pyguide.html?showone=Comments#Comments>`_.
This is so that we can generate documentation with
`sphinx <http://sphinxcontrib-napoleon.readthedocs.org/en/latest/>`_. See the
`examples <http://sphinxcontrib-napoleon.readthedocs.io/en/latest/example_google.html>`_
in the sphinx documentation.
Example::
Code should pass pep8 --max-line-length=100 without any warnings.
from synapse.types import UserID
...
user_id = UserID(local, server)
is preferred over::
from synapse import types
...
user_id = types.UserID(local, server)
(or any other variant).
This goes against the advice in the Google style guide, but it means that
errors in the name are caught early (at import time).
- Multiple imports from the same package can be combined onto one line::
from synapse.types import GroupID, RoomID, UserID
An effort should be made to keep the individual imports in alphabetical
order.
If the list becomes long, wrap it with parentheses and split it over
multiple lines.
- As per `PEP-8 <https://www.python.org/dev/peps/pep-0008/#imports>`_,
imports should be grouped in the following order, with a blank line between
each group:
1. standard library imports
2. related third party imports
3. local application/library specific imports
- Imports within each group should be sorted alphabetically by module name.
- Avoid wildcard imports (``from synapse.types import *``) and relative
imports (``from .types import UserID``).
-4
View File
@@ -298,10 +298,6 @@ It can be used like this:
# this will now be logged against the request context
logger.debug("Request handling complete")
XXX: I think ``preserve_context_over_fn`` is supposed to do the first option,
but the fact that it does ``preserve_context_over_deferred`` on its results
means that its use is fraught with difficulty.
Passing synapse deferreds into third-party functions
----------------------------------------------------
+99
View File
@@ -0,0 +1,99 @@
Password auth provider modules
==============================
Password auth providers offer a way for server administrators to integrate
their Synapse installation with an existing authentication system.
A password auth provider is a Python class which is dynamically loaded into
Synapse, and provides a number of methods by which it can integrate with the
authentication system.
This document serves as a reference for those looking to implement their own
password auth providers.
Required methods
----------------
Password auth provider classes must provide the following methods:
*class* ``SomeProvider.parse_config``\(*config*)
This method is passed the ``config`` object for this module from the
homeserver configuration file.
It should perform any appropriate sanity checks on the provided
configuration, and return an object which is then passed into ``__init__``.
*class* ``SomeProvider``\(*config*, *account_handler*)
The constructor is passed the config object returned by ``parse_config``,
and a ``synapse.module_api.ModuleApi`` object which allows the
password provider to check if accounts exist and/or create new ones.
Optional methods
----------------
Password auth provider classes may optionally provide the following methods.
*class* ``SomeProvider.get_db_schema_files``\()
This method, if implemented, should return an Iterable of ``(name,
stream)`` pairs of database schema files. Each file is applied in turn at
initialisation, and a record is then made in the database so that it is
not re-applied on the next start.
``someprovider.get_supported_login_types``\()
This method, if implemented, should return a ``dict`` mapping from a login
type identifier (such as ``m.login.password``) to an iterable giving the
fields which must be provided by the user in the submission to the
``/login`` api. These fields are passed in the ``login_dict`` dictionary
to ``check_auth``.
For example, if a password auth provider wants to implement a custom login
type of ``com.example.custom_login``, where the client is expected to pass
the fields ``secret1`` and ``secret2``, the provider should implement this
method and return the following dict::
{"com.example.custom_login": ("secret1", "secret2")}
``someprovider.check_auth``\(*username*, *login_type*, *login_dict*)
This method is the one that does the real work. If implemented, it will be
called for each login attempt where the login type matches one of the keys
returned by ``get_supported_login_types``.
It is passed the (possibly UNqualified) ``user`` provided by the client,
the login type, and a dictionary of login secrets passed by the client.
The method should return a Twisted ``Deferred`` object, which resolves to
the canonical ``@localpart:domain`` user id if authentication is successful,
and ``None`` if not.
Alternatively, the ``Deferred`` can resolve to a ``(str, func)`` tuple, in
which case the second field is a callback which will be called with the
result from the ``/login`` call (including ``access_token``, ``device_id``,
etc.)
``someprovider.check_password``\(*user_id*, *password*)
This method provides a simpler interface than ``get_supported_login_types``
and ``check_auth`` for password auth providers that just want to provide a
mechanism for validating ``m.login.password`` logins.
Iif implemented, it will be called to check logins with an
``m.login.password`` login type. It is passed a qualified
``@localpart:domain`` user id, and the password provided by the user.
The method should return a Twisted ``Deferred`` object, which resolves to
``True`` if authentication is successful, and ``False`` if not.
``someprovider.on_logged_out``\(*user_id*, *device_id*, *access_token*)
This method, if implemented, is called when a user logs out. It is passed
the qualified user ID, the ID of the deactivated device (if any: access
tokens are occasionally created without an associated device ID), and the
(now deactivated) access token.
It may return a Twisted ``Deferred`` object; the logout request will wait
for the deferred to complete but the result is ignored.
+2
View File
@@ -1,6 +1,8 @@
Using Postgres
--------------
Postgres version 9.4 or later is known to work.
Set up database
===============
+1 -1
View File
@@ -50,7 +50,7 @@ master_doc = 'index'
# General information about the project.
project = u'Synapse'
copyright = u'2014, TNG'
copyright = u'Copyright 2014-2017 OpenMarket Ltd, 2017 Vector Creations Ltd, 2017 New Vector Ltd'
# The version info for the project you're documenting, acts as replacement for
# |version| and |release|, also used in various other places throughout the
@@ -56,6 +56,7 @@ As a first cut, let's do #2 and have the receiver hit the API to calculate its o
API
---
```
GET /_matrix/media/r0/preview_url?url=http://wherever.com
200 OK
{
@@ -66,6 +67,7 @@ GET /_matrix/media/r0/preview_url?url=http://wherever.com
"og:description" : "“Synapse 0.12 is out! Lots of polishing, performance &amp;amp; bugfixes: /sync API, /r0 prefix, fulltext search, 3PID invites https://t.co/5alhXLLEGP”"
"og:site_name" : "Twitter"
}
```
* Downloads the URL
* If HTML, just stores it in RAM and parses it for OG meta tags
+17
View File
@@ -0,0 +1,17 @@
User Directory API Implementation
=================================
The user directory is currently maintained based on the 'visible' users
on this particular server - i.e. ones which your account shares a room with, or
who are present in a publicly viewable room present on the server.
The directory info is stored in various tables, which can (typically after
DB corruption) get stale or out of sync. If this happens, for now the
quickest solution to fix it is:
```
UPDATE user_directory_stream_pos SET stream_id = NULL;
```
and restart the synapse, which should then start a background task to
flush the current tables and regenerate the directory.
+136 -21
View File
@@ -1,11 +1,15 @@
Scaling synapse via workers
---------------------------
===========================
Synapse has experimental support for splitting out functionality into
multiple separate python processes, helping greatly with scalability. These
processes are called 'workers', and are (eventually) intended to scale
horizontally independently.
All of the below is highly experimental and subject to change as Synapse evolves,
but documenting it here to help folks needing highly scalable Synapses similar
to the one running matrix.org!
All processes continue to share the same database instance, and as such, workers
only work with postgres based synapse deployments (sharing a single sqlite
across multiple processes is a recipe for disaster, plus you should be using
@@ -16,6 +20,16 @@ TCP protocol called 'replication' - analogous to MySQL or Postgres style
database replication; feeding a stream of relevant data to the workers so they
can be kept in sync with the main synapse process and database state.
Configuration
-------------
To make effective use of the workers, you will need to configure an HTTP
reverse-proxy such as nginx or haproxy, which will direct incoming requests to
the correct worker, or to the main synapse instance. Note that this includes
requests made to the federation port. The caveats regarding running a
reverse-proxy on the federation port still apply (see
https://github.com/matrix-org/synapse/blob/master/README.rst#reverse-proxying-the-federation-port).
To enable workers, you need to add a replication listener to the master synapse, e.g.::
listeners:
@@ -27,26 +41,19 @@ Under **no circumstances** should this replication API listener be exposed to th
public internet; it currently implements no authentication whatsoever and is
unencrypted.
You then create a set of configs for the various worker processes. These should be
worker configuration files should be stored in a dedicated subdirectory, to allow
synctl to manipulate them.
The current available worker applications are:
* synapse.app.pusher - handles sending push notifications to sygnal and email
* synapse.app.synchrotron - handles /sync endpoints. can scales horizontally through multiple instances.
* synapse.app.appservice - handles output traffic to Application Services
* synapse.app.federation_reader - handles receiving federation traffic (including public_rooms API)
* synapse.app.media_repository - handles the media repository.
* synapse.app.client_reader - handles client API endpoints like /publicRooms
You then create a set of configs for the various worker processes. These
should be worker configuration files, and should be stored in a dedicated
subdirectory, to allow synctl to manipulate them.
Each worker configuration file inherits the configuration of the main homeserver
configuration file. You can then override configuration specific to that worker,
e.g. the HTTP listener that it provides (if any); logging configuration; etc.
You should minimise the number of overrides though to maintain a usable config.
You must specify the type of worker application (worker_app) and the replication
endpoint that it's talking to on the main synapse process (worker_replication_host
and worker_replication_port).
You must specify the type of worker application (``worker_app``). The currently
available worker applications are listed below. You must also specify the
replication endpoint that it's talking to on the main synapse process
(``worker_replication_host`` and ``worker_replication_port``).
For instance::
@@ -68,11 +75,11 @@ For instance::
worker_log_config: /home/matrix/synapse/config/synchrotron_log_config.yaml
...is a full configuration for a synchrotron worker instance, which will expose a
plain HTTP /sync endpoint on port 8083 separately from the /sync endpoint provided
plain HTTP ``/sync`` endpoint on port 8083 separately from the ``/sync`` endpoint provided
by the main synapse.
Obviously you should configure your loadbalancer to route the /sync endpoint to
the synchrotron instance(s) in this instance.
Obviously you should configure your reverse-proxy to route the relevant
endpoints to the worker (``localhost:8083`` in the above example).
Finally, to actually run your worker-based synapse, you must pass synctl the -a
commandline option to tell it to operate on all the worker configurations found
@@ -89,6 +96,114 @@ To manipulate a specific worker, you pass the -w option to synctl::
synctl -w $CONFIG/workers/synchrotron.yaml restart
All of the above is highly experimental and subject to change as Synapse evolves,
but documenting it here to help folks needing highly scalable Synapses similar
to the one running matrix.org!
Available worker applications
-----------------------------
``synapse.app.pusher``
~~~~~~~~~~~~~~~~~~~~~~
Handles sending push notifications to sygnal and email. Doesn't handle any
REST endpoints itself, but you should set ``start_pushers: False`` in the
shared configuration file to stop the main synapse sending these notifications.
Note this worker cannot be load-balanced: only one instance should be active.
``synapse.app.synchrotron``
~~~~~~~~~~~~~~~~~~~~~~~~~~~
The synchrotron handles ``sync`` requests from clients. In particular, it can
handle REST endpoints matching the following regular expressions::
^/_matrix/client/(v2_alpha|r0)/sync$
^/_matrix/client/(api/v1|v2_alpha|r0)/events$
^/_matrix/client/(api/v1|r0)/initialSync$
^/_matrix/client/(api/v1|r0)/rooms/[^/]+/initialSync$
The above endpoints should all be routed to the synchrotron worker by the
reverse-proxy configuration.
It is possible to run multiple instances of the synchrotron to scale
horizontally. In this case the reverse-proxy should be configured to
load-balance across the instances, though it will be more efficient if all
requests from a particular user are routed to a single instance. Extracting
a userid from the access token is currently left as an exercise for the reader.
``synapse.app.appservice``
~~~~~~~~~~~~~~~~~~~~~~~~~~
Handles sending output traffic to Application Services. Doesn't handle any
REST endpoints itself, but you should set ``notify_appservices: False`` in the
shared configuration file to stop the main synapse sending these notifications.
Note this worker cannot be load-balanced: only one instance should be active.
``synapse.app.federation_reader``
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Handles a subset of federation endpoints. In particular, it can handle REST
endpoints matching the following regular expressions::
^/_matrix/federation/v1/event/
^/_matrix/federation/v1/state/
^/_matrix/federation/v1/state_ids/
^/_matrix/federation/v1/backfill/
^/_matrix/federation/v1/get_missing_events/
^/_matrix/federation/v1/publicRooms
The above endpoints should all be routed to the federation_reader worker by the
reverse-proxy configuration.
``synapse.app.federation_sender``
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Handles sending federation traffic to other servers. Doesn't handle any
REST endpoints itself, but you should set ``send_federation: False`` in the
shared configuration file to stop the main synapse sending this traffic.
Note this worker cannot be load-balanced: only one instance should be active.
``synapse.app.media_repository``
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Handles the media repository. It can handle all endpoints starting with::
/_matrix/media/
You should also set ``enable_media_repo: False`` in the shared configuration
file to stop the main synapse running background jobs related to managing the
media repository.
Note this worker cannot be load-balanced: only one instance should be active.
``synapse.app.client_reader``
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Handles client API endpoints. It can handle REST endpoints matching the
following regular expressions::
^/_matrix/client/(api/v1|r0|unstable)/publicRooms$
``synapse.app.user_dir``
~~~~~~~~~~~~~~~~~~~~~~~~
Handles searches in the user directory. It can handle REST endpoints matching
the following regular expressions::
^/_matrix/client/(api/v1|r0|unstable)/user_directory/search$
``synapse.app.frontend_proxy``
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Proxies some frequently-requested client endpoints to add caching and remove
load from the main synapse. It can handle REST endpoints matching the following
regular expressions::
^/_matrix/client/(api/v1|r0|unstable)/keys/upload
It will proxy any requests it cannot handle to the main synapse instance. It
must therefore be configured with the location of the main instance, via
the ``worker_main_http_uri`` setting in the frontend_proxy worker configuration
file. For example::
worker_main_http_uri: http://127.0.0.1:8008
+1
View File
@@ -17,6 +17,7 @@ export HAPROXY_BIN=/home/haproxy/haproxy-1.6.11/haproxy
./sytest/jenkins/prep_sytest_for_postgres.sh
./sytest/jenkins/install_and_run.sh \
--python $WORKSPACE/.tox/py27/bin/python \
--synapse-directory $WORKSPACE \
--dendron $WORKSPACE/dendron/bin/dendron \
--haproxy \
+1
View File
@@ -15,5 +15,6 @@ export SYNAPSE_CACHE_FACTOR=1
./sytest/jenkins/prep_sytest_for_postgres.sh
./sytest/jenkins/install_and_run.sh \
--python $WORKSPACE/.tox/py27/bin/python \
--synapse-directory $WORKSPACE \
--dendron $WORKSPACE/dendron/bin/dendron \
+1
View File
@@ -14,4 +14,5 @@ export SYNAPSE_CACHE_FACTOR=1
./sytest/jenkins/prep_sytest_for_postgres.sh
./sytest/jenkins/install_and_run.sh \
--python $WORKSPACE/.tox/py27/bin/python \
--synapse-directory $WORKSPACE \
+1
View File
@@ -12,4 +12,5 @@ export SYNAPSE_CACHE_FACTOR=1
./jenkins/clone.sh sytest https://github.com/matrix-org/sytest.git
./sytest/jenkins/install_and_run.sh \
--python $WORKSPACE/.tox/py27/bin/python \
--synapse-directory $WORKSPACE \
Regular → Executable
+111 -14
View File
@@ -1,10 +1,30 @@
#!/usr/bin/env python
#
# Copyright 2015, 2016 OpenMarket Ltd
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from __future__ import print_function
import argparse
import nacl.signing
import json
import base64
import requests
import sys
import srvlookup
import yaml
def encode_base64(input_bytes):
"""Encode bytes as a base64 string without any padding."""
@@ -103,15 +123,25 @@ def lookup(destination, path):
except:
return "https://%s:%d%s" % (destination, 8448, path)
def get_json(origin_name, origin_key, destination, path):
request_json = {
"method": "GET",
def request_json(method, origin_name, origin_key, destination, path, content):
if method is None:
if content is None:
method = "GET"
else:
method = "POST"
json_to_sign = {
"method": method,
"uri": path,
"origin": origin_name,
"destination": destination,
}
signed_json = sign_json(request_json, origin_key, origin_name)
if content is not None:
json_to_sign["content"] = json.loads(content)
signed_json = sign_json(json_to_sign, origin_key, origin_name)
authorization_headers = []
@@ -120,30 +150,97 @@ def get_json(origin_name, origin_key, destination, path):
origin_name, key, sig,
)
authorization_headers.append(bytes(header))
sys.stderr.write(header)
sys.stderr.write("\n")
print ("Authorization: %s" % header, file=sys.stderr)
result = requests.get(
lookup(destination, path),
dest = lookup(destination, path)
print ("Requesting %s" % dest, file=sys.stderr)
result = requests.request(
method=method,
url=dest,
headers={"Authorization": authorization_headers[0]},
verify=False,
data=content,
)
sys.stderr.write("Status Code: %d\n" % (result.status_code,))
return result.json()
def main():
origin_name, keyfile, destination, path = sys.argv[1:]
parser = argparse.ArgumentParser(
description=
"Signs and sends a federation request to a matrix homeserver",
)
with open(keyfile) as f:
parser.add_argument(
"-N", "--server-name",
help="Name to give as the local homeserver. If unspecified, will be "
"read from the config file.",
)
parser.add_argument(
"-k", "--signing-key-path",
help="Path to the file containing the private ed25519 key to sign the "
"request with.",
)
parser.add_argument(
"-c", "--config",
default="homeserver.yaml",
help="Path to server config file. Ignored if --server-name and "
"--signing-key-path are both given.",
)
parser.add_argument(
"-d", "--destination",
default="matrix.org",
help="name of the remote homeserver. We will do SRV lookups and "
"connect appropriately.",
)
parser.add_argument(
"-X", "--method",
help="HTTP method to use for the request. Defaults to GET if --data is"
"unspecified, POST if it is."
)
parser.add_argument(
"--body",
help="Data to send as the body of the HTTP request"
)
parser.add_argument(
"path",
help="request path. We will add '/_matrix/federation/v1/' to this."
)
args = parser.parse_args()
if not args.server_name or not args.signing_key_path:
read_args_from_config(args)
with open(args.signing_key_path) as f:
key = read_signing_keys(f)[0]
result = get_json(
origin_name, key, destination, "/_matrix/federation/v1/" + path
result = request_json(
args.method,
args.server_name, key, args.destination,
"/_matrix/federation/v1/" + args.path,
content=args.body,
)
json.dump(result, sys.stdout)
print ""
print ("")
def read_args_from_config(args):
with open(args.config, 'r') as fh:
config = yaml.safe_load(fh)
if not args.server_name:
args.server_name = config['server_name']
if not args.signing_key_path:
args.signing_key_path = config['signing_key_path']
if __name__ == "__main__":
main()
+56 -12
View File
@@ -41,6 +41,15 @@ BOOLEAN_COLUMNS = {
"presence_stream": ["currently_active"],
"public_room_list_stream": ["visibility"],
"device_lists_outbound_pokes": ["sent"],
"users_who_share_rooms": ["share_private"],
"groups": ["is_public"],
"group_rooms": ["is_public"],
"group_users": ["is_public", "is_admin"],
"group_summary_rooms": ["is_public"],
"group_room_categories": ["is_public"],
"group_summary_users": ["is_public"],
"group_roles": ["is_public"],
"local_group_membership": ["is_publicised", "is_admin"],
}
@@ -111,6 +120,7 @@ class Store(object):
_simple_update_one = SQLBaseStore.__dict__["_simple_update_one"]
_simple_update_one_txn = SQLBaseStore.__dict__["_simple_update_one_txn"]
_simple_update_txn = SQLBaseStore.__dict__["_simple_update_txn"]
def runInteraction(self, desc, func, *args, **kwargs):
def r(conn):
@@ -121,7 +131,7 @@ class Store(object):
try:
txn = conn.cursor()
return func(
LoggingTransaction(txn, desc, self.database_engine, []),
LoggingTransaction(txn, desc, self.database_engine, [], []),
*args, **kwargs
)
except self.database_engine.module.DatabaseError as e:
@@ -251,6 +261,25 @@ class Porter(object):
)
return
if table in (
"user_directory", "user_directory_search", "users_who_share_rooms",
"users_in_pubic_room",
):
# We don't port these tables, as they're a faff and we can regenreate
# them anyway.
self.progress.update(table, table_size) # Mark table as done
return
if table == "user_directory_stream_pos":
# We need to make sure there is a single row, `(X, null), as that is
# what synapse expects to be there.
yield self.postgres_store._simple_insert(
table=table,
values={"stream_id": None},
)
self.progress.update(table, table_size) # Mark table as done
return
forward_select = (
"SELECT rowid, * FROM %s WHERE rowid >= ? ORDER BY rowid LIMIT ?"
% (table,)
@@ -298,7 +327,7 @@ class Porter(object):
backward_chunk = min(row[0] for row in brows) - 1
rows = frows + brows
self._convert_rows(table, headers, rows)
rows = self._convert_rows(table, headers, rows)
def insert(txn):
self.postgres_store.insert_many_txn(
@@ -356,10 +385,13 @@ class Porter(object):
" VALUES (?,?,?,?,to_tsvector('english', ?),?,?)"
)
rows_dict = [
dict(zip(headers, row))
for row in rows
]
rows_dict = []
for row in rows:
d = dict(zip(headers, row))
if "\0" in d['value']:
logger.warn('dropping search row %s', d)
else:
rows_dict.append(d)
txn.executemany(sql, [
(
@@ -531,17 +563,29 @@ class Porter(object):
i for i, h in enumerate(headers) if h in bool_col_names
]
class BadValueException(Exception):
pass
def conv(j, col):
if j in bool_cols:
return bool(col)
elif isinstance(col, basestring) and "\0" in col:
logger.warn("DROPPING ROW: NUL value in table %s col %s: %r", table, headers[j], col)
raise BadValueException();
return col
outrows = []
for i, row in enumerate(rows):
rows[i] = tuple(
conv(j, col)
for j, col in enumerate(row)
if j > 0
)
try:
outrows.append(tuple(
conv(j, col)
for j, col in enumerate(row)
if j > 0
))
except BadValueException:
pass
return outrows
@defer.inlineCallbacks
def _setup_sent_transactions(self):
@@ -569,7 +613,7 @@ class Porter(object):
"select", r,
)
self._convert_rows("sent_transactions", headers, rows)
rows = self._convert_rows("sent_transactions", headers, rows)
inserted_rows = len(rows)
if inserted_rows:
+45
View File
@@ -0,0 +1,45 @@
#!/usr/bin/env perl
use strict;
use warnings;
use JSON::XS;
use LWP::UserAgent;
use URI::Escape;
if (@ARGV < 4) {
die "usage: $0 <homeserver url> <access_token> <room_id|room_alias> <group_id>\n";
}
my ($hs, $access_token, $room_id, $group_id) = @ARGV;
my $ua = LWP::UserAgent->new();
$ua->timeout(10);
if ($room_id =~ /^#/) {
$room_id = uri_escape($room_id);
$room_id = decode_json($ua->get("${hs}/_matrix/client/r0/directory/room/${room_id}?access_token=${access_token}")->decoded_content)->{room_id};
}
my $room_users = [ keys %{decode_json($ua->get("${hs}/_matrix/client/r0/rooms/${room_id}/joined_members?access_token=${access_token}")->decoded_content)->{joined}} ];
my $group_users = [
(map { $_->{user_id} } @{decode_json($ua->get("${hs}/_matrix/client/unstable/groups/${group_id}/users?access_token=${access_token}" )->decoded_content)->{chunk}}),
(map { $_->{user_id} } @{decode_json($ua->get("${hs}/_matrix/client/unstable/groups/${group_id}/invited_users?access_token=${access_token}" )->decoded_content)->{chunk}}),
];
die "refusing to sync from empty room" unless (@$room_users);
die "refusing to sync to empty group" unless (@$group_users);
my $diff = {};
foreach my $user (@$room_users) { $diff->{$user}++ }
foreach my $user (@$group_users) { $diff->{$user}-- }
foreach my $user (keys %$diff) {
if ($diff->{$user} == 1) {
warn "inviting $user";
print STDERR $ua->put("${hs}/_matrix/client/unstable/groups/${group_id}/admin/users/invite/${user}?access_token=${access_token}", Content=>'{}')->status_line."\n";
}
elsif ($diff->{$user} == -1) {
warn "removing $user";
print STDERR $ua->put("${hs}/_matrix/client/unstable/groups/${group_id}/admin/users/remove/${user}?access_token=${access_token}", Content=>'{}')->status_line."\n";
}
}
+1 -1
View File
@@ -16,4 +16,4 @@
""" This is a reference implementation of a Matrix home server.
"""
__version__ = "0.21.0"
__version__ = "0.25.1"
+77 -28
View File
@@ -23,7 +23,8 @@ from synapse import event_auth
from synapse.api.constants import EventTypes, Membership, JoinRules
from synapse.api.errors import AuthError, Codes
from synapse.types import UserID
from synapse.util import logcontext
from synapse.util.caches import register_cache, CACHE_SIZE_FACTOR
from synapse.util.caches.lrucache import LruCache
from synapse.util.metrics import Measure
logger = logging.getLogger(__name__)
@@ -39,6 +40,10 @@ AuthEventTypes = (
GUEST_DEVICE_ID = "guest_device"
class _InvalidMacaroonException(Exception):
pass
class Auth(object):
"""
FIXME: This class contains a mix of functions for authenticating users
@@ -51,6 +56,9 @@ class Auth(object):
self.state = hs.get_state_handler()
self.TOKEN_NOT_FOUND_HTTP_STATUS = 401
self.token_cache = LruCache(CACHE_SIZE_FACTOR * 10000)
register_cache("token_cache", self.token_cache)
@defer.inlineCallbacks
def check_from_context(self, event, context, do_sig_check=True):
auth_events_ids = yield self.compute_auth_events(
@@ -144,17 +152,8 @@ class Auth(object):
@defer.inlineCallbacks
def check_host_in_room(self, room_id, host):
with Measure(self.clock, "check_host_in_room"):
latest_event_ids = yield self.store.get_latest_event_ids_in_room(room_id)
logger.debug("calling resolve_state_groups from check_host_in_room")
entry = yield self.state.resolve_state_groups(
room_id, latest_event_ids
)
ret = yield self.store.is_host_joined(
room_id, host, entry.state_group, entry.state
)
defer.returnValue(ret)
latest_event_ids = yield self.store.is_host_joined(room_id, host)
defer.returnValue(latest_event_ids)
def _check_joined_room(self, member, user_id, room_id):
if not member or member.membership != Membership.JOIN:
@@ -209,8 +208,8 @@ class Auth(object):
default=[""]
)[0]
if user and access_token and ip_addr:
logcontext.preserve_fn(self.store.insert_client_ip)(
user=user,
self.store.insert_client_ip(
user_id=user.to_string(),
access_token=access_token,
ip=ip_addr,
user_agent=user_agent,
@@ -271,13 +270,17 @@ class Auth(object):
rights (str): The operation being performed; the access token must
allow this.
Returns:
dict : dict that includes the user and the ID of their access token.
Deferred[dict]: dict that includes:
`user` (UserID)
`is_guest` (bool)
`token_id` (int|None): access token id. May be None if guest
`device_id` (str|None): device corresponding to access token
Raises:
AuthError if no user by that token exists or the token is invalid.
"""
try:
macaroon = pymacaroons.Macaroon.deserialize(token)
except Exception: # deserialize can throw more-or-less anything
user_id, guest = self._parse_and_validate_macaroon(token, rights)
except _InvalidMacaroonException:
# doesn't look like a macaroon: treat it as an opaque token which
# must be in the database.
# TODO: it would be nice to get rid of this, but apparently some
@@ -286,19 +289,8 @@ class Auth(object):
defer.returnValue(r)
try:
user_id = self.get_user_id_from_macaroon(macaroon)
user = UserID.from_string(user_id)
self.validate_macaroon(
macaroon, rights, self.hs.config.expire_access_token,
user_id=user_id,
)
guest = False
for caveat in macaroon.caveats:
if caveat.caveat_id == "guest = true":
guest = True
if guest:
# Guest access tokens are not stored in the database (there can
# only be one access token per guest, anyway).
@@ -370,6 +362,55 @@ class Auth(object):
errcode=Codes.UNKNOWN_TOKEN
)
def _parse_and_validate_macaroon(self, token, rights="access"):
"""Takes a macaroon and tries to parse and validate it. This is cached
if and only if rights == access and there isn't an expiry.
On invalid macaroon raises _InvalidMacaroonException
Returns:
(user_id, is_guest)
"""
if rights == "access":
cached = self.token_cache.get(token, None)
if cached:
return cached
try:
macaroon = pymacaroons.Macaroon.deserialize(token)
except Exception: # deserialize can throw more-or-less anything
# doesn't look like a macaroon: treat it as an opaque token which
# must be in the database.
# TODO: it would be nice to get rid of this, but apparently some
# people use access tokens which aren't macaroons
raise _InvalidMacaroonException()
try:
user_id = self.get_user_id_from_macaroon(macaroon)
has_expiry = False
guest = False
for caveat in macaroon.caveats:
if caveat.caveat_id.startswith("time "):
has_expiry = True
elif caveat.caveat_id == "guest = true":
guest = True
self.validate_macaroon(
macaroon, rights, self.hs.config.expire_access_token,
user_id=user_id,
)
except (pymacaroons.exceptions.MacaroonException, TypeError, ValueError):
raise AuthError(
self.TOKEN_NOT_FOUND_HTTP_STATUS, "Invalid macaroon passed.",
errcode=Codes.UNKNOWN_TOKEN
)
if not has_expiry and rights == "access":
self.token_cache[token] = (user_id, guest)
return user_id, guest
def get_user_id_from_macaroon(self, macaroon):
"""Retrieve the user_id given by the caveats on the macaroon.
@@ -482,6 +523,14 @@ class Auth(object):
)
def is_server_admin(self, user):
""" Check if the given user is a local server admin.
Args:
user (str): mxid of user to check
Returns:
bool: True if the user is an admin
"""
return self.store.is_server_admin(user)
@defer.inlineCallbacks
+16
View File
@@ -140,6 +140,22 @@ class RegistrationError(SynapseError):
pass
class InteractiveAuthIncompleteError(Exception):
"""An error raised when UI auth is not yet complete
(This indicates we should return a 401 with 'result' as the body)
Attributes:
result (dict): the server response to the request, which should be
passed back to the client
"""
def __init__(self, result):
super(InteractiveAuthIncompleteError, self).__init__(
"Interactive auth not yet complete",
)
self.result = result
class UnrecognizedRequestError(SynapseError):
"""An error indicating we don't understand the request you're trying to make"""
def __init__(self, *args, **kwargs):
+178
View File
@@ -0,0 +1,178 @@
# -*- coding: utf-8 -*-
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import gc
import logging
import sys
try:
import affinity
except Exception:
affinity = None
from daemonize import Daemonize
from synapse.util import PreserveLoggingContext
from synapse.util.rlimit import change_resource_limit
from twisted.internet import error, reactor
logger = logging.getLogger(__name__)
def start_worker_reactor(appname, config):
""" Run the reactor in the main process
Daemonizes if necessary, and then configures some resources, before starting
the reactor. Pulls configuration from the 'worker' settings in 'config'.
Args:
appname (str): application name which will be sent to syslog
config (synapse.config.Config): config object
"""
logger = logging.getLogger(config.worker_app)
start_reactor(
appname,
config.soft_file_limit,
config.gc_thresholds,
config.worker_pid_file,
config.worker_daemonize,
config.worker_cpu_affinity,
logger,
)
def start_reactor(
appname,
soft_file_limit,
gc_thresholds,
pid_file,
daemonize,
cpu_affinity,
logger,
):
""" Run the reactor in the main process
Daemonizes if necessary, and then configures some resources, before starting
the reactor
Args:
appname (str): application name which will be sent to syslog
soft_file_limit (int):
gc_thresholds:
pid_file (str): name of pid file to write to if daemonize is True
daemonize (bool): true to run the reactor in a background process
cpu_affinity (int|None): cpu affinity mask
logger (logging.Logger): logger instance to pass to Daemonize
"""
def run():
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
logger.info("Running")
if cpu_affinity is not None:
if not affinity:
quit_with_error(
"Missing package 'affinity' required for cpu_affinity\n"
"option\n\n"
"Install by running:\n\n"
" pip install affinity\n\n"
)
logger.info("Setting CPU affinity to %s" % cpu_affinity)
affinity.set_process_affinity_mask(0, cpu_affinity)
change_resource_limit(soft_file_limit)
if gc_thresholds:
gc.set_threshold(*gc_thresholds)
reactor.run()
if daemonize:
daemon = Daemonize(
app=appname,
pid=pid_file,
action=run,
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
run()
def quit_with_error(error_string):
message_lines = error_string.split("\n")
line_length = max([len(l) for l in message_lines if len(l) < 80]) + 2
sys.stderr.write("*" * line_length + '\n')
for line in message_lines:
sys.stderr.write(" %s\n" % (line.rstrip(),))
sys.stderr.write("*" * line_length + '\n')
sys.exit(1)
def listen_tcp(bind_addresses, port, factory, backlog=50):
"""
Create a TCP socket for a port and several addresses
"""
for address in bind_addresses:
try:
reactor.listenTCP(
port,
factory,
backlog,
address
)
except error.CannotListenError as e:
check_bind_error(e, address, bind_addresses)
def listen_ssl(bind_addresses, port, factory, context_factory, backlog=50):
"""
Create an SSL socket for a port and several addresses
"""
for address in bind_addresses:
try:
reactor.listenSSL(
port,
factory,
context_factory,
backlog,
address
)
except error.CannotListenError as e:
check_bind_error(e, address, bind_addresses)
def check_bind_error(e, address, bind_addresses):
"""
This method checks an exception occurred while binding on 0.0.0.0.
If :: is specified in the bind addresses a warning is shown.
The exception is still raised otherwise.
Binding on both 0.0.0.0 and :: causes an exception on Linux and macOS
because :: binds on both IPv4 and IPv6 (as per RFC 3493).
When binding on 0.0.0.0 after :: this can safely be ignored.
Args:
e (Exception): Exception that was caught.
address (str): Address on which binding was attempted.
bind_addresses (list): Addresses on which the service listens.
"""
if address == '0.0.0.0' and '::' in bind_addresses:
logger.warn('Failed to listen on 0.0.0.0, continuing because listening on [::]')
else:
raise e
+27 -61
View File
@@ -13,38 +13,31 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import sys
import synapse
from synapse.server import HomeServer
from synapse import events
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.logger import setup_logging
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.http.site import SynapseSite
from synapse.metrics.resource import MetricsResource, METRICS_PREFIX
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage.appservice import SlavedApplicationServiceStore
from synapse.replication.slave.storage.directory import DirectoryStore
from synapse.replication.slave.storage.events import SlavedEventStore
from synapse.replication.slave.storage.appservice import SlavedApplicationServiceStore
from synapse.replication.slave.storage.registration import SlavedRegistrationStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.server import HomeServer
from synapse.storage.engines import create_engine
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext, PreserveLoggingContext, preserve_fn
from synapse.util.logcontext import LoggingContext, preserve_fn
from synapse.util.manhole import manhole
from synapse.util.rlimit import change_resource_limit
from synapse.util.versionstring import get_version_string
from synapse import events
from twisted.internet import reactor
from twisted.web.resource import Resource
from daemonize import Daemonize
import sys
import logging
import gc
logger = logging.getLogger("synapse.app.appservice")
@@ -86,17 +79,16 @@ class AppserviceServer(HomeServer):
root_resource = create_resource_tree(resources, Resource())
for address in bind_addresses:
reactor.listenTCP(
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
interface=address
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse appservice now listening on port %d", port)
@@ -105,18 +97,15 @@ class AppserviceServer(HomeServer):
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
reactor.listenTCP(
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
),
interface=address
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
@@ -181,36 +170,13 @@ def start(config_options):
ps.setup()
ps.start_listening(config.worker_listeners)
def run():
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
logger.info("Running")
change_resource_limit(config.soft_file_limit)
if config.gc_thresholds:
gc.set_threshold(*config.gc_thresholds)
reactor.run()
def start():
ps.get_datastore().start_profiling()
ps.get_state_handler().start_caching()
reactor.callWhenRunning(start)
if config.worker_daemonize:
daemon = Daemonize(
app="synapse-appservice",
pid=config.worker_pid_file,
action=run,
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
run()
_base.start_worker_reactor("synapse-appservice", config)
if __name__ == '__main__':
+31 -65
View File
@@ -13,47 +13,39 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import sys
import synapse
from synapse import events
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.http.site import SynapseSite
from synapse.crypto import context_factory
from synapse.http.server import JsonResource
from synapse.metrics.resource import MetricsResource, METRICS_PREFIX
from synapse.http.site import SynapseSite
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage._base import BaseSlavedStore
from synapse.replication.slave.storage.appservice import SlavedApplicationServiceStore
from synapse.replication.slave.storage.client_ips import SlavedClientIpStore
from synapse.replication.slave.storage.directory import DirectoryStore
from synapse.replication.slave.storage.events import SlavedEventStore
from synapse.replication.slave.storage.keys import SlavedKeyStore
from synapse.replication.slave.storage.room import RoomStore
from synapse.replication.slave.storage.directory import DirectoryStore
from synapse.replication.slave.storage.registration import SlavedRegistrationStore
from synapse.replication.slave.storage.room import RoomStore
from synapse.replication.slave.storage.transactions import TransactionStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.rest.client.v1.room import PublicRoomListRestServlet
from synapse.server import HomeServer
from synapse.storage.client_ips import ClientIpStore
from synapse.storage.engines import create_engine
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext, PreserveLoggingContext
from synapse.util.logcontext import LoggingContext
from synapse.util.manhole import manhole
from synapse.util.rlimit import change_resource_limit
from synapse.util.versionstring import get_version_string
from synapse.crypto import context_factory
from synapse import events
from twisted.internet import reactor
from twisted.web.resource import Resource
from daemonize import Daemonize
import sys
import logging
import gc
logger = logging.getLogger("synapse.app.client_reader")
@@ -65,8 +57,8 @@ class ClientReaderSlavedStore(
SlavedApplicationServiceStore,
SlavedRegistrationStore,
TransactionStore,
SlavedClientIpStore,
BaseSlavedStore,
ClientIpStore, # After BaseSlavedStore because the constructor is different
):
pass
@@ -111,17 +103,16 @@ class ClientReaderServer(HomeServer):
root_resource = create_resource_tree(resources, Resource())
for address in bind_addresses:
reactor.listenTCP(
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
interface=address
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse client reader now listening on port %d", port)
@@ -130,18 +121,16 @@ class ClientReaderServer(HomeServer):
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
reactor.listenTCP(
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
),
interface=address
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
@@ -183,36 +172,13 @@ def start(config_options):
ss.get_handlers()
ss.start_listening(config.worker_listeners)
def run():
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
logger.info("Running")
change_resource_limit(config.soft_file_limit)
if config.gc_thresholds:
gc.set_threshold(*config.gc_thresholds)
reactor.run()
def start():
ss.get_state_handler().start_caching()
ss.get_datastore().start_profiling()
reactor.callWhenRunning(start)
if config.worker_daemonize:
daemon = Daemonize(
app="synapse-client-reader",
pid=config.worker_pid_file,
action=run,
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
run()
_base.start_worker_reactor("synapse-client-reader", config)
if __name__ == '__main__':
+28 -63
View File
@@ -13,44 +13,36 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import sys
import synapse
from synapse import events
from synapse.api.urls import FEDERATION_PREFIX
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.crypto import context_factory
from synapse.federation.transport.server import TransportLayerServer
from synapse.http.site import SynapseSite
from synapse.metrics.resource import MetricsResource, METRICS_PREFIX
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage._base import BaseSlavedStore
from synapse.replication.slave.storage.directory import DirectoryStore
from synapse.replication.slave.storage.events import SlavedEventStore
from synapse.replication.slave.storage.keys import SlavedKeyStore
from synapse.replication.slave.storage.room import RoomStore
from synapse.replication.slave.storage.transactions import TransactionStore
from synapse.replication.slave.storage.directory import DirectoryStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.server import HomeServer
from synapse.storage.engines import create_engine
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext, PreserveLoggingContext
from synapse.util.logcontext import LoggingContext
from synapse.util.manhole import manhole
from synapse.util.rlimit import change_resource_limit
from synapse.util.versionstring import get_version_string
from synapse.api.urls import FEDERATION_PREFIX
from synapse.federation.transport.server import TransportLayerServer
from synapse.crypto import context_factory
from synapse import events
from twisted.internet import reactor
from twisted.web.resource import Resource
from daemonize import Daemonize
import sys
import logging
import gc
logger = logging.getLogger("synapse.app.federation_reader")
@@ -100,17 +92,16 @@ class FederationReaderServer(HomeServer):
root_resource = create_resource_tree(resources, Resource())
for address in bind_addresses:
reactor.listenTCP(
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
interface=address
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse federation reader now listening on port %d", port)
@@ -119,18 +110,15 @@ class FederationReaderServer(HomeServer):
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
reactor.listenTCP(
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
),
interface=address
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
@@ -172,36 +160,13 @@ def start(config_options):
ss.get_handlers()
ss.start_listening(config.worker_listeners)
def run():
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
logger.info("Running")
change_resource_limit(config.soft_file_limit)
if config.gc_thresholds:
gc.set_threshold(*config.gc_thresholds)
reactor.run()
def start():
ss.get_state_handler().start_caching()
ss.get_datastore().start_profiling()
reactor.callWhenRunning(start)
if config.worker_daemonize:
daemon = Daemonize(
app="synapse-federation-reader",
pid=config.worker_pid_file,
action=run,
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
run()
_base.start_worker_reactor("synapse-federation-reader", config)
if __name__ == '__main__':
+31 -66
View File
@@ -13,45 +13,38 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import sys
import synapse
from synapse.server import HomeServer
from synapse import events
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.logger import setup_logging
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.crypto import context_factory
from synapse.http.site import SynapseSite
from synapse.federation import send_queue
from synapse.metrics.resource import MetricsResource, METRICS_PREFIX
from synapse.http.site import SynapseSite
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage.deviceinbox import SlavedDeviceInboxStore
from synapse.replication.slave.storage.devices import SlavedDeviceStore
from synapse.replication.slave.storage.events import SlavedEventStore
from synapse.replication.slave.storage.presence import SlavedPresenceStore
from synapse.replication.slave.storage.receipts import SlavedReceiptsStore
from synapse.replication.slave.storage.registration import SlavedRegistrationStore
from synapse.replication.slave.storage.presence import SlavedPresenceStore
from synapse.replication.slave.storage.transactions import TransactionStore
from synapse.replication.slave.storage.devices import SlavedDeviceStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.server import HomeServer
from synapse.storage.engines import create_engine
from synapse.util.async import Linearizer
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext, PreserveLoggingContext, preserve_fn
from synapse.util.logcontext import LoggingContext, preserve_fn
from synapse.util.manhole import manhole
from synapse.util.rlimit import change_resource_limit
from synapse.util.versionstring import get_version_string
from synapse import events
from twisted.internet import reactor, defer
from twisted.internet import defer, reactor
from twisted.web.resource import Resource
from daemonize import Daemonize
import sys
import logging
import gc
logger = logging.getLogger("synapse.app.appservice")
logger = logging.getLogger("synapse.app.federation_sender")
class FederationSenderSlaveStore(
@@ -113,17 +106,16 @@ class FederationSenderServer(HomeServer):
root_resource = create_resource_tree(resources, Resource())
for address in bind_addresses:
reactor.listenTCP(
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
interface=address
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse federation_sender now listening on port %d", port)
@@ -132,18 +124,15 @@ class FederationSenderServer(HomeServer):
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
reactor.listenTCP(
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
),
interface=address
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
@@ -213,36 +202,12 @@ def start(config_options):
ps.setup()
ps.start_listening(config.worker_listeners)
def run():
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
logger.info("Running")
change_resource_limit(config.soft_file_limit)
if config.gc_thresholds:
gc.set_threshold(*config.gc_thresholds)
reactor.run()
def start():
ps.get_datastore().start_profiling()
ps.get_state_handler().start_caching()
reactor.callWhenRunning(start)
if config.worker_daemonize:
daemon = Daemonize(
app="synapse-federation-sender",
pid=config.worker_pid_file,
action=run,
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
run()
_base.start_worker_reactor("synapse-federation-sender", config)
class FederationSenderHandler(object):
+241
View File
@@ -0,0 +1,241 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# Copyright 2016 OpenMarket Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import sys
import synapse
from synapse import events
from synapse.api.errors import SynapseError
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.crypto import context_factory
from synapse.http.server import JsonResource
from synapse.http.servlet import (
RestServlet, parse_json_object_from_request,
)
from synapse.http.site import SynapseSite
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage._base import BaseSlavedStore
from synapse.replication.slave.storage.appservice import SlavedApplicationServiceStore
from synapse.replication.slave.storage.client_ips import SlavedClientIpStore
from synapse.replication.slave.storage.devices import SlavedDeviceStore
from synapse.replication.slave.storage.registration import SlavedRegistrationStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.rest.client.v2_alpha._base import client_v2_patterns
from synapse.server import HomeServer
from synapse.storage.engines import create_engine
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext
from synapse.util.manhole import manhole
from synapse.util.versionstring import get_version_string
from twisted.internet import defer, reactor
from twisted.web.resource import Resource
logger = logging.getLogger("synapse.app.frontend_proxy")
class KeyUploadServlet(RestServlet):
PATTERNS = client_v2_patterns("/keys/upload(/(?P<device_id>[^/]+))?$")
def __init__(self, hs):
"""
Args:
hs (synapse.server.HomeServer): server
"""
super(KeyUploadServlet, self).__init__()
self.auth = hs.get_auth()
self.store = hs.get_datastore()
self.http_client = hs.get_simple_http_client()
self.main_uri = hs.config.worker_main_http_uri
@defer.inlineCallbacks
def on_POST(self, request, device_id):
requester = yield self.auth.get_user_by_req(request, allow_guest=True)
user_id = requester.user.to_string()
body = parse_json_object_from_request(request)
if device_id is not None:
# passing the device_id here is deprecated; however, we allow it
# for now for compatibility with older clients.
if (requester.device_id is not None and
device_id != requester.device_id):
logger.warning("Client uploading keys for a different device "
"(logged in as %s, uploading for %s)",
requester.device_id, device_id)
else:
device_id = requester.device_id
if device_id is None:
raise SynapseError(
400,
"To upload keys, you must pass device_id when authenticating"
)
if body:
# They're actually trying to upload something, proxy to main synapse.
# Pass through the auth headers, if any, in case the access token
# is there.
auth_headers = request.requestHeaders.getRawHeaders("Authorization", [])
headers = {
"Authorization": auth_headers,
}
result = yield self.http_client.post_json_get_json(
self.main_uri + request.uri,
body,
headers=headers,
)
defer.returnValue((200, result))
else:
# Just interested in counts.
result = yield self.store.count_e2e_one_time_keys(user_id, device_id)
defer.returnValue((200, {"one_time_key_counts": result}))
class FrontendProxySlavedStore(
SlavedDeviceStore,
SlavedClientIpStore,
SlavedApplicationServiceStore,
SlavedRegistrationStore,
BaseSlavedStore,
):
pass
class FrontendProxyServer(HomeServer):
def get_db_conn(self, run_new_connection=True):
# Any param beginning with cp_ is a parameter for adbapi, and should
# not be passed to the database engine.
db_params = {
k: v for k, v in self.db_config.get("args", {}).items()
if not k.startswith("cp_")
}
db_conn = self.database_engine.module.connect(**db_params)
if run_new_connection:
self.database_engine.on_new_connection(db_conn)
return db_conn
def setup(self):
logger.info("Setting up.")
self.datastore = FrontendProxySlavedStore(self.get_db_conn(), self)
logger.info("Finished setting up.")
def _listen_http(self, listener_config):
port = listener_config["port"]
bind_addresses = listener_config["bind_addresses"]
site_tag = listener_config.get("tag", port)
resources = {}
for res in listener_config["resources"]:
for name in res["names"]:
if name == "metrics":
resources[METRICS_PREFIX] = MetricsResource(self)
elif name == "client":
resource = JsonResource(self, canonical_json=False)
KeyUploadServlet(self).register(resource)
resources.update({
"/_matrix/client/r0": resource,
"/_matrix/client/unstable": resource,
"/_matrix/client/v2_alpha": resource,
"/_matrix/client/api/v1": resource,
})
root_resource = create_resource_tree(resources, Resource())
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse client reader now listening on port %d", port)
def start_listening(self, listeners):
for listener in listeners:
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
self.get_tcp_replication().start_replication(self)
def build_tcp_replication(self):
return ReplicationClientHandler(self.get_datastore())
def start(config_options):
try:
config = HomeServerConfig.load_config(
"Synapse frontend proxy", config_options
)
except ConfigError as e:
sys.stderr.write("\n" + e.message + "\n")
sys.exit(1)
assert config.worker_app == "synapse.app.frontend_proxy"
assert config.worker_main_http_uri is not None
setup_logging(config, use_worker_options=True)
events.USE_FROZEN_DICTS = config.use_frozen_dicts
database_engine = create_engine(config.database_config)
tls_server_context_factory = context_factory.ServerContextFactory(config)
ss = FrontendProxyServer(
config.server_name,
db_config=config.database_config,
tls_server_context_factory=tls_server_context_factory,
config=config,
version_string="Synapse/" + get_version_string(synapse),
database_engine=database_engine,
)
ss.setup()
ss.get_handlers()
ss.start_listening(config.worker_listeners)
def start():
ss.get_state_handler().start_caching()
ss.get_datastore().start_profiling()
reactor.callWhenRunning(start)
_base.start_worker_reactor("synapse-frontend-proxy", config)
if __name__ == '__main__':
with LoggingContext("main"):
start(sys.argv[1:])
+164 -189
View File
@@ -13,61 +13,51 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import synapse
import gc
import logging
import os
import sys
import synapse
import synapse.config.logger
from synapse import events
from synapse.api.urls import CONTENT_REPO_PREFIX, FEDERATION_PREFIX, \
LEGACY_MEDIA_PREFIX, MEDIA_PREFIX, SERVER_KEY_PREFIX, SERVER_KEY_V2_PREFIX, \
STATIC_PREFIX, WEB_CLIENT_PREFIX
from synapse.app import _base
from synapse.app._base import quit_with_error, listen_ssl, listen_tcp
from synapse.config._base import ConfigError
from synapse.python_dependencies import (
check_requirements, CONDITIONAL_REQUIREMENTS
)
from synapse.rest import ClientRestResource
from synapse.storage.engines import create_engine, IncorrectDatabaseSetup
from synapse.storage import are_all_users_on_domain
from synapse.storage.prepare_database import UpgradeDatabaseException, prepare_database
from synapse.server import HomeServer
from twisted.internet import reactor, task, defer
from twisted.application import service
from twisted.web.resource import Resource, EncodingResourceWrapper
from twisted.web.static import File
from twisted.web.server import GzipEncoderFactory
from synapse.http.server import RootRedirect
from synapse.rest.media.v0.content_repository import ContentRepoResource
from synapse.rest.media.v1.media_repository import MediaRepositoryResource
from synapse.rest.key.v1.server_key_resource import LocalKey
from synapse.rest.key.v2 import KeyApiV2Resource
from synapse.api.urls import (
FEDERATION_PREFIX, WEB_CLIENT_PREFIX, CONTENT_REPO_PREFIX,
SERVER_KEY_PREFIX, LEGACY_MEDIA_PREFIX, MEDIA_PREFIX, STATIC_PREFIX,
SERVER_KEY_V2_PREFIX,
)
from synapse.config.homeserver import HomeServerConfig
from synapse.crypto import context_factory
from synapse.util.logcontext import LoggingContext, PreserveLoggingContext
from synapse.metrics import register_memory_metrics, get_metrics_for
from synapse.metrics.resource import MetricsResource, METRICS_PREFIX
from synapse.replication.tcp.resource import ReplicationStreamProtocolFactory
from synapse.federation.transport.server import TransportLayerServer
from synapse.module_api import ModuleApi
from synapse.http.additional_resource import AdditionalResource
from synapse.http.server import RootRedirect
from synapse.http.site import SynapseSite
from synapse.metrics import register_memory_metrics
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.python_dependencies import CONDITIONAL_REQUIREMENTS, \
check_requirements
from synapse.replication.tcp.resource import ReplicationStreamProtocolFactory
from synapse.rest import ClientRestResource
from synapse.rest.key.v1.server_key_resource import LocalKey
from synapse.rest.key.v2 import KeyApiV2Resource
from synapse.rest.media.v0.content_repository import ContentRepoResource
from synapse.server import HomeServer
from synapse.storage import are_all_users_on_domain
from synapse.storage.engines import IncorrectDatabaseSetup, create_engine
from synapse.storage.prepare_database import UpgradeDatabaseException, prepare_database
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext
from synapse.util.manhole import manhole
from synapse.util.module_loader import load_module
from synapse.util.rlimit import change_resource_limit
from synapse.util.versionstring import get_version_string
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.manhole import manhole
from synapse.http.site import SynapseSite
from synapse import events
from daemonize import Daemonize
from twisted.application import service
from twisted.internet import defer, reactor
from twisted.web.resource import EncodingResourceWrapper, Resource
from twisted.web.server import GzipEncoderFactory
from twisted.web.static import File
logger = logging.getLogger("synapse.app.homeserver")
@@ -119,52 +109,18 @@ class SynapseHomeServer(HomeServer):
resources = {}
for res in listener_config["resources"]:
for name in res["names"]:
if name == "client":
client_resource = ClientRestResource(self)
if res["compress"]:
client_resource = gz_wrap(client_resource)
resources.update(self._configure_named_resource(
name, res.get("compress", False),
))
resources.update({
"/_matrix/client/api/v1": client_resource,
"/_matrix/client/r0": client_resource,
"/_matrix/client/unstable": client_resource,
"/_matrix/client/v2_alpha": client_resource,
"/_matrix/client/versions": client_resource,
})
if name == "federation":
resources.update({
FEDERATION_PREFIX: TransportLayerServer(self),
})
if name in ["static", "client"]:
resources.update({
STATIC_PREFIX: File(
os.path.join(os.path.dirname(synapse.__file__), "static")
),
})
if name in ["media", "federation", "client"]:
media_repo = MediaRepositoryResource(self)
resources.update({
MEDIA_PREFIX: media_repo,
LEGACY_MEDIA_PREFIX: media_repo,
CONTENT_REPO_PREFIX: ContentRepoResource(
self, self.config.uploads_path
),
})
if name in ["keys", "federation"]:
resources.update({
SERVER_KEY_PREFIX: LocalKey(self),
SERVER_KEY_V2_PREFIX: KeyApiV2Resource(self),
})
if name == "webclient":
resources[WEB_CLIENT_PREFIX] = build_resource_for_web_client(self)
if name == "metrics" and self.get_config().enable_metrics:
resources[METRICS_PREFIX] = MetricsResource(self)
additional_resources = listener_config.get("additional_resources", {})
logger.debug("Configuring additional resources: %r",
additional_resources)
module_api = ModuleApi(self, self.get_auth_handler())
for path, resmodule in additional_resources.items():
handler_cls, config = load_module(resmodule)
handler = handler_cls(config, module_api)
resources[path] = AdditionalResource(self, handler.handle_request)
if WEB_CLIENT_PREFIX in resources:
root_resource = RootRedirect(WEB_CLIENT_PREFIX)
@@ -174,32 +130,97 @@ class SynapseHomeServer(HomeServer):
root_resource = create_resource_tree(resources, root_resource)
if tls:
for address in bind_addresses:
reactor.listenSSL(
port,
SynapseSite(
"synapse.access.https.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
self.tls_server_context_factory,
interface=address
)
listen_ssl(
bind_addresses,
port,
SynapseSite(
"synapse.access.https.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
self.tls_server_context_factory,
)
else:
for address in bind_addresses:
reactor.listenTCP(
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
interface=address
listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse now listening on port %d", port)
def _configure_named_resource(self, name, compress=False):
"""Build a resource map for a named resource
Args:
name (str): named resource: one of "client", "federation", etc
compress (bool): whether to enable gzip compression for this
resource
Returns:
dict[str, Resource]: map from path to HTTP resource
"""
resources = {}
if name == "client":
client_resource = ClientRestResource(self)
if compress:
client_resource = gz_wrap(client_resource)
resources.update({
"/_matrix/client/api/v1": client_resource,
"/_matrix/client/r0": client_resource,
"/_matrix/client/unstable": client_resource,
"/_matrix/client/v2_alpha": client_resource,
"/_matrix/client/versions": client_resource,
})
if name == "federation":
resources.update({
FEDERATION_PREFIX: TransportLayerServer(self),
})
if name in ["static", "client"]:
resources.update({
STATIC_PREFIX: File(
os.path.join(os.path.dirname(synapse.__file__), "static")
),
})
if name in ["media", "federation", "client"]:
if self.get_config().enable_media_repo:
media_repo = self.get_media_repository_resource()
resources.update({
MEDIA_PREFIX: media_repo,
LEGACY_MEDIA_PREFIX: media_repo,
CONTENT_REPO_PREFIX: ContentRepoResource(
self, self.config.uploads_path
),
})
elif name == "media":
raise ConfigError(
"'media' resource conflicts with enable_media_repo=False",
)
if name in ["keys", "federation"]:
resources.update({
SERVER_KEY_PREFIX: LocalKey(self),
SERVER_KEY_V2_PREFIX: KeyApiV2Resource(self),
})
if name == "webclient":
resources[WEB_CLIENT_PREFIX] = build_resource_for_web_client(self)
if name == "metrics" and self.get_config().enable_metrics:
resources[METRICS_PREFIX] = MetricsResource(self)
return resources
def start_listening(self):
config = self.get_config()
@@ -207,18 +228,15 @@ class SynapseHomeServer(HomeServer):
if listener["type"] == "http":
self._listener_http(config, listener)
elif listener["type"] == "manhole":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
reactor.listenTCP(
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
),
interface=address
listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
elif listener["type"] == "replication":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
@@ -262,16 +280,6 @@ class SynapseHomeServer(HomeServer):
return db_conn
def quit_with_error(error_string):
message_lines = error_string.split("\n")
line_length = max([len(l) for l in message_lines if len(l) < 80]) + 2
sys.stderr.write("*" * line_length + '\n')
for line in message_lines:
sys.stderr.write(" %s\n" % (line.rstrip(),))
sys.stderr.write("*" * line_length + '\n')
sys.exit(1)
def setup(config_options):
"""
Args:
@@ -398,7 +406,8 @@ def run(hs):
ThreadPool._worker = profile(ThreadPool._worker)
reactor.run = profile(reactor.run)
start_time = hs.get_clock().time()
clock = hs.get_clock()
start_time = clock.time()
stats = {}
@@ -410,41 +419,23 @@ def run(hs):
if uptime < 0:
uptime = 0
# If the stats directory is empty then this is the first time we've
# reported stats.
first_time = not stats
stats["homeserver"] = hs.config.server_name
stats["timestamp"] = now
stats["uptime_seconds"] = uptime
stats["total_users"] = yield hs.get_datastore().count_all_users()
total_nonbridged_users = yield hs.get_datastore().count_nonbridged_users()
stats["total_nonbridged_users"] = total_nonbridged_users
room_count = yield hs.get_datastore().get_room_count()
stats["total_room_count"] = room_count
stats["daily_active_users"] = yield hs.get_datastore().count_daily_users()
daily_messages = yield hs.get_datastore().count_daily_messages()
if daily_messages is not None:
stats["daily_messages"] = daily_messages
else:
stats.pop("daily_messages", None)
stats["daily_active_rooms"] = yield hs.get_datastore().count_daily_active_rooms()
stats["daily_messages"] = yield hs.get_datastore().count_daily_messages()
if first_time:
# Add callbacks to report the synapse stats as metrics whenever
# prometheus requests them, typically every 30s.
# As some of the stats are expensive to calculate we only update
# them when synapse phones home to matrix.org every 24 hours.
metrics = get_metrics_for("synapse.usage")
metrics.add_callback("timestamp", lambda: stats["timestamp"])
metrics.add_callback("uptime_seconds", lambda: stats["uptime_seconds"])
metrics.add_callback("total_users", lambda: stats["total_users"])
metrics.add_callback("total_room_count", lambda: stats["total_room_count"])
metrics.add_callback(
"daily_active_users", lambda: stats["daily_active_users"]
)
metrics.add_callback(
"daily_messages", lambda: stats.get("daily_messages", 0)
)
daily_sent_messages = yield hs.get_datastore().count_daily_sent_messages()
stats["daily_sent_messages"] = daily_sent_messages
logger.info("Reporting stats to matrix.org: %s" % (stats,))
try:
@@ -456,41 +447,25 @@ def run(hs):
logger.warn("Error reporting stats: %s", e)
if hs.config.report_stats:
phone_home_task = task.LoopingCall(phone_stats_home)
logger.info("Scheduling stats reporting for 24 hour intervals")
phone_home_task.start(60 * 60 * 24, now=False)
logger.info("Scheduling stats reporting for 3 hour intervals")
clock.looping_call(phone_stats_home, 3 * 60 * 60 * 1000)
def in_thread():
# Uncomment to enable tracing of log context changes.
# sys.settrace(logcontext_tracer)
# We wait 5 minutes to send the first set of stats as the server can
# be quite busy the first few minutes
clock.call_later(5 * 60, phone_stats_home)
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
change_resource_limit(hs.config.soft_file_limit)
if hs.config.gc_thresholds:
gc.set_threshold(*hs.config.gc_thresholds)
reactor.run()
if hs.config.daemonize and hs.config.print_pidfile:
print (hs.config.pid_file)
if hs.config.daemonize:
if hs.config.print_pidfile:
print (hs.config.pid_file)
daemon = Daemonize(
app="synapse-homeserver",
pid=hs.config.pid_file,
action=lambda: in_thread(),
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
in_thread()
_base.start_reactor(
"synapse-homeserver",
hs.config.soft_file_limit,
hs.config.gc_thresholds,
hs.config.pid_file,
hs.config.daemonize,
hs.config.cpu_affinity,
logger,
)
def main():
+38 -67
View File
@@ -13,57 +13,48 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import sys
import synapse
from synapse import events
from synapse.api.urls import (
CONTENT_REPO_PREFIX, LEGACY_MEDIA_PREFIX, MEDIA_PREFIX
)
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.crypto import context_factory
from synapse.http.site import SynapseSite
from synapse.metrics.resource import MetricsResource, METRICS_PREFIX
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage._base import BaseSlavedStore
from synapse.replication.slave.storage.appservice import SlavedApplicationServiceStore
from synapse.replication.slave.storage.client_ips import SlavedClientIpStore
from synapse.replication.slave.storage.registration import SlavedRegistrationStore
from synapse.replication.slave.storage.transactions import TransactionStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.rest.media.v0.content_repository import ContentRepoResource
from synapse.rest.media.v1.media_repository import MediaRepositoryResource
from synapse.server import HomeServer
from synapse.storage.client_ips import ClientIpStore
from synapse.storage.engines import create_engine
from synapse.storage.media_repository import MediaRepositoryStore
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext, PreserveLoggingContext
from synapse.util.logcontext import LoggingContext
from synapse.util.manhole import manhole
from synapse.util.rlimit import change_resource_limit
from synapse.util.versionstring import get_version_string
from synapse.api.urls import (
CONTENT_REPO_PREFIX, LEGACY_MEDIA_PREFIX, MEDIA_PREFIX
)
from synapse.crypto import context_factory
from synapse import events
from twisted.internet import reactor
from twisted.web.resource import Resource
from daemonize import Daemonize
import sys
import logging
import gc
logger = logging.getLogger("synapse.app.media_repository")
class MediaRepositorySlavedStore(
SlavedApplicationServiceStore,
SlavedRegistrationStore,
SlavedClientIpStore,
TransactionStore,
BaseSlavedStore,
MediaRepositoryStore,
ClientIpStore,
):
pass
@@ -97,7 +88,7 @@ class MediaRepositoryServer(HomeServer):
if name == "metrics":
resources[METRICS_PREFIX] = MetricsResource(self)
elif name == "media":
media_repo = MediaRepositoryResource(self)
media_repo = self.get_media_repository_resource()
resources.update({
MEDIA_PREFIX: media_repo,
LEGACY_MEDIA_PREFIX: media_repo,
@@ -108,17 +99,16 @@ class MediaRepositoryServer(HomeServer):
root_resource = create_resource_tree(resources, Resource())
for address in bind_addresses:
reactor.listenTCP(
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
interface=address
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse media repository now listening on port %d", port)
@@ -127,18 +117,15 @@ class MediaRepositoryServer(HomeServer):
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
reactor.listenTCP(
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
),
interface=address
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
@@ -159,6 +146,13 @@ def start(config_options):
assert config.worker_app == "synapse.app.media_repository"
if config.enable_media_repo:
_base.quit_with_error(
"enable_media_repo must be disabled in the main synapse process\n"
"before the media repo can be run in a separate worker.\n"
"Please add ``enable_media_repo: false`` to the main config\n"
)
setup_logging(config, use_worker_options=True)
events.USE_FROZEN_DICTS = config.use_frozen_dicts
@@ -180,36 +174,13 @@ def start(config_options):
ss.get_handlers()
ss.start_listening(config.worker_listeners)
def run():
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
logger.info("Running")
change_resource_limit(config.soft_file_limit)
if config.gc_thresholds:
gc.set_threshold(*config.gc_thresholds)
reactor.run()
def start():
ss.get_state_handler().start_caching()
ss.get_datastore().start_profiling()
reactor.callWhenRunning(start)
if config.worker_daemonize:
daemon = Daemonize(
app="synapse-media-repository",
pid=config.worker_pid_file,
action=run,
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
run()
_base.start_worker_reactor("synapse-media-repository", config)
if __name__ == '__main__':
+30 -65
View File
@@ -13,41 +13,33 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import sys
import synapse
from synapse.server import HomeServer
from synapse import events
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.logger import setup_logging
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.http.site import SynapseSite
from synapse.metrics.resource import MetricsResource, METRICS_PREFIX
from synapse.storage.roommember import RoomMemberStore
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage.account_data import SlavedAccountDataStore
from synapse.replication.slave.storage.events import SlavedEventStore
from synapse.replication.slave.storage.pushers import SlavedPusherStore
from synapse.replication.slave.storage.receipts import SlavedReceiptsStore
from synapse.replication.slave.storage.account_data import SlavedAccountDataStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.storage.engines import create_engine
from synapse.server import HomeServer
from synapse.storage import DataStore
from synapse.storage.engines import create_engine
from synapse.storage.roommember import RoomMemberStore
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext, preserve_fn, \
PreserveLoggingContext
from synapse.util.logcontext import LoggingContext, preserve_fn
from synapse.util.manhole import manhole
from synapse.util.rlimit import change_resource_limit
from synapse.util.versionstring import get_version_string
from synapse import events
from twisted.internet import reactor, defer
from twisted.internet import defer, reactor
from twisted.web.resource import Resource
from daemonize import Daemonize
import sys
import logging
import gc
logger = logging.getLogger("synapse.app.pusher")
@@ -122,17 +114,16 @@ class PusherServer(HomeServer):
root_resource = create_resource_tree(resources, Resource())
for address in bind_addresses:
reactor.listenTCP(
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
interface=address
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse pusher now listening on port %d", port)
@@ -141,18 +132,15 @@ class PusherServer(HomeServer):
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
reactor.listenTCP(
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
),
interface=address
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
@@ -244,18 +232,6 @@ def start(config_options):
ps.setup()
ps.start_listening(config.worker_listeners)
def run():
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
logger.info("Running")
change_resource_limit(config.soft_file_limit)
if config.gc_thresholds:
gc.set_threshold(*config.gc_thresholds)
reactor.run()
def start():
ps.get_pusherpool().start()
ps.get_datastore().start_profiling()
@@ -263,18 +239,7 @@ def start(config_options):
reactor.callWhenRunning(start)
if config.worker_daemonize:
daemon = Daemonize(
app="synapse-pusher",
pid=config.worker_pid_file,
action=run,
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
run()
_base.start_worker_reactor("synapse-pusher", config)
if __name__ == '__main__':
+45 -73
View File
@@ -13,56 +13,51 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import contextlib
import logging
import sys
import synapse
from synapse.api.constants import EventTypes
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.handlers.presence import PresenceHandler, get_interested_parties
from synapse.http.site import SynapseSite
from synapse.http.server import JsonResource
from synapse.metrics.resource import MetricsResource, METRICS_PREFIX
from synapse.rest.client.v2_alpha import sync
from synapse.rest.client.v1 import events
from synapse.rest.client.v1.room import RoomInitialSyncRestServlet
from synapse.rest.client.v1.initial_sync import InitialSyncRestServlet
from synapse.http.site import SynapseSite
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage._base import BaseSlavedStore
from synapse.replication.slave.storage.events import SlavedEventStore
from synapse.replication.slave.storage.receipts import SlavedReceiptsStore
from synapse.replication.slave.storage.account_data import SlavedAccountDataStore
from synapse.replication.slave.storage.appservice import SlavedApplicationServiceStore
from synapse.replication.slave.storage.registration import SlavedRegistrationStore
from synapse.replication.slave.storage.filtering import SlavedFilteringStore
from synapse.replication.slave.storage.push_rule import SlavedPushRuleStore
from synapse.replication.slave.storage.presence import SlavedPresenceStore
from synapse.replication.slave.storage.client_ips import SlavedClientIpStore
from synapse.replication.slave.storage.deviceinbox import SlavedDeviceInboxStore
from synapse.replication.slave.storage.devices import SlavedDeviceStore
from synapse.replication.slave.storage.events import SlavedEventStore
from synapse.replication.slave.storage.filtering import SlavedFilteringStore
from synapse.replication.slave.storage.presence import SlavedPresenceStore
from synapse.replication.slave.storage.push_rule import SlavedPushRuleStore
from synapse.replication.slave.storage.receipts import SlavedReceiptsStore
from synapse.replication.slave.storage.registration import SlavedRegistrationStore
from synapse.replication.slave.storage.room import RoomStore
from synapse.replication.slave.storage.groups import SlavedGroupServerStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.rest.client.v1 import events
from synapse.rest.client.v1.initial_sync import InitialSyncRestServlet
from synapse.rest.client.v1.room import RoomInitialSyncRestServlet
from synapse.rest.client.v2_alpha import sync
from synapse.server import HomeServer
from synapse.storage.client_ips import ClientIpStore
from synapse.storage.engines import create_engine
from synapse.storage.presence import UserPresenceState
from synapse.storage.roommember import RoomMemberStore
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext, PreserveLoggingContext, preserve_fn
from synapse.util.logcontext import LoggingContext, preserve_fn
from synapse.util.manhole import manhole
from synapse.util.rlimit import change_resource_limit
from synapse.util.stringutils import random_string
from synapse.util.versionstring import get_version_string
from twisted.internet import reactor, defer
from twisted.internet import defer, reactor
from twisted.web.resource import Resource
from daemonize import Daemonize
import sys
import logging
import contextlib
import gc
logger = logging.getLogger("synapse.app.synchrotron")
@@ -75,11 +70,12 @@ class SynchrotronSlavedStore(
SlavedRegistrationStore,
SlavedFilteringStore,
SlavedPresenceStore,
SlavedGroupServerStore,
SlavedDeviceInboxStore,
SlavedDeviceStore,
SlavedClientIpStore,
RoomStore,
BaseSlavedStore,
ClientIpStore, # After BaseSlavedStore because the constructor is different
):
who_forgot_in_room = (
RoomMemberStore.__dict__["who_forgot_in_room"]
@@ -292,17 +288,16 @@ class SynchrotronServer(HomeServer):
root_resource = create_resource_tree(resources, Resource())
for address in bind_addresses:
reactor.listenTCP(
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
),
interface=address
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse synchrotron now listening on port %d", port)
@@ -311,18 +306,15 @@ class SynchrotronServer(HomeServer):
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
bind_addresses = listener["bind_addresses"]
for address in bind_addresses:
reactor.listenTCP(
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
),
interface=address
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
@@ -344,11 +336,10 @@ class SyncReplicationHandler(ReplicationClientHandler):
self.store = hs.get_datastore()
self.typing_handler = hs.get_typing_handler()
# NB this is a SynchrotronPresence, not a normal PresenceHandler
self.presence_handler = hs.get_presence_handler()
self.notifier = hs.get_notifier()
self.presence_handler.sync_callback = self.send_user_sync
def on_rdata(self, stream_name, token, rows):
super(SyncReplicationHandler, self).on_rdata(stream_name, token, rows)
@@ -409,6 +400,10 @@ class SyncReplicationHandler(ReplicationClientHandler):
)
elif stream_name == "presence":
yield self.presence_handler.process_replication_rows(token, rows)
elif stream_name == "receipts":
self.notifier.on_new_event(
"groups_key", token, users=[row.user_id for row in rows],
)
def start(config_options):
@@ -440,36 +435,13 @@ def start(config_options):
ss.setup()
ss.start_listening(config.worker_listeners)
def run():
# make sure that we run the reactor with the sentinel log context,
# otherwise other PreserveLoggingContext instances will get confused
# and complain when they see the logcontext arbitrarily swapping
# between the sentinel and `run` logcontexts.
with PreserveLoggingContext():
logger.info("Running")
change_resource_limit(config.soft_file_limit)
if config.gc_thresholds:
gc.set_threshold(*config.gc_thresholds)
reactor.run()
def start():
ss.get_datastore().start_profiling()
ss.get_state_handler().start_caching()
reactor.callWhenRunning(start)
if config.worker_daemonize:
daemon = Daemonize(
app="synapse-synchrotron",
pid=config.worker_pid_file,
action=run,
auto_close_fds=False,
verbose=True,
logger=logger,
)
daemon.start()
else:
run()
_base.start_worker_reactor("synapse-synchrotron", config)
if __name__ == '__main__':
+237
View File
@@ -0,0 +1,237 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# Copyright 2017 Vector Creations Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import sys
import synapse
from synapse import events
from synapse.app import _base
from synapse.config._base import ConfigError
from synapse.config.homeserver import HomeServerConfig
from synapse.config.logger import setup_logging
from synapse.crypto import context_factory
from synapse.http.server import JsonResource
from synapse.http.site import SynapseSite
from synapse.metrics.resource import METRICS_PREFIX, MetricsResource
from synapse.replication.slave.storage._base import BaseSlavedStore
from synapse.replication.slave.storage.appservice import SlavedApplicationServiceStore
from synapse.replication.slave.storage.client_ips import SlavedClientIpStore
from synapse.replication.slave.storage.events import SlavedEventStore
from synapse.replication.slave.storage.registration import SlavedRegistrationStore
from synapse.replication.tcp.client import ReplicationClientHandler
from synapse.rest.client.v2_alpha import user_directory
from synapse.server import HomeServer
from synapse.storage.engines import create_engine
from synapse.storage.user_directory import UserDirectoryStore
from synapse.util.caches.stream_change_cache import StreamChangeCache
from synapse.util.httpresourcetree import create_resource_tree
from synapse.util.logcontext import LoggingContext, preserve_fn
from synapse.util.manhole import manhole
from synapse.util.versionstring import get_version_string
from twisted.internet import reactor
from twisted.web.resource import Resource
logger = logging.getLogger("synapse.app.user_dir")
class UserDirectorySlaveStore(
SlavedEventStore,
SlavedApplicationServiceStore,
SlavedRegistrationStore,
SlavedClientIpStore,
UserDirectoryStore,
BaseSlavedStore,
):
def __init__(self, db_conn, hs):
super(UserDirectorySlaveStore, self).__init__(db_conn, hs)
events_max = self._stream_id_gen.get_current_token()
curr_state_delta_prefill, min_curr_state_delta_id = self._get_cache_dict(
db_conn, "current_state_delta_stream",
entity_column="room_id",
stream_column="stream_id",
max_value=events_max, # As we share the stream id with events token
limit=1000,
)
self._curr_state_delta_stream_cache = StreamChangeCache(
"_curr_state_delta_stream_cache", min_curr_state_delta_id,
prefilled_cache=curr_state_delta_prefill,
)
self._current_state_delta_pos = events_max
def stream_positions(self):
result = super(UserDirectorySlaveStore, self).stream_positions()
result["current_state_deltas"] = self._current_state_delta_pos
return result
def process_replication_rows(self, stream_name, token, rows):
if stream_name == "current_state_deltas":
self._current_state_delta_pos = token
for row in rows:
self._curr_state_delta_stream_cache.entity_has_changed(
row.room_id, token
)
return super(UserDirectorySlaveStore, self).process_replication_rows(
stream_name, token, rows
)
class UserDirectoryServer(HomeServer):
def get_db_conn(self, run_new_connection=True):
# Any param beginning with cp_ is a parameter for adbapi, and should
# not be passed to the database engine.
db_params = {
k: v for k, v in self.db_config.get("args", {}).items()
if not k.startswith("cp_")
}
db_conn = self.database_engine.module.connect(**db_params)
if run_new_connection:
self.database_engine.on_new_connection(db_conn)
return db_conn
def setup(self):
logger.info("Setting up.")
self.datastore = UserDirectorySlaveStore(self.get_db_conn(), self)
logger.info("Finished setting up.")
def _listen_http(self, listener_config):
port = listener_config["port"]
bind_addresses = listener_config["bind_addresses"]
site_tag = listener_config.get("tag", port)
resources = {}
for res in listener_config["resources"]:
for name in res["names"]:
if name == "metrics":
resources[METRICS_PREFIX] = MetricsResource(self)
elif name == "client":
resource = JsonResource(self, canonical_json=False)
user_directory.register_servlets(self, resource)
resources.update({
"/_matrix/client/r0": resource,
"/_matrix/client/unstable": resource,
"/_matrix/client/v2_alpha": resource,
"/_matrix/client/api/v1": resource,
})
root_resource = create_resource_tree(resources, Resource())
_base.listen_tcp(
bind_addresses,
port,
SynapseSite(
"synapse.access.http.%s" % (site_tag,),
site_tag,
listener_config,
root_resource,
)
)
logger.info("Synapse user_dir now listening on port %d", port)
def start_listening(self, listeners):
for listener in listeners:
if listener["type"] == "http":
self._listen_http(listener)
elif listener["type"] == "manhole":
_base.listen_tcp(
listener["bind_addresses"],
listener["port"],
manhole(
username="matrix",
password="rabbithole",
globals={"hs": self},
)
)
else:
logger.warn("Unrecognized listener type: %s", listener["type"])
self.get_tcp_replication().start_replication(self)
def build_tcp_replication(self):
return UserDirectoryReplicationHandler(self)
class UserDirectoryReplicationHandler(ReplicationClientHandler):
def __init__(self, hs):
super(UserDirectoryReplicationHandler, self).__init__(hs.get_datastore())
self.user_directory = hs.get_user_directory_handler()
def on_rdata(self, stream_name, token, rows):
super(UserDirectoryReplicationHandler, self).on_rdata(
stream_name, token, rows
)
if stream_name == "current_state_deltas":
preserve_fn(self.user_directory.notify_new_event)()
def start(config_options):
try:
config = HomeServerConfig.load_config(
"Synapse user directory", config_options
)
except ConfigError as e:
sys.stderr.write("\n" + e.message + "\n")
sys.exit(1)
assert config.worker_app == "synapse.app.user_dir"
setup_logging(config, use_worker_options=True)
events.USE_FROZEN_DICTS = config.use_frozen_dicts
database_engine = create_engine(config.database_config)
if config.update_user_directory:
sys.stderr.write(
"\nThe update_user_directory must be disabled in the main synapse process"
"\nbefore they can be run in a separate worker."
"\nPlease add ``update_user_directory: false`` to the main config"
"\n"
)
sys.exit(1)
# Force the pushers to start since they will be disabled in the main config
config.update_user_directory = True
tls_server_context_factory = context_factory.ServerContextFactory(config)
ps = UserDirectoryServer(
config.server_name,
db_config=config.database_config,
tls_server_context_factory=tls_server_context_factory,
config=config,
version_string="Synapse/" + get_version_string(synapse),
database_engine=database_engine,
)
ps.setup()
ps.start_listening(config.worker_listeners)
def start():
ps.get_datastore().start_profiling()
ps.get_state_handler().start_caching()
reactor.callWhenRunning(start)
_base.start_worker_reactor("synapse-user-dir", config)
if __name__ == '__main__':
with LoggingContext("main"):
start(sys.argv[1:])
+36 -1
View File
@@ -14,6 +14,7 @@
# limitations under the License.
from synapse.api.constants import EventTypes
from synapse.util.caches.descriptors import cachedInlineCallbacks
from synapse.types import GroupID, get_domain_from_id
from twisted.internet import defer
@@ -81,12 +82,13 @@ class ApplicationService(object):
# values.
NS_LIST = [NS_USERS, NS_ALIASES, NS_ROOMS]
def __init__(self, token, url=None, namespaces=None, hs_token=None,
def __init__(self, token, hostname, url=None, namespaces=None, hs_token=None,
sender=None, id=None, protocols=None, rate_limited=True):
self.token = token
self.url = url
self.hs_token = hs_token
self.sender = sender
self.server_name = hostname
self.namespaces = self._check_namespaces(namespaces)
self.id = id
@@ -125,6 +127,24 @@ class ApplicationService(object):
raise ValueError(
"Expected bool for 'exclusive' in ns '%s'" % ns
)
group_id = regex_obj.get("group_id")
if group_id:
if not isinstance(group_id, str):
raise ValueError(
"Expected string for 'group_id' in ns '%s'" % ns
)
try:
GroupID.from_string(group_id)
except Exception:
raise ValueError(
"Expected valid group ID for 'group_id' in ns '%s'" % ns
)
if get_domain_from_id(group_id) != self.server_name:
raise ValueError(
"Expected 'group_id' to be this host in ns '%s'" % ns
)
regex = regex_obj.get("regex")
if isinstance(regex, basestring):
regex_obj["regex"] = re.compile(regex) # Pre-compile regex
@@ -251,6 +271,21 @@ class ApplicationService(object):
if regex_obj["exclusive"]
]
def get_groups_for_user(self, user_id):
"""Get the groups that this user is associated with by this AS
Args:
user_id (str): The ID of the user.
Returns:
iterable[str]: an iterable that yields group_id strings.
"""
return (
regex_obj["group_id"]
for regex_obj in self.namespaces[ApplicationService.NS_USERS]
if "group_id" in regex_obj and regex_obj["regex"].match(user_id)
)
def is_rate_limited(self):
return self.rate_limited
+7 -3
View File
@@ -18,6 +18,7 @@ from synapse.api.constants import ThirdPartyEntityKind
from synapse.api.errors import CodeMessageException
from synapse.http.client import SimpleHttpClient
from synapse.events.utils import serialize_event
from synapse.util.logcontext import preserve_fn, make_deferred_yieldable
from synapse.util.caches.response_cache import ResponseCache
from synapse.types import ThirdPartyInstanceID
@@ -192,9 +193,12 @@ class ApplicationServiceApi(SimpleHttpClient):
defer.returnValue(None)
key = (service.id, protocol)
return self.protocol_meta_cache.get(key) or (
self.protocol_meta_cache.set(key, _get())
)
result = self.protocol_meta_cache.get(key)
if not result:
result = self.protocol_meta_cache.set(
key, preserve_fn(_get)()
)
return make_deferred_yieldable(result)
@defer.inlineCallbacks
def push_bulk(self, service, events, txn_id=None):
+1 -1
View File
@@ -123,7 +123,7 @@ class _ServiceQueuer(object):
with Measure(self.clock, "servicequeuer.send"):
try:
yield self.txn_ctrl.send(service, events)
except:
except Exception:
logger.exception("AS request failed")
finally:
self.requests_in_flight.discard(service.id)
+26 -10
View File
@@ -81,22 +81,38 @@ class Config(object):
def abspath(file_path):
return os.path.abspath(file_path) if file_path else file_path
@classmethod
def path_exists(cls, file_path):
"""Check if a file exists
Unlike os.path.exists, this throws an exception if there is an error
checking if the file exists (for example, if there is a perms error on
the parent dir).
Returns:
bool: True if the file exists; False if not.
"""
try:
os.stat(file_path)
return True
except OSError as e:
if e.errno != errno.ENOENT:
raise e
return False
@classmethod
def check_file(cls, file_path, config_name):
if file_path is None:
raise ConfigError(
"Missing config for %s."
" You must specify a path for the config file. You can "
"do this with the -c or --config-path option. "
"Adding --generate-config along with --server-name "
"<server name> will generate a config file at the given path."
% (config_name,)
)
if not os.path.exists(file_path):
try:
os.stat(file_path)
except OSError as e:
raise ConfigError(
"File %s config for %s doesn't exist."
" Try running again with --generate-config"
% (file_path, config_name,)
"Error accessing file '%s' (config for %s): %s"
% (file_path, config_name, e.strerror)
)
return cls.abspath(file_path)
@@ -248,7 +264,7 @@ class Config(object):
" -c CONFIG-FILE\""
)
(config_path,) = config_files
if not os.path.exists(config_path):
if not cls.path_exists(config_path):
if config_args.keys_directory:
config_dir_path = config_args.keys_directory
else:
@@ -261,7 +277,7 @@ class Config(object):
"Must specify a server_name to a generate config for."
" Pass -H server.name."
)
if not os.path.exists(config_dir_path):
if not cls.path_exists(config_dir_path):
os.makedirs(config_dir_path)
with open(config_path, "wb") as config_file:
config_bytes, config = obj.generate_config(
+1
View File
@@ -154,6 +154,7 @@ def _load_appservice(hostname, as_info, config_filename):
)
return ApplicationService(
token=as_info["as_token"],
hostname=hostname,
url=as_info["url"],
namespaces=as_info["namespaces"],
hs_token=as_info["hs_token"],
+1 -1
View File
@@ -41,7 +41,7 @@ class CasConfig(Config):
#cas_config:
# enabled: true
# server_url: "https://cas-server.com"
# service_url: "https://homesever.domain.com:8448"
# service_url: "https://homeserver.domain.com:8448"
# #required_attributes:
# # name: value
"""
+32
View File
@@ -0,0 +1,32 @@
# -*- coding: utf-8 -*-
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from ._base import Config
class GroupsConfig(Config):
def read_config(self, config):
self.enable_group_creation = config.get("enable_group_creation", False)
self.group_creation_prefix = config.get("group_creation_prefix", "")
def default_config(self, **kwargs):
return """\
# Whether to allow non server admins to create groups on this server
enable_group_creation: false
# If enabled, non server admins can only create groups with local parts
# starting with this prefix
# group_creation_prefix: "unofficial/"
"""
+6 -1
View File
@@ -33,6 +33,10 @@ from .jwt import JWTConfig
from .password_auth_providers import PasswordAuthProviderConfig
from .emailconfig import EmailConfig
from .workers import WorkerConfig
from .push import PushConfig
from .spam_checker import SpamCheckerConfig
from .groups import GroupsConfig
from .user_directory import UserDirectoryConfig
class HomeServerConfig(TlsConfig, ServerConfig, DatabaseConfig, LoggingConfig,
@@ -40,7 +44,8 @@ class HomeServerConfig(TlsConfig, ServerConfig, DatabaseConfig, LoggingConfig,
VoipConfig, RegistrationConfig, MetricsConfig, ApiConfig,
AppServiceConfig, KeyConfig, SAML2Config, CasConfig,
JWTConfig, PasswordConfig, EmailConfig,
WorkerConfig, PasswordAuthProviderConfig,):
WorkerConfig, PasswordAuthProviderConfig, PushConfig,
SpamCheckerConfig, GroupsConfig, UserDirectoryConfig,):
pass
+4 -4
View File
@@ -118,10 +118,9 @@ class KeyConfig(Config):
signing_keys = self.read_file(signing_key_path, "signing_key")
try:
return read_signing_keys(signing_keys.splitlines(True))
except Exception:
except Exception as e:
raise ConfigError(
"Error reading signing_key."
" Try running again with --generate-config"
"Error reading signing_key: %s" % (str(e))
)
def read_old_signing_keys(self, old_signing_keys):
@@ -141,7 +140,8 @@ class KeyConfig(Config):
def generate_files(self, config):
signing_key_path = config["signing_key_path"]
if not os.path.exists(signing_key_path):
if not self.path_exists(signing_key_path):
with open(signing_key_path, "w") as signing_key_file:
key_id = "a_" + random_string(4)
write_signing_keys(
+5 -1
View File
@@ -148,8 +148,8 @@ def setup_logging(config, use_worker_options=False):
"%(asctime)s - %(name)s - %(lineno)d - %(levelname)s - %(request)s"
" - %(message)s"
)
if log_config is None:
if log_config is None:
level = logging.INFO
level_for_storage = logging.INFO
if config.verbosity:
@@ -176,6 +176,10 @@ def setup_logging(config, use_worker_options=False):
logger.info("Opened new log file due to SIGHUP")
else:
handler = logging.StreamHandler()
def sighup(signum, stack):
pass
handler.setFormatter(formatter)
handler.addFilter(LoggingContextFilter(request=""))
+20 -23
View File
@@ -13,44 +13,41 @@
# See the License for the specific language governing permissions and
# limitations under the License.
from ._base import Config, ConfigError
from ._base import Config
import importlib
from synapse.util.module_loader import load_module
LDAP_PROVIDER = 'ldap_auth_provider.LdapAuthProvider'
class PasswordAuthProviderConfig(Config):
def read_config(self, config):
self.password_providers = []
providers = []
# We want to be backwards compatible with the old `ldap_config`
# param.
ldap_config = config.get("ldap_config", {})
self.ldap_enabled = ldap_config.get("enabled", False)
if self.ldap_enabled:
from ldap_auth_provider import LdapAuthProvider
parsed_config = LdapAuthProvider.parse_config(ldap_config)
self.password_providers.append((LdapAuthProvider, parsed_config))
if ldap_config.get("enabled", False):
providers.append({
'module': LDAP_PROVIDER,
'config': ldap_config,
})
providers = config.get("password_providers", [])
providers.extend(config.get("password_providers", []))
for provider in providers:
mod_name = provider['module']
# This is for backwards compat when the ldap auth provider resided
# in this package.
if provider['module'] == "synapse.util.ldap_auth_provider.LdapAuthProvider":
from ldap_auth_provider import LdapAuthProvider
provider_class = LdapAuthProvider
else:
# We need to import the module, and then pick the class out of
# that, so we split based on the last dot.
module, clz = provider['module'].rsplit(".", 1)
module = importlib.import_module(module)
provider_class = getattr(module, clz)
if mod_name == "synapse.util.ldap_auth_provider.LdapAuthProvider":
mod_name = LDAP_PROVIDER
(provider_class, provider_config) = load_module({
"module": mod_name,
"config": provider['config'],
})
try:
provider_config = provider_class.parse_config(provider["config"])
except Exception as e:
raise ConfigError(
"Failed to parse config for %r: %r" % (provider['module'], e)
)
self.password_providers.append((provider_class, provider_config))
def default_config(self, **kwargs):
+61
View File
@@ -0,0 +1,61 @@
# -*- coding: utf-8 -*-
# Copyright 2015, 2016 OpenMarket Ltd
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from ._base import Config
class PushConfig(Config):
def read_config(self, config):
push_config = config.get("push", {})
self.push_include_content = push_config.get("include_content", True)
# There was a a 'redact_content' setting but mistakenly read from the
# 'email'section'. Check for the flag in the 'push' section, and log,
# but do not honour it to avoid nasty surprises when people upgrade.
if push_config.get("redact_content") is not None:
print(
"The push.redact_content content option has never worked. "
"Please set push.include_content if you want this behaviour"
)
# Now check for the one in the 'email' section and honour it,
# with a warning.
push_config = config.get("email", {})
redact_content = push_config.get("redact_content")
if redact_content is not None:
print(
"The 'email.redact_content' option is deprecated: "
"please set push.include_content instead"
)
self.push_include_content = not redact_content
def default_config(self, config_dir_path, server_name, **kwargs):
return """
# Clients requesting push notifications can either have the body of
# the message sent in the notification poke along with other details
# like the sender, or just the event ID and room ID (`event_id_only`).
# If clients choose the former, this option controls whether the
# notification request includes the content of the event (other details
# like the sender are still included). For `event_id_only` push, it
# has no effect.
# For modern android devices the notification content will still appear
# because it is loaded by the app. iPhone, however will send a
# notification saying only that a message arrived and who it came from.
#
#push:
# include_content: true
"""
+7
View File
@@ -41,6 +41,8 @@ class RegistrationConfig(Config):
self.allow_guest_access and config.get("invite_3pid_guest", False)
)
self.auto_join_rooms = config.get("auto_join_rooms", [])
def default_config(self, **kwargs):
registration_shared_secret = random_string_with_symbols(50)
@@ -70,6 +72,11 @@ class RegistrationConfig(Config):
- matrix.org
- vector.im
- riot.im
# Users who register on this homeserver will automatically be joined
# to these rooms
#auto_join_rooms:
# - "#example:example.com"
""" % locals()
def add_arguments(self, parser):
+20
View File
@@ -70,7 +70,19 @@ class ContentRepositoryConfig(Config):
self.max_upload_size = self.parse_size(config["max_upload_size"])
self.max_image_pixels = self.parse_size(config["max_image_pixels"])
self.max_spider_size = self.parse_size(config["max_spider_size"])
self.media_store_path = self.ensure_directory(config["media_store_path"])
self.backup_media_store_path = config.get("backup_media_store_path")
if self.backup_media_store_path:
self.backup_media_store_path = self.ensure_directory(
self.backup_media_store_path
)
self.synchronous_backup_media_store = config.get(
"synchronous_backup_media_store", False
)
self.uploads_path = self.ensure_directory(config["uploads_path"])
self.dynamic_thumbnails = config["dynamic_thumbnails"]
self.thumbnail_requirements = parse_thumbnail_requirements(
@@ -115,6 +127,14 @@ class ContentRepositoryConfig(Config):
# Directory where uploaded images and attachments are stored.
media_store_path: "%(media_store)s"
# A secondary directory where uploaded images and attachments are
# stored as a backup.
# backup_media_store_path: "%(media_store)s"
# Whether to wait for successful write to backup media store before
# returning successfully.
# synchronous_backup_media_store: false
# Directory where in-progress uploads are stored.
uploads_path: "%(uploads_path)s"
+57 -8
View File
@@ -1,5 +1,6 @@
# -*- coding: utf-8 -*-
# Copyright 2014-2016 OpenMarket Ltd
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
@@ -29,14 +30,31 @@ class ServerConfig(Config):
self.user_agent_suffix = config.get("user_agent_suffix")
self.use_frozen_dicts = config.get("use_frozen_dicts", False)
self.public_baseurl = config.get("public_baseurl")
self.cpu_affinity = config.get("cpu_affinity")
# Whether to send federation traffic out in this process. This only
# applies to some federation traffic, and so shouldn't be used to
# "disable" federation
self.send_federation = config.get("send_federation", True)
# Whether to update the user directory or not. This should be set to
# false only if we are updating the user directory in a worker
self.update_user_directory = config.get("update_user_directory", True)
# whether to enable the media repository endpoints. This should be set
# to false if the media repository is running as a separate endpoint;
# doing so ensures that we will not run cache cleanup jobs on the
# master, potentially causing inconsistency.
self.enable_media_repo = config.get("enable_media_repo", True)
self.filter_timeline_limit = config.get("filter_timeline_limit", -1)
# Whether we should block invites sent to users on this server
# (other than those sent by local server admins)
self.block_non_admin_invites = config.get(
"block_non_admin_invites", False,
)
if self.public_baseurl is not None:
if self.public_baseurl[-1] != '/':
self.public_baseurl += '/'
@@ -143,6 +161,27 @@ class ServerConfig(Config):
# When running as a daemon, the file to store the pid in
pid_file: %(pid_file)s
# CPU affinity mask. Setting this restricts the CPUs on which the
# process will be scheduled. It is represented as a bitmask, with the
# lowest order bit corresponding to the first logical CPU and the
# highest order bit corresponding to the last logical CPU. Not all CPUs
# may exist on a given system but a mask may specify more CPUs than are
# present.
#
# For example:
# 0x00000001 is processor #0,
# 0x00000003 is processors #0 and #1,
# 0xFFFFFFFF is all processors (#0 through #31).
#
# Pinning a Python process to a single CPU is desirable, because Python
# is inherently single-threaded due to the GIL, and can suffer a
# 30-40%% slowdown due to cache blow-out and thread context switching
# if the scheduler happens to schedule the underlying threads across
# different cores. See
# https://www.mirantis.com/blog/improve-performance-python-programs-restricting-single-cpu/.
#
# cpu_affinity: 0xFFFFFFFF
# Whether to serve a web client from the HTTP/HTTPS root resource.
web_client: True
@@ -167,6 +206,10 @@ class ServerConfig(Config):
# and sync operations. The default value is -1, means no upper limit.
# filter_timeline_limit: 5000
# Whether room invites to users on this server should be blocked
# (except those sent by local server admins). The default is False.
# block_non_admin_invites: True
# List of ports that Synapse should listen on, their purpose and their
# configuration.
listeners:
@@ -177,13 +220,12 @@ class ServerConfig(Config):
port: %(bind_port)s
# Local addresses to listen on.
# This will listen on all IPv4 addresses by default.
# On Linux and Mac OS, `::` will listen on all IPv4 and IPv6
# addresses by default. For most other OSes, this will only listen
# on IPv6.
bind_addresses:
- '::'
- '0.0.0.0'
# Uncomment to listen on all IPv6 interfaces
# N.B: On at least Linux this will also listen on all IPv4
# addresses, so you will need to comment out the line above.
# - '::'
# This is a 'http' listener, allows us to specify 'resources'.
type: http
@@ -210,11 +252,18 @@ class ServerConfig(Config):
- names: [federation] # Federation APIs
compress: false
# optional list of additional endpoints which can be loaded via
# dynamic modules
# additional_resources:
# "/_matrix/my/custom/endpoint":
# module: my_module.CustomRequestHandler
# config: {}
# Unsecure HTTP listener,
# For when matrix traffic passes through loadbalancer that unwraps TLS.
- port: %(unsecure_port)s
tls: false
bind_addresses: ['0.0.0.0']
bind_addresses: ['::', '0.0.0.0']
type: http
x_forwarded: false
@@ -228,7 +277,7 @@ class ServerConfig(Config):
# Turn on the twisted ssh manhole service on localhost on the given
# port.
# - port: 9000
# bind_address: 127.0.0.1
# bind_addresses: ['::1', '127.0.0.1']
# type: manhole
""" % locals()
@@ -266,7 +315,7 @@ def read_gc_thresholds(thresholds):
return (
int(thresholds[0]), int(thresholds[1]), int(thresholds[2]),
)
except:
except Exception:
raise ConfigError(
"Value of `gc_threshold` must be a list of three integers if set"
)
+35
View File
@@ -0,0 +1,35 @@
# -*- coding: utf-8 -*-
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from synapse.util.module_loader import load_module
from ._base import Config
class SpamCheckerConfig(Config):
def read_config(self, config):
self.spam_checker = None
provider = config.get("spam_checker", None)
if provider is not None:
self.spam_checker = load_module(provider)
def default_config(self, **kwargs):
return """\
# spam_checker:
# module: "my_custom_project.SuperSpamChecker"
# config:
# example_option: 'things'
"""
+9 -3
View File
@@ -109,6 +109,12 @@ class TlsConfig(Config):
# key. It may be necessary to publish the fingerprints of a new
# certificate and wait until the "valid_until_ts" of the previous key
# responses have passed before deploying it.
#
# You can calculate a fingerprint from a given TLS listener via:
# openssl s_client -connect $host:$port < /dev/null 2> /dev/null |
# openssl x509 -outform DER | openssl sha256 -binary | base64 | tr -d '='
# or by checking matrix.org/federationtester/api/report?server_name=$host
#
tls_fingerprints: []
# tls_fingerprints: [{"sha256": "<base64_encoded_sha256_fingerprint>"}]
""" % locals()
@@ -126,7 +132,7 @@ class TlsConfig(Config):
tls_private_key_path = config["tls_private_key_path"]
tls_dh_params_path = config["tls_dh_params_path"]
if not os.path.exists(tls_private_key_path):
if not self.path_exists(tls_private_key_path):
with open(tls_private_key_path, "w") as private_key_file:
tls_private_key = crypto.PKey()
tls_private_key.generate_key(crypto.TYPE_RSA, 2048)
@@ -141,7 +147,7 @@ class TlsConfig(Config):
crypto.FILETYPE_PEM, private_key_pem
)
if not os.path.exists(tls_certificate_path):
if not self.path_exists(tls_certificate_path):
with open(tls_certificate_path, "w") as certificate_file:
cert = crypto.X509()
subject = cert.get_subject()
@@ -159,7 +165,7 @@ class TlsConfig(Config):
certificate_file.write(cert_pem)
if not os.path.exists(tls_dh_params_path):
if not self.path_exists(tls_dh_params_path):
if GENERATE_DH_PARAMS:
subprocess.check_call([
"openssl", "dhparam",
+44
View File
@@ -0,0 +1,44 @@
# -*- coding: utf-8 -*-
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from ._base import Config
class UserDirectoryConfig(Config):
"""User Directory Configuration
Configuration for the behaviour of the /user_directory API
"""
def read_config(self, config):
self.user_directory_search_all_users = False
user_directory_config = config.get("user_directory", None)
if user_directory_config:
self.user_directory_search_all_users = (
user_directory_config.get("search_all_users", False)
)
def default_config(self, config_dir_path, server_name, **kwargs):
return """
# User Directory configuration
#
# 'search_all_users' defines whether to search all users visible to your HS
# when searching the user directory, rather than limiting to users visible
# in public rooms. Defaults to false. If you set it True, you'll have to run
# UPDATE user_directory_stream_pos SET stream_id = NULL;
# on your database to tell it to rebuild the user_directory search indexes.
#
#user_directory:
# search_all_users: false
"""
+3
View File
@@ -32,6 +32,9 @@ class WorkerConfig(Config):
self.worker_replication_port = config.get("worker_replication_port", None)
self.worker_name = config.get("worker_name", self.worker_app)
self.worker_main_http_uri = config.get("worker_main_http_uri", None)
self.worker_cpu_affinity = config.get("worker_cpu_affinity")
if self.worker_listeners:
for listener in self.worker_listeners:
bind_address = listener.pop("bind_address", None)
+1 -1
View File
@@ -34,7 +34,7 @@ class ServerContextFactory(ssl.ContextFactory):
try:
_ecCurve = _OpenSSLECCurve(_defaultCurveName)
_ecCurve.addECKeyToContext(context)
except:
except Exception:
logger.exception("Failed to enable elliptic curve for TLS")
context.set_options(SSL.OP_NO_SSLv2 | SSL.OP_NO_SSLv3)
context.use_certificate_chain_file(config.tls_certificate_file)
+11 -4
View File
@@ -32,18 +32,25 @@ def check_event_content_hash(event, hash_algorithm=hashlib.sha256):
"""Check whether the hash for this PDU matches the contents"""
name, expected_hash = compute_content_hash(event, hash_algorithm)
logger.debug("Expecting hash: %s", encode_base64(expected_hash))
if name not in event.hashes:
# some malformed events lack a 'hashes'. Protect against it being missing
# or a weird type by basically treating it the same as an unhashed event.
hashes = event.get("hashes")
if not isinstance(hashes, dict):
raise SynapseError(400, "Malformed 'hashes'", Codes.UNAUTHORIZED)
if name not in hashes:
raise SynapseError(
400,
"Algorithm %s not in hashes %s" % (
name, list(event.hashes),
name, list(hashes),
),
Codes.UNAUTHORIZED,
)
message_hash_base64 = event.hashes[name]
message_hash_base64 = hashes[name]
try:
message_hash_bytes = decode_base64(message_hash_base64)
except:
except Exception:
raise SynapseError(
400,
"Invalid base64: %s" % (message_hash_base64,),
+5 -12
View File
@@ -13,14 +13,11 @@
# See the License for the specific language governing permissions and
# limitations under the License.
from synapse.util import logcontext
from twisted.web.http import HTTPClient
from twisted.internet.protocol import Factory
from twisted.internet import defer, reactor
from synapse.http.endpoint import matrix_federation_endpoint
from synapse.util.logcontext import (
preserve_context_over_fn, preserve_context_over_deferred
)
import simplejson as json
import logging
@@ -43,14 +40,10 @@ def fetch_server_key(server_name, ssl_context_factory, path=KEY_API_V1):
for i in range(5):
try:
protocol = yield preserve_context_over_fn(
endpoint.connect, factory
)
server_response, server_certificate = yield preserve_context_over_deferred(
protocol.remote_key
)
defer.returnValue((server_response, server_certificate))
return
with logcontext.PreserveLoggingContext():
protocol = yield endpoint.connect(factory)
server_response, server_certificate = yield protocol.remote_key
defer.returnValue((server_response, server_certificate))
except SynapseKeyClientError as e:
logger.exception("Error getting key for %r" % (server_name,))
if e.status.startswith("4"):
+166 -129
View File
@@ -1,5 +1,6 @@
# -*- coding: utf-8 -*-
# Copyright 2014-2016 OpenMarket Ltd
# Copyright 2017 New Vector Ltd.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
@@ -15,10 +16,9 @@
from synapse.crypto.keyclient import fetch_server_key
from synapse.api.errors import SynapseError, Codes
from synapse.util import unwrapFirstError
from synapse.util.async import ObservableDeferred
from synapse.util import unwrapFirstError, logcontext
from synapse.util.logcontext import (
preserve_context_over_deferred, preserve_context_over_fn, PreserveLoggingContext,
PreserveLoggingContext,
preserve_fn
)
from synapse.util.metrics import Measure
@@ -57,7 +57,8 @@ Attributes:
json_object(dict): The JSON object to verify.
deferred(twisted.internet.defer.Deferred):
A deferred (server_name, key_id, verify_key) tuple that resolves when
a verify key has been fetched
a verify key has been fetched. The deferreds' callbacks are run with no
logcontext.
"""
@@ -74,23 +75,32 @@ class Keyring(object):
self.perspective_servers = self.config.perspectives
self.hs = hs
# map from server name to Deferred. Has an entry for each server with
# an ongoing key download; the Deferred completes once the download
# completes.
#
# These are regular, logcontext-agnostic Deferreds.
self.key_downloads = {}
def verify_json_for_server(self, server_name, json_object):
return self.verify_json_objects_for_server(
[(server_name, json_object)]
)[0]
return logcontext.make_deferred_yieldable(
self.verify_json_objects_for_server(
[(server_name, json_object)]
)[0]
)
def verify_json_objects_for_server(self, server_and_json):
"""Bulk verfies signatures of json objects, bulk fetching keys as
"""Bulk verifies signatures of json objects, bulk fetching keys as
necessary.
Args:
server_and_json (list): List of pairs of (server_name, json_object)
Returns:
list of deferreds indicating success or failure to verify each
json object's signature for the given server_name.
List<Deferred>: for each input pair, a deferred indicating success
or failure to verify each json object's signature for the given
server_name. The deferreds run their callbacks in the sentinel
logcontext.
"""
verify_requests = []
@@ -117,94 +127,72 @@ class Keyring(object):
verify_requests.append(verify_request)
@defer.inlineCallbacks
def handle_key_deferred(verify_request):
server_name = verify_request.server_name
try:
_, key_id, verify_key = yield verify_request.deferred
except IOError as e:
logger.warn(
"Got IOError when downloading keys for %s: %s %s",
server_name, type(e).__name__, str(e.message),
)
raise SynapseError(
502,
"Error downloading keys for %s" % (server_name,),
Codes.UNAUTHORIZED,
)
except Exception as e:
logger.exception(
"Got Exception when downloading keys for %s: %s %s",
server_name, type(e).__name__, str(e.message),
)
raise SynapseError(
401,
"No key for %s with id %s" % (server_name, key_ids),
Codes.UNAUTHORIZED,
)
json_object = verify_request.json_object
logger.debug("Got key %s %s:%s for server %s, verifying" % (
key_id, verify_key.alg, verify_key.version, server_name,
))
try:
verify_signed_json(json_object, server_name, verify_key)
except:
raise SynapseError(
401,
"Invalid signature for server %s with key %s:%s" % (
server_name, verify_key.alg, verify_key.version
),
Codes.UNAUTHORIZED,
)
server_to_deferred = {
server_name: defer.Deferred()
for server_name, _ in server_and_json
}
with PreserveLoggingContext():
# We want to wait for any previous lookups to complete before
# proceeding.
wait_on_deferred = self.wait_for_previous_lookups(
[server_name for server_name, _ in server_and_json],
server_to_deferred,
)
# Actually start fetching keys.
wait_on_deferred.addBoth(
lambda _: self.get_server_verify_keys(verify_requests)
)
# When we've finished fetching all the keys for a given server_name,
# resolve the deferred passed to `wait_for_previous_lookups` so that
# any lookups waiting will proceed.
server_to_request_ids = {}
def remove_deferreds(res, server_name, verify_request):
request_id = id(verify_request)
server_to_request_ids[server_name].discard(request_id)
if not server_to_request_ids[server_name]:
d = server_to_deferred.pop(server_name, None)
if d:
d.callback(None)
return res
for verify_request in verify_requests:
server_name = verify_request.server_name
request_id = id(verify_request)
server_to_request_ids.setdefault(server_name, set()).add(request_id)
deferred.addBoth(remove_deferreds, server_name, verify_request)
preserve_fn(self._start_key_lookups)(verify_requests)
# Pass those keys to handle_key_deferred so that the json object
# signatures can be verified
handle = preserve_fn(_handle_key_deferred)
return [
preserve_context_over_fn(handle_key_deferred, verify_request)
for verify_request in verify_requests
handle(rq) for rq in verify_requests
]
@defer.inlineCallbacks
def _start_key_lookups(self, verify_requests):
"""Sets off the key fetches for each verify request
Once each fetch completes, verify_request.deferred will be resolved.
Args:
verify_requests (List[VerifyKeyRequest]):
"""
# create a deferred for each server we're going to look up the keys
# for; we'll resolve them once we have completed our lookups.
# These will be passed into wait_for_previous_lookups to block
# any other lookups until we have finished.
# The deferreds are called with no logcontext.
server_to_deferred = {
rq.server_name: defer.Deferred()
for rq in verify_requests
}
# We want to wait for any previous lookups to complete before
# proceeding.
yield self.wait_for_previous_lookups(
[rq.server_name for rq in verify_requests],
server_to_deferred,
)
# Actually start fetching keys.
self._get_server_verify_keys(verify_requests)
# When we've finished fetching all the keys for a given server_name,
# resolve the deferred passed to `wait_for_previous_lookups` so that
# any lookups waiting will proceed.
#
# map from server name to a set of request ids
server_to_request_ids = {}
for verify_request in verify_requests:
server_name = verify_request.server_name
request_id = id(verify_request)
server_to_request_ids.setdefault(server_name, set()).add(request_id)
def remove_deferreds(res, verify_request):
server_name = verify_request.server_name
request_id = id(verify_request)
server_to_request_ids[server_name].discard(request_id)
if not server_to_request_ids[server_name]:
d = server_to_deferred.pop(server_name, None)
if d:
d.callback(None)
return res
for verify_request in verify_requests:
verify_request.deferred.addBoth(
remove_deferreds, verify_request,
)
@defer.inlineCallbacks
def wait_for_previous_lookups(self, server_names, server_to_deferred):
"""Waits for any previous key lookups for the given servers to finish.
@@ -212,7 +200,13 @@ class Keyring(object):
Args:
server_names (list): list of server_names we want to lookup
server_to_deferred (dict): server_name to deferred which gets
resolved once we've finished looking up keys for that server
resolved once we've finished looking up keys for that server.
The Deferreds should be regular twisted ones which call their
callbacks with no logcontext.
Returns: a Deferred which resolves once all key lookups for the given
servers have completed. Follows the synapse rules of logcontext
preservation.
"""
while True:
wait_on = [
@@ -226,17 +220,15 @@ class Keyring(object):
else:
break
def rm(r, server_name_):
self.key_downloads.pop(server_name_, None)
return r
for server_name, deferred in server_to_deferred.items():
d = ObservableDeferred(preserve_context_over_deferred(deferred))
self.key_downloads[server_name] = d
self.key_downloads[server_name] = deferred
deferred.addBoth(rm, server_name)
def rm(r, server_name):
self.key_downloads.pop(server_name, None)
return r
d.addBoth(rm, server_name)
def get_server_verify_keys(self, verify_requests):
def _get_server_verify_keys(self, verify_requests):
"""Tries to find at least one key for each verify request
For each verify_request, verify_request.deferred is called back with
@@ -305,21 +297,23 @@ class Keyring(object):
if not missing_keys:
break
for verify_request in requests_missing_keys.values():
verify_request.deferred.errback(SynapseError(
401,
"No key for %s with id %s" % (
verify_request.server_name, verify_request.key_ids,
),
Codes.UNAUTHORIZED,
))
with PreserveLoggingContext():
for verify_request in requests_missing_keys:
verify_request.deferred.errback(SynapseError(
401,
"No key for %s with id %s" % (
verify_request.server_name, verify_request.key_ids,
),
Codes.UNAUTHORIZED,
))
def on_err(err):
for verify_request in verify_requests:
if not verify_request.deferred.called:
verify_request.deferred.errback(err)
with PreserveLoggingContext():
for verify_request in verify_requests:
if not verify_request.deferred.called:
verify_request.deferred.errback(err)
do_iterations().addErrback(on_err)
preserve_fn(do_iterations)().addErrback(on_err)
@defer.inlineCallbacks
def get_keys_from_store(self, server_name_and_key_ids):
@@ -333,7 +327,7 @@ class Keyring(object):
Deferred: resolves to dict[str, dict[str, VerifyKey]]: map from
server_name -> key_id -> VerifyKey
"""
res = yield preserve_context_over_deferred(defer.gatherResults(
res = yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(self.store.get_server_verify_keys)(
server_name, key_ids
@@ -341,7 +335,7 @@ class Keyring(object):
for server_name, key_ids in server_name_and_key_ids
],
consumeErrors=True,
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError))
defer.returnValue(dict(res))
@@ -362,13 +356,13 @@ class Keyring(object):
)
defer.returnValue({})
results = yield preserve_context_over_deferred(defer.gatherResults(
results = yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(get_key)(p_name, p_keys)
for p_name, p_keys in self.perspective_servers.items()
],
consumeErrors=True,
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError))
union_of_keys = {}
for result in results:
@@ -402,13 +396,13 @@ class Keyring(object):
defer.returnValue(keys)
results = yield preserve_context_over_deferred(defer.gatherResults(
results = yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(get_key)(server_name, key_ids)
for server_name, key_ids in server_name_and_key_ids
],
consumeErrors=True,
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError))
merged = {}
for result in results:
@@ -485,7 +479,7 @@ class Keyring(object):
for server_name, response_keys in processed_response.items():
keys.setdefault(server_name, {}).update(response_keys)
yield preserve_context_over_deferred(defer.gatherResults(
yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(self.store_keys)(
server_name=server_name,
@@ -495,7 +489,7 @@ class Keyring(object):
for server_name, response_keys in keys.items()
],
consumeErrors=True
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError))
defer.returnValue(keys)
@@ -543,7 +537,7 @@ class Keyring(object):
keys.update(response_keys)
yield preserve_context_over_deferred(defer.gatherResults(
yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(self.store_keys)(
server_name=key_server_name,
@@ -553,7 +547,7 @@ class Keyring(object):
for key_server_name, verify_keys in keys.items()
],
consumeErrors=True
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError))
defer.returnValue(keys)
@@ -619,7 +613,7 @@ class Keyring(object):
response_keys.update(verify_keys)
response_keys.update(old_verify_keys)
yield preserve_context_over_deferred(defer.gatherResults(
yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(self.store.store_server_keys_json)(
server_name=server_name,
@@ -632,7 +626,7 @@ class Keyring(object):
for key_id in updated_key_ids
],
consumeErrors=True,
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError))
results[server_name] = response_keys
@@ -710,7 +704,6 @@ class Keyring(object):
defer.returnValue(verify_keys)
@defer.inlineCallbacks
def store_keys(self, server_name, from_server, verify_keys):
"""Store a collection of verify keys for a given server
Args:
@@ -721,7 +714,7 @@ class Keyring(object):
A deferred that completes when the keys are stored.
"""
# TODO(markjh): Store whether the keys have expired.
yield preserve_context_over_deferred(defer.gatherResults(
return logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(self.store.store_server_verify_key)(
server_name, server_name, key.time_added, key
@@ -729,4 +722,48 @@ class Keyring(object):
for key_id, key in verify_keys.items()
],
consumeErrors=True,
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError))
@defer.inlineCallbacks
def _handle_key_deferred(verify_request):
server_name = verify_request.server_name
try:
with PreserveLoggingContext():
_, key_id, verify_key = yield verify_request.deferred
except IOError as e:
logger.warn(
"Got IOError when downloading keys for %s: %s %s",
server_name, type(e).__name__, str(e.message),
)
raise SynapseError(
502,
"Error downloading keys for %s" % (server_name,),
Codes.UNAUTHORIZED,
)
except Exception as e:
logger.exception(
"Got Exception when downloading keys for %s: %s %s",
server_name, type(e).__name__, str(e.message),
)
raise SynapseError(
401,
"No key for %s with id %s" % (server_name, verify_request.key_ids),
Codes.UNAUTHORIZED,
)
json_object = verify_request.json_object
logger.debug("Got key %s %s:%s for server %s, verifying" % (
key_id, verify_key.alg, verify_key.version, server_name,
))
try:
verify_signed_json(json_object, server_name, verify_key)
except Exception:
raise SynapseError(
401,
"Invalid signature for server %s with key %s:%s" % (
server_name, verify_key.alg, verify_key.version
),
Codes.UNAUTHORIZED,
)
+5 -5
View File
@@ -443,12 +443,12 @@ def _check_power_levels(event, auth_events):
for k, v in user_list.items():
try:
UserID.from_string(k)
except:
except Exception:
raise SynapseError(400, "Not a valid user_id: %s" % (k,))
try:
int(v)
except:
except Exception:
raise SynapseError(400, "Not a valid power level: %s" % (v,))
key = (event.type, event.state_key, )
@@ -470,14 +470,14 @@ def _check_power_levels(event, auth_events):
("invite", None),
]
old_list = current_state.content.get("users")
old_list = current_state.content.get("users", {})
for user in set(old_list.keys() + user_list.keys()):
levels_to_check.append(
(user, "users")
)
old_list = current_state.content.get("events")
new_list = event.content.get("events")
old_list = current_state.content.get("events", {})
new_list = event.content.get("events", {})
for ev_id in set(old_list.keys() + new_list.keys()):
levels_to_check.append(
(ev_id, "events")
+1 -1
View File
@@ -55,7 +55,7 @@ class EventBuilderFactory(object):
local_part = str(int(self.clock.time())) + i + random_string(5)
e_id = EventID.create(local_part, self.hostname)
e_id = EventID(local_part, self.hostname)
return e_id.to_string()
+113
View File
@@ -0,0 +1,113 @@
# -*- coding: utf-8 -*-
# Copyright 2017 New Vector Ltd.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
class SpamChecker(object):
def __init__(self, hs):
self.spam_checker = None
module = None
config = None
try:
module, config = hs.config.spam_checker
except Exception:
pass
if module is not None:
self.spam_checker = module(config=config)
def check_event_for_spam(self, event):
"""Checks if a given event is considered "spammy" by this server.
If the server considers an event spammy, then it will be rejected if
sent by a local user. If it is sent by a user on another server, then
users receive a blank event.
Args:
event (synapse.events.EventBase): the event to be checked
Returns:
bool: True if the event is spammy.
"""
if self.spam_checker is None:
return False
return self.spam_checker.check_event_for_spam(event)
def user_may_invite(self, inviter_userid, invitee_userid, room_id):
"""Checks if a given user may send an invite
If this method returns false, the invite will be rejected.
Args:
userid (string): The sender's user ID
Returns:
bool: True if the user may send an invite, otherwise False
"""
if self.spam_checker is None:
return True
return self.spam_checker.user_may_invite(inviter_userid, invitee_userid, room_id)
def user_may_create_room(self, userid):
"""Checks if a given user may create a room
If this method returns false, the creation request will be rejected.
Args:
userid (string): The sender's user ID
Returns:
bool: True if the user may create a room, otherwise False
"""
if self.spam_checker is None:
return True
return self.spam_checker.user_may_create_room(userid)
def user_may_create_room_alias(self, userid, room_alias):
"""Checks if a given user may create a room alias
If this method returns false, the association request will be rejected.
Args:
userid (string): The sender's user ID
room_alias (string): The alias to be created
Returns:
bool: True if the user may create a room alias, otherwise False
"""
if self.spam_checker is None:
return True
return self.spam_checker.user_may_create_room_alias(userid, room_alias)
def user_may_publish_room(self, userid, room_id):
"""Checks if a given user may publish a room to the directory
If this method returns false, the publish request will be rejected.
Args:
userid (string): The sender's user ID
room_id (string): The ID of the room that would be published
Returns:
bool: True if the user may publish the room, otherwise False
"""
if self.spam_checker is None:
return True
return self.spam_checker.user_may_publish_room(userid, room_id)
+73 -64
View File
@@ -12,28 +12,20 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from twisted.internet import defer
from synapse.events.utils import prune_event
from synapse.crypto.event_signing import check_event_content_hash
from synapse.api.errors import SynapseError
from synapse.util import unwrapFirstError
from synapse.util.logcontext import preserve_fn, preserve_context_over_deferred
import logging
from synapse.api.errors import SynapseError
from synapse.crypto.event_signing import check_event_content_hash
from synapse.events.utils import prune_event
from synapse.util import unwrapFirstError, logcontext
from twisted.internet import defer
logger = logging.getLogger(__name__)
class FederationBase(object):
def __init__(self, hs):
pass
self.spam_checker = hs.get_spam_checker()
@defer.inlineCallbacks
def _check_sigs_and_hash_and_fetch(self, origin, pdus, outlier=False,
@@ -57,56 +49,52 @@ class FederationBase(object):
"""
deferreds = self._check_sigs_and_hashes(pdus)
def callback(pdu):
return pdu
@defer.inlineCallbacks
def handle_check_result(pdu, deferred):
try:
res = yield logcontext.make_deferred_yieldable(deferred)
except SynapseError:
res = None
def errback(failure, pdu):
failure.trap(SynapseError)
return None
def try_local_db(res, pdu):
if not res:
# Check local db.
return self.store.get_event(
res = yield self.store.get_event(
pdu.event_id,
allow_rejected=True,
allow_none=True,
)
return res
def try_remote(res, pdu):
if not res and pdu.origin != origin:
return self.get_pdu(
destinations=[pdu.origin],
event_id=pdu.event_id,
outlier=outlier,
timeout=10000,
).addErrback(lambda e: None)
return res
try:
res = yield self.get_pdu(
destinations=[pdu.origin],
event_id=pdu.event_id,
outlier=outlier,
timeout=10000,
)
except SynapseError:
pass
def warn(res, pdu):
if not res:
logger.warn(
"Failed to find copy of %s with valid signature",
pdu.event_id,
)
return res
for pdu, deferred in zip(pdus, deferreds):
deferred.addCallbacks(
callback, errback, errbackArgs=[pdu]
).addCallback(
try_local_db, pdu
).addCallback(
try_remote, pdu
).addCallback(
warn, pdu
defer.returnValue(res)
handle = logcontext.preserve_fn(handle_check_result)
deferreds2 = [
handle(pdu, deferred)
for pdu, deferred in zip(pdus, deferreds)
]
valid_pdus = yield logcontext.make_deferred_yieldable(
defer.gatherResults(
deferreds2,
consumeErrors=True,
)
valid_pdus = yield preserve_context_over_deferred(defer.gatherResults(
deferreds,
consumeErrors=True
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError)
if include_none:
defer.returnValue(valid_pdus)
@@ -114,15 +102,24 @@ class FederationBase(object):
defer.returnValue([p for p in valid_pdus if p])
def _check_sigs_and_hash(self, pdu):
return self._check_sigs_and_hashes([pdu])[0]
return logcontext.make_deferred_yieldable(
self._check_sigs_and_hashes([pdu])[0],
)
def _check_sigs_and_hashes(self, pdus):
"""Throws a SynapseError if a PDU does not have the correct
signatures.
"""Checks that each of the received events is correctly signed by the
sending server.
Args:
pdus (list[FrozenEvent]): the events to be checked
Returns:
FrozenEvent: Either the given event or it redacted if it failed the
content hash check.
list[Deferred]: for each input event, a deferred which:
* returns the original event if the checks pass
* returns a redacted version of the event (if the signature
matched but the hash did not)
* throws a SynapseError if the signature check failed.
The deferreds run their callbacks in the sentinel logcontext.
"""
redacted_pdus = [
@@ -130,26 +127,38 @@ class FederationBase(object):
for pdu in pdus
]
deferreds = preserve_fn(self.keyring.verify_json_objects_for_server)([
deferreds = self.keyring.verify_json_objects_for_server([
(p.origin, p.get_pdu_json())
for p in redacted_pdus
])
ctx = logcontext.LoggingContext.current_context()
def callback(_, pdu, redacted):
if not check_event_content_hash(pdu):
logger.warn(
"Event content has been tampered, redacting %s: %s",
pdu.event_id, pdu.get_pdu_json()
)
return redacted
return pdu
with logcontext.PreserveLoggingContext(ctx):
if not check_event_content_hash(pdu):
logger.warn(
"Event content has been tampered, redacting %s: %s",
pdu.event_id, pdu.get_pdu_json()
)
return redacted
if self.spam_checker.check_event_for_spam(pdu):
logger.warn(
"Event contains spam, redacting %s: %s",
pdu.event_id, pdu.get_pdu_json()
)
return redacted
return pdu
def errback(failure, pdu):
failure.trap(SynapseError)
logger.warn(
"Signature check failed for %s",
pdu.event_id,
)
with logcontext.PreserveLoggingContext(ctx):
logger.warn(
"Signature check failed for %s",
pdu.event_id,
)
return failure
for deferred, pdu, redacted in zip(deferreds, pdus, redacted_pdus):
+6 -6
View File
@@ -22,10 +22,10 @@ from synapse.api.constants import Membership
from synapse.api.errors import (
CodeMessageException, HttpResponseException, SynapseError,
)
from synapse.util import unwrapFirstError
from synapse.util import unwrapFirstError, logcontext
from synapse.util.caches.expiringcache import ExpiringCache
from synapse.util.logutils import log_function
from synapse.util.logcontext import preserve_fn, preserve_context_over_deferred
from synapse.util.logcontext import make_deferred_yieldable, preserve_fn
from synapse.events import FrozenEvent, builder
import synapse.metrics
@@ -189,10 +189,10 @@ class FederationClient(FederationBase):
]
# FIXME: We should handle signature failures more gracefully.
pdus[:] = yield preserve_context_over_deferred(defer.gatherResults(
pdus[:] = yield logcontext.make_deferred_yieldable(defer.gatherResults(
self._check_sigs_and_hashes(pdus),
consumeErrors=True,
)).addErrback(unwrapFirstError)
).addErrback(unwrapFirstError))
defer.returnValue(pdus)
@@ -252,7 +252,7 @@ class FederationClient(FederationBase):
pdu = pdu_list[0]
# Check signatures are correct.
signed_pdu = yield self._check_sigs_and_hashes([pdu])[0]
signed_pdu = yield self._check_sigs_and_hash(pdu)
break
@@ -420,7 +420,7 @@ class FederationClient(FederationBase):
for e_id in batch
]
res = yield preserve_context_over_deferred(
res = yield make_deferred_yieldable(
defer.DeferredList(deferreds, consumeErrors=True)
)
for success, result in res:
+112 -62
View File
@@ -12,14 +12,13 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from twisted.internet import defer
from .federation_base import FederationBase
from .units import Transaction, Edu
from synapse.util.async import Linearizer
from synapse.util import async
from synapse.util.logcontext import make_deferred_yieldable, preserve_fn
from synapse.util.logutils import log_function
from synapse.util.caches.response_cache import ResponseCache
from synapse.events import FrozenEvent
@@ -33,6 +32,9 @@ from synapse.crypto.event_signing import compute_event_signature
import simplejson as json
import logging
# when processing incoming transactions, we try to handle multiple rooms in
# parallel, up to this limit.
TRANSACTION_CONCURRENCY_LIMIT = 10
logger = logging.getLogger(__name__)
@@ -52,7 +54,8 @@ class FederationServer(FederationBase):
self.auth = hs.get_auth()
self._server_linearizer = Linearizer("fed_server")
self._server_linearizer = async.Linearizer("fed_server")
self._transaction_linearizer = async.Linearizer("fed_txn_handler")
# We cache responses to state queries, as they take a while and often
# come in waves.
@@ -109,25 +112,41 @@ class FederationServer(FederationBase):
@defer.inlineCallbacks
@log_function
def on_incoming_transaction(self, transaction_data):
# keep this as early as possible to make the calculated origin ts as
# accurate as possible.
request_time = self._clock.time_msec()
transaction = Transaction(**transaction_data)
received_pdus_counter.inc_by(len(transaction.pdus))
for p in transaction.pdus:
if "unsigned" in p:
unsigned = p["unsigned"]
if "age" in unsigned:
p["age"] = unsigned["age"]
if "age" in p:
p["age_ts"] = int(self._clock.time_msec()) - int(p["age"])
del p["age"]
pdu_list = [
self.event_from_pdu_json(p) for p in transaction.pdus
]
if not transaction.transaction_id:
raise Exception("Transaction missing transaction_id")
if not transaction.origin:
raise Exception("Transaction missing origin")
logger.debug("[%s] Got transaction", transaction.transaction_id)
# use a linearizer to ensure that we don't process the same transaction
# multiple times in parallel.
with (yield self._transaction_linearizer.queue(
(transaction.origin, transaction.transaction_id),
)):
result = yield self._handle_incoming_transaction(
transaction, request_time,
)
defer.returnValue(result)
@defer.inlineCallbacks
def _handle_incoming_transaction(self, transaction, request_time):
""" Process an incoming transaction and return the HTTP response
Args:
transaction (Transaction): incoming transaction
request_time (int): timestamp that the HTTP request arrived at
Returns:
Deferred[(int, object)]: http response code and body
"""
response = yield self.transaction_actions.have_responded(transaction)
if response:
@@ -140,42 +159,49 @@ class FederationServer(FederationBase):
logger.debug("[%s] Transaction is new", transaction.transaction_id)
results = []
received_pdus_counter.inc_by(len(transaction.pdus))
for pdu in pdu_list:
# check that it's actually being sent from a valid destination to
# workaround bug #1753 in 0.18.5 and 0.18.6
if transaction.origin != get_domain_from_id(pdu.event_id):
# We continue to accept join events from any server; this is
# necessary for the federation join dance to work correctly.
# (When we join over federation, the "helper" server is
# responsible for sending out the join event, rather than the
# origin. See bug #1893).
if not (
pdu.type == 'm.room.member' and
pdu.content and
pdu.content.get("membership", None) == 'join'
):
logger.info(
"Discarding PDU %s from invalid origin %s",
pdu.event_id, transaction.origin
)
continue
else:
logger.info(
"Accepting join PDU %s from %s",
pdu.event_id, transaction.origin
)
pdus_by_room = {}
try:
yield self._handle_received_pdu(transaction.origin, pdu)
results.append({})
except FederationError as e:
self.send_failure(e, transaction.origin)
results.append({"error": str(e)})
except Exception as e:
results.append({"error": str(e)})
logger.exception("Failed to handle PDU")
for p in transaction.pdus:
if "unsigned" in p:
unsigned = p["unsigned"]
if "age" in unsigned:
p["age"] = unsigned["age"]
if "age" in p:
p["age_ts"] = request_time - int(p["age"])
del p["age"]
event = self.event_from_pdu_json(p)
room_id = event.room_id
pdus_by_room.setdefault(room_id, []).append(event)
pdu_results = {}
# we can process different rooms in parallel (which is useful if they
# require callouts to other servers to fetch missing events), but
# impose a limit to avoid going too crazy with ram/cpu.
@defer.inlineCallbacks
def process_pdus_for_room(room_id):
logger.debug("Processing PDUs for %s", room_id)
for pdu in pdus_by_room[room_id]:
event_id = pdu.event_id
try:
yield self._handle_received_pdu(
transaction.origin, pdu
)
pdu_results[event_id] = {}
except FederationError as e:
logger.warn("Error handling PDU %s: %s", event_id, e)
pdu_results[event_id] = {"error": str(e)}
except Exception as e:
pdu_results[event_id] = {"error": str(e)}
logger.exception("Failed to handle PDU %s", event_id)
yield async.concurrently_execute(
process_pdus_for_room, pdus_by_room.keys(),
TRANSACTION_CONCURRENCY_LIMIT,
)
if hasattr(transaction, "edus"):
for edu in (Edu(**x) for x in transaction.edus):
@@ -185,17 +211,16 @@ class FederationServer(FederationBase):
edu.content
)
for failure in getattr(transaction, "pdu_failures", []):
logger.info("Got failure %r", failure)
logger.debug("Returning: %s", str(results))
pdu_failures = getattr(transaction, "pdu_failures", [])
for failure in pdu_failures:
logger.info("Got failure %r", failure)
response = {
"pdus": dict(zip(
(p.event_id for p in pdu_list), results
)),
"pdus": pdu_results,
}
logger.debug("Returning: %s", str(response))
yield self.transaction_actions.set_response(
transaction,
200, response
@@ -229,12 +254,13 @@ class FederationServer(FederationBase):
result = self._state_resp_cache.get((room_id, event_id))
if not result:
with (yield self._server_linearizer.queue((origin, room_id))):
resp = yield self._state_resp_cache.set(
d = self._state_resp_cache.set(
(room_id, event_id),
self._on_context_state_request_compute(room_id, event_id)
preserve_fn(self._on_context_state_request_compute)(room_id, event_id)
)
resp = yield make_deferred_yieldable(d)
else:
resp = yield result
resp = yield make_deferred_yieldable(result)
defer.returnValue((200, resp))
@@ -520,6 +546,30 @@ class FederationServer(FederationBase):
Returns (Deferred): completes with None
Raises: FederationError if the signatures / hash do not match
"""
# check that it's actually being sent from a valid destination to
# workaround bug #1753 in 0.18.5 and 0.18.6
if origin != get_domain_from_id(pdu.event_id):
# We continue to accept join events from any server; this is
# necessary for the federation join dance to work correctly.
# (When we join over federation, the "helper" server is
# responsible for sending out the join event, rather than the
# origin. See bug #1893).
if not (
pdu.type == 'm.room.member' and
pdu.content and
pdu.content.get("membership", None) == 'join'
):
logger.info(
"Discarding PDU %s from invalid origin %s",
pdu.event_id, origin
)
return
else:
logger.info(
"Accepting join PDU %s from %s",
pdu.event_id, origin
)
# Check signature.
try:
pdu = yield self._check_sigs_and_hash(pdu)
+40 -17
View File
@@ -20,8 +20,8 @@ from .persistence import TransactionActions
from .units import Transaction, Edu
from synapse.api.errors import HttpResponseException
from synapse.util import logcontext, PreserveLoggingContext
from synapse.util.async import run_on_reactor
from synapse.util.logcontext import preserve_context_over_fn, preserve_fn
from synapse.util.retryutils import NotRetryingDestination, get_retry_limiter
from synapse.util.metrics import measure_func
from synapse.handlers.presence import format_user_presence_state, get_interested_remotes
@@ -146,7 +146,6 @@ class TransactionQueue(object):
else:
return not destination.startswith("localhost")
@defer.inlineCallbacks
def notify_new_events(self, current_id):
"""This gets called when we have some new events we might want to
send out to other servers.
@@ -156,6 +155,13 @@ class TransactionQueue(object):
if self._is_processing:
return
# fire off a processing loop in the background. It's likely it will
# outlast the current request, so run it in the sentinel logcontext.
with PreserveLoggingContext():
self._process_event_queue_loop()
@defer.inlineCallbacks
def _process_event_queue_loop(self):
try:
self._is_processing = True
while True:
@@ -187,6 +193,7 @@ class TransactionQueue(object):
prev_id for prev_id, _ in event.prev_events
],
)
destinations = set(destinations)
if send_on_behalf_of is not None:
# If we are sending the event on behalf of another server
@@ -230,11 +237,9 @@ class TransactionQueue(object):
(pdu, order)
)
preserve_context_over_fn(
self._attempt_new_transaction, destination
)
self._attempt_new_transaction(destination)
@preserve_fn # the caller should not yield on this
@logcontext.preserve_fn # the caller should not yield on this
@defer.inlineCallbacks
def send_presence(self, states):
"""Send the new presence states to the appropriate destinations.
@@ -298,7 +303,7 @@ class TransactionQueue(object):
state.user_id: state for state in states
})
preserve_fn(self._attempt_new_transaction)(destination)
self._attempt_new_transaction(destination)
def send_edu(self, destination, edu_type, content, key=None):
edu = Edu(
@@ -320,9 +325,7 @@ class TransactionQueue(object):
else:
self.pending_edus_by_dest.setdefault(destination, []).append(edu)
preserve_context_over_fn(
self._attempt_new_transaction, destination
)
self._attempt_new_transaction(destination)
def send_failure(self, failure, destination):
if destination == self.server_name or destination == "localhost":
@@ -335,9 +338,7 @@ class TransactionQueue(object):
destination, []
).append(failure)
preserve_context_over_fn(
self._attempt_new_transaction, destination
)
self._attempt_new_transaction(destination)
def send_device_messages(self, destination):
if destination == self.server_name or destination == "localhost":
@@ -346,15 +347,24 @@ class TransactionQueue(object):
if not self.can_send_to(destination):
return
preserve_context_over_fn(
self._attempt_new_transaction, destination
)
self._attempt_new_transaction(destination)
def get_current_token(self):
return 0
@defer.inlineCallbacks
def _attempt_new_transaction(self, destination):
"""Try to start a new transaction to this destination
If there is already a transaction in progress to this destination,
returns immediately. Otherwise kicks off the process of sending a
transaction in the background.
Args:
destination (str):
Returns:
None
"""
# list of (pending_pdu, deferred, order)
if destination in self.pending_transactions:
# XXX: pending_transactions can get stuck on by a never-ending
@@ -367,6 +377,19 @@ class TransactionQueue(object):
)
return
logger.debug("TX [%s] Starting transaction loop", destination)
# Drop the logcontext before starting the transaction. It doesn't
# really make sense to log all the outbound transactions against
# whatever path led us to this point: that's pretty arbitrary really.
#
# (this also means we can fire off _perform_transaction without
# yielding)
with logcontext.PreserveLoggingContext():
self._transaction_transmission_loop(destination)
@defer.inlineCallbacks
def _transaction_transmission_loop(self, destination):
pending_pdus = []
try:
self.pending_transactions[destination] = 1
+415
View File
@@ -471,3 +471,418 @@ class TransportLayerClient(object):
)
defer.returnValue(content)
@log_function
def get_group_profile(self, destination, group_id, requester_user_id):
"""Get a group profile
"""
path = PREFIX + "/groups/%s/profile" % (group_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def update_group_profile(self, destination, group_id, requester_user_id, content):
"""Update a remote group profile
Args:
destination (str)
group_id (str)
requester_user_id (str)
content (dict): The new profile of the group
"""
path = PREFIX + "/groups/%s/profile" % (group_id,)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
data=content,
ignore_backoff=True,
)
@log_function
def get_group_summary(self, destination, group_id, requester_user_id):
"""Get a group summary
"""
path = PREFIX + "/groups/%s/summary" % (group_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def get_rooms_in_group(self, destination, group_id, requester_user_id):
"""Get all rooms in a group
"""
path = PREFIX + "/groups/%s/rooms" % (group_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
def add_room_to_group(self, destination, group_id, requester_user_id, room_id,
content):
"""Add a room to a group
"""
path = PREFIX + "/groups/%s/room/%s" % (group_id, room_id,)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
data=content,
ignore_backoff=True,
)
def update_room_in_group(self, destination, group_id, requester_user_id, room_id,
config_key, content):
"""Update room in group
"""
path = PREFIX + "/groups/%s/room/%s/config/%s" % (group_id, room_id, config_key,)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
data=content,
ignore_backoff=True,
)
def remove_room_from_group(self, destination, group_id, requester_user_id, room_id):
"""Remove a room from a group
"""
path = PREFIX + "/groups/%s/room/%s" % (group_id, room_id,)
return self.client.delete_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def get_users_in_group(self, destination, group_id, requester_user_id):
"""Get users in a group
"""
path = PREFIX + "/groups/%s/users" % (group_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def get_invited_users_in_group(self, destination, group_id, requester_user_id):
"""Get users that have been invited to a group
"""
path = PREFIX + "/groups/%s/invited_users" % (group_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def accept_group_invite(self, destination, group_id, user_id, content):
"""Accept a group invite
"""
path = PREFIX + "/groups/%s/users/%s/accept_invite" % (group_id, user_id)
return self.client.post_json(
destination=destination,
path=path,
data=content,
ignore_backoff=True,
)
@log_function
def invite_to_group(self, destination, group_id, user_id, requester_user_id, content):
"""Invite a user to a group
"""
path = PREFIX + "/groups/%s/users/%s/invite" % (group_id, user_id)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
data=content,
ignore_backoff=True,
)
@log_function
def invite_to_group_notification(self, destination, group_id, user_id, content):
"""Sent by group server to inform a user's server that they have been
invited.
"""
path = PREFIX + "/groups/local/%s/users/%s/invite" % (group_id, user_id)
return self.client.post_json(
destination=destination,
path=path,
data=content,
ignore_backoff=True,
)
@log_function
def remove_user_from_group(self, destination, group_id, requester_user_id,
user_id, content):
"""Remove a user fron a group
"""
path = PREFIX + "/groups/%s/users/%s/remove" % (group_id, user_id)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
data=content,
ignore_backoff=True,
)
@log_function
def remove_user_from_group_notification(self, destination, group_id, user_id,
content):
"""Sent by group server to inform a user's server that they have been
kicked from the group.
"""
path = PREFIX + "/groups/local/%s/users/%s/remove" % (group_id, user_id)
return self.client.post_json(
destination=destination,
path=path,
data=content,
ignore_backoff=True,
)
@log_function
def renew_group_attestation(self, destination, group_id, user_id, content):
"""Sent by either a group server or a user's server to periodically update
the attestations
"""
path = PREFIX + "/groups/%s/renew_attestation/%s" % (group_id, user_id)
return self.client.post_json(
destination=destination,
path=path,
data=content,
ignore_backoff=True,
)
@log_function
def update_group_summary_room(self, destination, group_id, user_id, room_id,
category_id, content):
"""Update a room entry in a group summary
"""
if category_id:
path = PREFIX + "/groups/%s/summary/categories/%s/rooms/%s" % (
group_id, category_id, room_id,
)
else:
path = PREFIX + "/groups/%s/summary/rooms/%s" % (group_id, room_id,)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": user_id},
data=content,
ignore_backoff=True,
)
@log_function
def delete_group_summary_room(self, destination, group_id, user_id, room_id,
category_id):
"""Delete a room entry in a group summary
"""
if category_id:
path = PREFIX + "/groups/%s/summary/categories/%s/rooms/%s" % (
group_id, category_id, room_id,
)
else:
path = PREFIX + "/groups/%s/summary/rooms/%s" % (group_id, room_id,)
return self.client.delete_json(
destination=destination,
path=path,
args={"requester_user_id": user_id},
ignore_backoff=True,
)
@log_function
def get_group_categories(self, destination, group_id, requester_user_id):
"""Get all categories in a group
"""
path = PREFIX + "/groups/%s/categories" % (group_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def get_group_category(self, destination, group_id, requester_user_id, category_id):
"""Get category info in a group
"""
path = PREFIX + "/groups/%s/categories/%s" % (group_id, category_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def update_group_category(self, destination, group_id, requester_user_id, category_id,
content):
"""Update a category in a group
"""
path = PREFIX + "/groups/%s/categories/%s" % (group_id, category_id,)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
data=content,
ignore_backoff=True,
)
@log_function
def delete_group_category(self, destination, group_id, requester_user_id,
category_id):
"""Delete a category in a group
"""
path = PREFIX + "/groups/%s/categories/%s" % (group_id, category_id,)
return self.client.delete_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def get_group_roles(self, destination, group_id, requester_user_id):
"""Get all roles in a group
"""
path = PREFIX + "/groups/%s/roles" % (group_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def get_group_role(self, destination, group_id, requester_user_id, role_id):
"""Get a roles info
"""
path = PREFIX + "/groups/%s/roles/%s" % (group_id, role_id,)
return self.client.get_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def update_group_role(self, destination, group_id, requester_user_id, role_id,
content):
"""Update a role in a group
"""
path = PREFIX + "/groups/%s/roles/%s" % (group_id, role_id,)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
data=content,
ignore_backoff=True,
)
@log_function
def delete_group_role(self, destination, group_id, requester_user_id, role_id):
"""Delete a role in a group
"""
path = PREFIX + "/groups/%s/roles/%s" % (group_id, role_id,)
return self.client.delete_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
@log_function
def update_group_summary_user(self, destination, group_id, requester_user_id,
user_id, role_id, content):
"""Update a users entry in a group
"""
if role_id:
path = PREFIX + "/groups/%s/summary/roles/%s/users/%s" % (
group_id, role_id, user_id,
)
else:
path = PREFIX + "/groups/%s/summary/users/%s" % (group_id, user_id,)
return self.client.post_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
data=content,
ignore_backoff=True,
)
@log_function
def delete_group_summary_user(self, destination, group_id, requester_user_id,
user_id, role_id):
"""Delete a users entry in a group
"""
if role_id:
path = PREFIX + "/groups/%s/summary/roles/%s/users/%s" % (
group_id, role_id, user_id,
)
else:
path = PREFIX + "/groups/%s/summary/users/%s" % (group_id, user_id,)
return self.client.delete_json(
destination=destination,
path=path,
args={"requester_user_id": requester_user_id},
ignore_backoff=True,
)
def bulk_get_publicised_groups(self, destination, user_ids):
"""Get the groups a list of users are publicising
"""
path = PREFIX + "/get_groups_publicised"
content = {"user_ids": user_ids}
return self.client.post_json(
destination=destination,
path=path,
data=content,
ignore_backoff=True,
)
+584 -12
View File
@@ -25,7 +25,7 @@ from synapse.http.servlet import (
from synapse.util.ratelimitutils import FederationRateLimiter
from synapse.util.versionstring import get_version_string
from synapse.util.logcontext import preserve_fn
from synapse.types import ThirdPartyInstanceID
from synapse.types import ThirdPartyInstanceID, get_domain_from_id
import functools
import logging
@@ -112,7 +112,7 @@ class Authenticator(object):
key = strip_quotes(param_dict["key"])
sig = strip_quotes(param_dict["sig"])
return (origin, key, sig)
except:
except Exception:
raise AuthenticationError(
400, "Malformed Authorization header", Codes.UNAUTHORIZED
)
@@ -153,12 +153,10 @@ class Authenticator(object):
class BaseFederationServlet(object):
REQUIRE_AUTH = True
def __init__(self, handler, authenticator, ratelimiter, server_name,
room_list_handler):
def __init__(self, handler, authenticator, ratelimiter, server_name):
self.handler = handler
self.authenticator = authenticator
self.ratelimiter = ratelimiter
self.room_list_handler = room_list_handler
def _wrap(self, func):
authenticator = self.authenticator
@@ -179,7 +177,7 @@ class BaseFederationServlet(object):
if self.REQUIRE_AUTH:
logger.exception("authenticate_request failed")
raise
except:
except Exception:
logger.exception("authenticate_request failed")
raise
@@ -272,7 +270,7 @@ class FederationSendServlet(BaseFederationServlet):
code, response = yield self.handler.on_incoming_transaction(
transaction_data
)
except:
except Exception:
logger.exception("on_incoming_transaction failed")
raise
@@ -590,7 +588,7 @@ class PublicRoomList(BaseFederationServlet):
else:
network_tuple = ThirdPartyInstanceID(None, None)
data = yield self.room_list_handler.get_local_public_room_list(
data = yield self.handler.get_local_public_room_list(
limit, since_token,
network_tuple=network_tuple
)
@@ -611,7 +609,515 @@ class FederationVersionServlet(BaseFederationServlet):
}))
SERVLET_CLASSES = (
class FederationGroupsProfileServlet(BaseFederationServlet):
"""Get/set the basic profile of a group on behalf of a user
"""
PATH = "/groups/(?P<group_id>[^/]*)/profile$"
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.get_group_profile(
group_id, requester_user_id
)
defer.returnValue((200, new_content))
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.update_group_profile(
group_id, requester_user_id, content
)
defer.returnValue((200, new_content))
class FederationGroupsSummaryServlet(BaseFederationServlet):
PATH = "/groups/(?P<group_id>[^/]*)/summary$"
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.get_group_summary(
group_id, requester_user_id
)
defer.returnValue((200, new_content))
class FederationGroupsRoomsServlet(BaseFederationServlet):
"""Get the rooms in a group on behalf of a user
"""
PATH = "/groups/(?P<group_id>[^/]*)/rooms$"
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.get_rooms_in_group(
group_id, requester_user_id
)
defer.returnValue((200, new_content))
class FederationGroupsAddRoomsServlet(BaseFederationServlet):
"""Add/remove room from group
"""
PATH = "/groups/(?P<group_id>[^/]*)/room/(?P<room_id>[^/]*)$"
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, room_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.add_room_to_group(
group_id, requester_user_id, room_id, content
)
defer.returnValue((200, new_content))
@defer.inlineCallbacks
def on_DELETE(self, origin, content, query, group_id, room_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.remove_room_from_group(
group_id, requester_user_id, room_id,
)
defer.returnValue((200, new_content))
class FederationGroupsAddRoomsConfigServlet(BaseFederationServlet):
"""Update room config in group
"""
PATH = (
"/groups/(?P<group_id>[^/]*)/room/(?P<room_id>[^/]*)"
"/config/(?P<config_key>[^/]*)$"
)
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, room_id, config_key):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
result = yield self.groups_handler.update_room_in_group(
group_id, requester_user_id, room_id, config_key, content,
)
defer.returnValue((200, result))
class FederationGroupsUsersServlet(BaseFederationServlet):
"""Get the users in a group on behalf of a user
"""
PATH = "/groups/(?P<group_id>[^/]*)/users$"
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.get_users_in_group(
group_id, requester_user_id
)
defer.returnValue((200, new_content))
class FederationGroupsInvitedUsersServlet(BaseFederationServlet):
"""Get the users that have been invited to a group
"""
PATH = "/groups/(?P<group_id>[^/]*)/invited_users$"
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.get_invited_users_in_group(
group_id, requester_user_id
)
defer.returnValue((200, new_content))
class FederationGroupsInviteServlet(BaseFederationServlet):
"""Ask a group server to invite someone to the group
"""
PATH = "/groups/(?P<group_id>[^/]*)/users/(?P<user_id>[^/]*)/invite$"
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, user_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.invite_to_group(
group_id, user_id, requester_user_id, content,
)
defer.returnValue((200, new_content))
class FederationGroupsAcceptInviteServlet(BaseFederationServlet):
"""Accept an invitation from the group server
"""
PATH = "/groups/(?P<group_id>[^/]*)/users/(?P<user_id>[^/]*)/accept_invite$"
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, user_id):
if get_domain_from_id(user_id) != origin:
raise SynapseError(403, "user_id doesn't match origin")
new_content = yield self.handler.accept_invite(
group_id, user_id, content,
)
defer.returnValue((200, new_content))
class FederationGroupsRemoveUserServlet(BaseFederationServlet):
"""Leave or kick a user from the group
"""
PATH = "/groups/(?P<group_id>[^/]*)/users/(?P<user_id>[^/]*)/remove$"
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, user_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
new_content = yield self.handler.remove_user_from_group(
group_id, user_id, requester_user_id, content,
)
defer.returnValue((200, new_content))
class FederationGroupsLocalInviteServlet(BaseFederationServlet):
"""A group server has invited a local user
"""
PATH = "/groups/local/(?P<group_id>[^/]*)/users/(?P<user_id>[^/]*)/invite$"
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, user_id):
if get_domain_from_id(group_id) != origin:
raise SynapseError(403, "group_id doesn't match origin")
new_content = yield self.handler.on_invite(
group_id, user_id, content,
)
defer.returnValue((200, new_content))
class FederationGroupsRemoveLocalUserServlet(BaseFederationServlet):
"""A group server has removed a local user
"""
PATH = "/groups/local/(?P<group_id>[^/]*)/users/(?P<user_id>[^/]*)/remove$"
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, user_id):
if get_domain_from_id(group_id) != origin:
raise SynapseError(403, "user_id doesn't match origin")
new_content = yield self.handler.user_removed_from_group(
group_id, user_id, content,
)
defer.returnValue((200, new_content))
class FederationGroupsRenewAttestaionServlet(BaseFederationServlet):
"""A group or user's server renews their attestation
"""
PATH = "/groups/(?P<group_id>[^/]*)/renew_attestation/(?P<user_id>[^/]*)$"
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, user_id):
# We don't need to check auth here as we check the attestation signatures
new_content = yield self.handler.on_renew_attestation(
group_id, user_id, content
)
defer.returnValue((200, new_content))
class FederationGroupsSummaryRoomsServlet(BaseFederationServlet):
"""Add/remove a room from the group summary, with optional category.
Matches both:
- /groups/:group/summary/rooms/:room_id
- /groups/:group/summary/categories/:category/rooms/:room_id
"""
PATH = (
"/groups/(?P<group_id>[^/]*)/summary"
"(/categories/(?P<category_id>[^/]+))?"
"/rooms/(?P<room_id>[^/]*)$"
)
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, category_id, room_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
if category_id == "":
raise SynapseError(400, "category_id cannot be empty string")
resp = yield self.handler.update_group_summary_room(
group_id, requester_user_id,
room_id=room_id,
category_id=category_id,
content=content,
)
defer.returnValue((200, resp))
@defer.inlineCallbacks
def on_DELETE(self, origin, content, query, group_id, category_id, room_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
if category_id == "":
raise SynapseError(400, "category_id cannot be empty string")
resp = yield self.handler.delete_group_summary_room(
group_id, requester_user_id,
room_id=room_id,
category_id=category_id,
)
defer.returnValue((200, resp))
class FederationGroupsCategoriesServlet(BaseFederationServlet):
"""Get all categories for a group
"""
PATH = (
"/groups/(?P<group_id>[^/]*)/categories/$"
)
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
resp = yield self.handler.get_group_categories(
group_id, requester_user_id,
)
defer.returnValue((200, resp))
class FederationGroupsCategoryServlet(BaseFederationServlet):
"""Add/remove/get a category in a group
"""
PATH = (
"/groups/(?P<group_id>[^/]*)/categories/(?P<category_id>[^/]+)$"
)
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id, category_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
resp = yield self.handler.get_group_category(
group_id, requester_user_id, category_id
)
defer.returnValue((200, resp))
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, category_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
if category_id == "":
raise SynapseError(400, "category_id cannot be empty string")
resp = yield self.handler.upsert_group_category(
group_id, requester_user_id, category_id, content,
)
defer.returnValue((200, resp))
@defer.inlineCallbacks
def on_DELETE(self, origin, content, query, group_id, category_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
if category_id == "":
raise SynapseError(400, "category_id cannot be empty string")
resp = yield self.handler.delete_group_category(
group_id, requester_user_id, category_id,
)
defer.returnValue((200, resp))
class FederationGroupsRolesServlet(BaseFederationServlet):
"""Get roles in a group
"""
PATH = (
"/groups/(?P<group_id>[^/]*)/roles/$"
)
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
resp = yield self.handler.get_group_roles(
group_id, requester_user_id,
)
defer.returnValue((200, resp))
class FederationGroupsRoleServlet(BaseFederationServlet):
"""Add/remove/get a role in a group
"""
PATH = (
"/groups/(?P<group_id>[^/]*)/roles/(?P<role_id>[^/]+)$"
)
@defer.inlineCallbacks
def on_GET(self, origin, content, query, group_id, role_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
resp = yield self.handler.get_group_role(
group_id, requester_user_id, role_id
)
defer.returnValue((200, resp))
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, role_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
if role_id == "":
raise SynapseError(400, "role_id cannot be empty string")
resp = yield self.handler.update_group_role(
group_id, requester_user_id, role_id, content,
)
defer.returnValue((200, resp))
@defer.inlineCallbacks
def on_DELETE(self, origin, content, query, group_id, role_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
if role_id == "":
raise SynapseError(400, "role_id cannot be empty string")
resp = yield self.handler.delete_group_role(
group_id, requester_user_id, role_id,
)
defer.returnValue((200, resp))
class FederationGroupsSummaryUsersServlet(BaseFederationServlet):
"""Add/remove a user from the group summary, with optional role.
Matches both:
- /groups/:group/summary/users/:user_id
- /groups/:group/summary/roles/:role/users/:user_id
"""
PATH = (
"/groups/(?P<group_id>[^/]*)/summary"
"(/roles/(?P<role_id>[^/]+))?"
"/users/(?P<user_id>[^/]*)$"
)
@defer.inlineCallbacks
def on_POST(self, origin, content, query, group_id, role_id, user_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
if role_id == "":
raise SynapseError(400, "role_id cannot be empty string")
resp = yield self.handler.update_group_summary_user(
group_id, requester_user_id,
user_id=user_id,
role_id=role_id,
content=content,
)
defer.returnValue((200, resp))
@defer.inlineCallbacks
def on_DELETE(self, origin, content, query, group_id, role_id, user_id):
requester_user_id = parse_string_from_args(query, "requester_user_id")
if get_domain_from_id(requester_user_id) != origin:
raise SynapseError(403, "requester_user_id doesn't match origin")
if role_id == "":
raise SynapseError(400, "role_id cannot be empty string")
resp = yield self.handler.delete_group_summary_user(
group_id, requester_user_id,
user_id=user_id,
role_id=role_id,
)
defer.returnValue((200, resp))
class FederationGroupsBulkPublicisedServlet(BaseFederationServlet):
"""Get roles in a group
"""
PATH = (
"/get_groups_publicised$"
)
@defer.inlineCallbacks
def on_POST(self, origin, content, query):
resp = yield self.handler.bulk_get_publicised_groups(
content["user_ids"], proxy=False,
)
defer.returnValue((200, resp))
FEDERATION_SERVLET_CLASSES = (
FederationSendServlet,
FederationPullServlet,
FederationEventServlet,
@@ -634,17 +1140,83 @@ SERVLET_CLASSES = (
FederationThirdPartyInviteExchangeServlet,
On3pidBindServlet,
OpenIdUserInfo,
PublicRoomList,
FederationVersionServlet,
)
ROOM_LIST_CLASSES = (
PublicRoomList,
)
GROUP_SERVER_SERVLET_CLASSES = (
FederationGroupsProfileServlet,
FederationGroupsSummaryServlet,
FederationGroupsRoomsServlet,
FederationGroupsUsersServlet,
FederationGroupsInvitedUsersServlet,
FederationGroupsInviteServlet,
FederationGroupsAcceptInviteServlet,
FederationGroupsRemoveUserServlet,
FederationGroupsSummaryRoomsServlet,
FederationGroupsCategoriesServlet,
FederationGroupsCategoryServlet,
FederationGroupsRolesServlet,
FederationGroupsRoleServlet,
FederationGroupsSummaryUsersServlet,
FederationGroupsAddRoomsServlet,
FederationGroupsAddRoomsConfigServlet,
)
GROUP_LOCAL_SERVLET_CLASSES = (
FederationGroupsLocalInviteServlet,
FederationGroupsRemoveLocalUserServlet,
FederationGroupsBulkPublicisedServlet,
)
GROUP_ATTESTATION_SERVLET_CLASSES = (
FederationGroupsRenewAttestaionServlet,
)
def register_servlets(hs, resource, authenticator, ratelimiter):
for servletclass in SERVLET_CLASSES:
for servletclass in FEDERATION_SERVLET_CLASSES:
servletclass(
handler=hs.get_replication_layer(),
authenticator=authenticator,
ratelimiter=ratelimiter,
server_name=hs.hostname,
room_list_handler=hs.get_room_list_handler(),
).register(resource)
for servletclass in ROOM_LIST_CLASSES:
servletclass(
handler=hs.get_room_list_handler(),
authenticator=authenticator,
ratelimiter=ratelimiter,
server_name=hs.hostname,
).register(resource)
for servletclass in GROUP_SERVER_SERVLET_CLASSES:
servletclass(
handler=hs.get_groups_server_handler(),
authenticator=authenticator,
ratelimiter=ratelimiter,
server_name=hs.hostname,
).register(resource)
for servletclass in GROUP_LOCAL_SERVLET_CLASSES:
servletclass(
handler=hs.get_groups_local_handler(),
authenticator=authenticator,
ratelimiter=ratelimiter,
server_name=hs.hostname,
).register(resource)
for servletclass in GROUP_ATTESTATION_SERVLET_CLASSES:
servletclass(
handler=hs.get_groups_attestation_renewer(),
authenticator=authenticator,
ratelimiter=ratelimiter,
server_name=hs.hostname,
).register(resource)
View File
+195
View File
@@ -0,0 +1,195 @@
# -*- coding: utf-8 -*-
# Copyright 2017 Vector Creations Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Attestations ensure that users and groups can't lie about their memberships.
When a user joins a group the HS and GS swap attestations, which allow them
both to independently prove to third parties their membership.These
attestations have a validity period so need to be periodically renewed.
If a user leaves (or gets kicked out of) a group, either side can still use
their attestation to "prove" their membership, until the attestation expires.
Therefore attestations shouldn't be relied on to prove membership in important
cases, but can for less important situtations, e.g. showing a users membership
of groups on their profile, showing flairs, etc.abs
An attestsation is a signed blob of json that looks like:
{
"user_id": "@foo:a.example.com",
"group_id": "+bar:b.example.com",
"valid_until_ms": 1507994728530,
"signatures":{"matrix.org":{"ed25519:auto":"..."}}
}
"""
import logging
import random
from twisted.internet import defer
from synapse.api.errors import SynapseError
from synapse.types import get_domain_from_id
from synapse.util.logcontext import preserve_fn
from signedjson.sign import sign_json
logger = logging.getLogger(__name__)
# Default validity duration for new attestations we create
DEFAULT_ATTESTATION_LENGTH_MS = 3 * 24 * 60 * 60 * 1000
# We add some jitter to the validity duration of attestations so that if we
# add lots of users at once we don't need to renew them all at once.
# The jitter is a multiplier picked randomly between the first and second number
DEFAULT_ATTESTATION_JITTER = (0.9, 1.3)
# Start trying to update our attestations when they come this close to expiring
UPDATE_ATTESTATION_TIME_MS = 1 * 24 * 60 * 60 * 1000
class GroupAttestationSigning(object):
"""Creates and verifies group attestations.
"""
def __init__(self, hs):
self.keyring = hs.get_keyring()
self.clock = hs.get_clock()
self.server_name = hs.hostname
self.signing_key = hs.config.signing_key[0]
@defer.inlineCallbacks
def verify_attestation(self, attestation, group_id, user_id, server_name=None):
"""Verifies that the given attestation matches the given parameters.
An optional server_name can be supplied to explicitly set which server's
signature is expected. Otherwise assumes that either the group_id or user_id
is local and uses the other's server as the one to check.
"""
if not server_name:
if get_domain_from_id(group_id) == self.server_name:
server_name = get_domain_from_id(user_id)
elif get_domain_from_id(user_id) == self.server_name:
server_name = get_domain_from_id(group_id)
else:
raise Exception("Expected either group_id or user_id to be local")
if user_id != attestation["user_id"]:
raise SynapseError(400, "Attestation has incorrect user_id")
if group_id != attestation["group_id"]:
raise SynapseError(400, "Attestation has incorrect group_id")
valid_until_ms = attestation["valid_until_ms"]
# TODO: We also want to check that *new* attestations that people give
# us to store are valid for at least a little while.
if valid_until_ms < self.clock.time_msec():
raise SynapseError(400, "Attestation expired")
yield self.keyring.verify_json_for_server(server_name, attestation)
def create_attestation(self, group_id, user_id):
"""Create an attestation for the group_id and user_id with default
validity length.
"""
validity_period = DEFAULT_ATTESTATION_LENGTH_MS
validity_period *= random.uniform(*DEFAULT_ATTESTATION_JITTER)
valid_until_ms = int(self.clock.time_msec() + validity_period)
return sign_json({
"group_id": group_id,
"user_id": user_id,
"valid_until_ms": valid_until_ms,
}, self.server_name, self.signing_key)
class GroupAttestionRenewer(object):
"""Responsible for sending and receiving attestation updates.
"""
def __init__(self, hs):
self.clock = hs.get_clock()
self.store = hs.get_datastore()
self.assestations = hs.get_groups_attestation_signing()
self.transport_client = hs.get_federation_transport_client()
self.is_mine_id = hs.is_mine_id
self.attestations = hs.get_groups_attestation_signing()
self._renew_attestations_loop = self.clock.looping_call(
self._renew_attestations, 30 * 60 * 1000,
)
@defer.inlineCallbacks
def on_renew_attestation(self, group_id, user_id, content):
"""When a remote updates an attestation
"""
attestation = content["attestation"]
if not self.is_mine_id(group_id) and not self.is_mine_id(user_id):
raise SynapseError(400, "Neither user not group are on this server")
yield self.attestations.verify_attestation(
attestation,
user_id=user_id,
group_id=group_id,
)
yield self.store.update_remote_attestion(group_id, user_id, attestation)
defer.returnValue({})
@defer.inlineCallbacks
def _renew_attestations(self):
"""Called periodically to check if we need to update any of our attestations
"""
now = self.clock.time_msec()
rows = yield self.store.get_attestations_need_renewals(
now + UPDATE_ATTESTATION_TIME_MS
)
@defer.inlineCallbacks
def _renew_attestation(group_id, user_id):
if not self.is_mine_id(group_id):
destination = get_domain_from_id(group_id)
elif not self.is_mine_id(user_id):
destination = get_domain_from_id(user_id)
else:
logger.warn(
"Incorrectly trying to do attestations for user: %r in %r",
user_id, group_id,
)
yield self.store.remove_attestation_renewal(group_id, user_id)
return
attestation = self.attestations.create_attestation(group_id, user_id)
yield self.transport_client.renew_group_attestation(
destination, group_id, user_id,
content={"attestation": attestation},
)
yield self.store.update_attestation_renewal(
group_id, user_id, attestation
)
for row in rows:
group_id = row["group_id"]
user_id = row["user_id"]
preserve_fn(_renew_attestation)(group_id, user_id)
+864
View File
@@ -0,0 +1,864 @@
# -*- coding: utf-8 -*-
# Copyright 2017 Vector Creations Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
from synapse.api.errors import SynapseError
from synapse.types import GroupID, RoomID, UserID, get_domain_from_id
from twisted.internet import defer
logger = logging.getLogger(__name__)
# TODO: Allow users to "knock" or simpkly join depending on rules
# TODO: Federation admin APIs
# TODO: is_priveged flag to users and is_public to users and rooms
# TODO: Audit log for admins (profile updates, membership changes, users who tried
# to join but were rejected, etc)
# TODO: Flairs
class GroupsServerHandler(object):
def __init__(self, hs):
self.hs = hs
self.store = hs.get_datastore()
self.room_list_handler = hs.get_room_list_handler()
self.auth = hs.get_auth()
self.clock = hs.get_clock()
self.keyring = hs.get_keyring()
self.is_mine_id = hs.is_mine_id
self.signing_key = hs.config.signing_key[0]
self.server_name = hs.hostname
self.attestations = hs.get_groups_attestation_signing()
self.transport_client = hs.get_federation_transport_client()
self.profile_handler = hs.get_profile_handler()
# Ensure attestations get renewed
hs.get_groups_attestation_renewer()
@defer.inlineCallbacks
def check_group_is_ours(self, group_id, requester_user_id,
and_exists=False, and_is_admin=None):
"""Check that the group is ours, and optionally if it exists.
If group does exist then return group.
Args:
group_id (str)
and_exists (bool): whether to also check if group exists
and_is_admin (str): whether to also check if given str is a user_id
that is an admin
"""
if not self.is_mine_id(group_id):
raise SynapseError(400, "Group not on this server")
group = yield self.store.get_group(group_id)
if and_exists and not group:
raise SynapseError(404, "Unknown group")
is_user_in_group = yield self.store.is_user_in_group(requester_user_id, group_id)
if group and not is_user_in_group and not group["is_public"]:
raise SynapseError(404, "Unknown group")
if and_is_admin:
is_admin = yield self.store.is_user_admin_in_group(group_id, and_is_admin)
if not is_admin:
raise SynapseError(403, "User is not admin in group")
defer.returnValue(group)
@defer.inlineCallbacks
def get_group_summary(self, group_id, requester_user_id):
"""Get the summary for a group as seen by requester_user_id.
The group summary consists of the profile of the room, and a curated
list of users and rooms. These list *may* be organised by role/category.
The roles/categories are ordered, and so are the users/rooms within them.
A user/room may appear in multiple roles/categories.
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
is_user_in_group = yield self.store.is_user_in_group(requester_user_id, group_id)
profile = yield self.get_group_profile(group_id, requester_user_id)
users, roles = yield self.store.get_users_for_summary_by_role(
group_id, include_private=is_user_in_group,
)
# TODO: Add profiles to users
rooms, categories = yield self.store.get_rooms_for_summary_by_category(
group_id, include_private=is_user_in_group,
)
for room_entry in rooms:
room_id = room_entry["room_id"]
joined_users = yield self.store.get_users_in_room(room_id)
entry = yield self.room_list_handler.generate_room_entry(
room_id, len(joined_users),
with_alias=False, allow_private=True,
)
entry = dict(entry) # so we don't change whats cached
entry.pop("room_id", None)
room_entry["profile"] = entry
rooms.sort(key=lambda e: e.get("order", 0))
for entry in users:
user_id = entry["user_id"]
if not self.is_mine_id(requester_user_id):
attestation = yield self.store.get_remote_attestation(group_id, user_id)
if not attestation:
continue
entry["attestation"] = attestation
else:
entry["attestation"] = self.attestations.create_attestation(
group_id, user_id,
)
user_profile = yield self.profile_handler.get_profile_from_cache(user_id)
entry.update(user_profile)
users.sort(key=lambda e: e.get("order", 0))
membership_info = yield self.store.get_users_membership_info_in_group(
group_id, requester_user_id,
)
defer.returnValue({
"profile": profile,
"users_section": {
"users": users,
"roles": roles,
"total_user_count_estimate": 0, # TODO
},
"rooms_section": {
"rooms": rooms,
"categories": categories,
"total_room_count_estimate": 0, # TODO
},
"user": membership_info,
})
@defer.inlineCallbacks
def update_group_summary_room(self, group_id, requester_user_id,
room_id, category_id, content):
"""Add/update a room to the group summary
"""
yield self.check_group_is_ours(
group_id,
requester_user_id,
and_exists=True,
and_is_admin=requester_user_id,
)
RoomID.from_string(room_id) # Ensure valid room id
order = content.get("order", None)
is_public = _parse_visibility_from_contents(content)
yield self.store.add_room_to_summary(
group_id=group_id,
room_id=room_id,
category_id=category_id,
order=order,
is_public=is_public,
)
defer.returnValue({})
@defer.inlineCallbacks
def delete_group_summary_room(self, group_id, requester_user_id,
room_id, category_id):
"""Remove a room from the summary
"""
yield self.check_group_is_ours(
group_id,
requester_user_id,
and_exists=True,
and_is_admin=requester_user_id,
)
yield self.store.remove_room_from_summary(
group_id=group_id,
room_id=room_id,
category_id=category_id,
)
defer.returnValue({})
@defer.inlineCallbacks
def get_group_categories(self, group_id, requester_user_id):
"""Get all categories in a group (as seen by user)
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
categories = yield self.store.get_group_categories(
group_id=group_id,
)
defer.returnValue({"categories": categories})
@defer.inlineCallbacks
def get_group_category(self, group_id, requester_user_id, category_id):
"""Get a specific category in a group (as seen by user)
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
res = yield self.store.get_group_category(
group_id=group_id,
category_id=category_id,
)
defer.returnValue(res)
@defer.inlineCallbacks
def update_group_category(self, group_id, requester_user_id, category_id, content):
"""Add/Update a group category
"""
yield self.check_group_is_ours(
group_id,
requester_user_id,
and_exists=True,
and_is_admin=requester_user_id,
)
is_public = _parse_visibility_from_contents(content)
profile = content.get("profile")
yield self.store.upsert_group_category(
group_id=group_id,
category_id=category_id,
is_public=is_public,
profile=profile,
)
defer.returnValue({})
@defer.inlineCallbacks
def delete_group_category(self, group_id, requester_user_id, category_id):
"""Delete a group category
"""
yield self.check_group_is_ours(
group_id,
requester_user_id,
and_exists=True,
and_is_admin=requester_user_id
)
yield self.store.remove_group_category(
group_id=group_id,
category_id=category_id,
)
defer.returnValue({})
@defer.inlineCallbacks
def get_group_roles(self, group_id, requester_user_id):
"""Get all roles in a group (as seen by user)
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
roles = yield self.store.get_group_roles(
group_id=group_id,
)
defer.returnValue({"roles": roles})
@defer.inlineCallbacks
def get_group_role(self, group_id, requester_user_id, role_id):
"""Get a specific role in a group (as seen by user)
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
res = yield self.store.get_group_role(
group_id=group_id,
role_id=role_id,
)
defer.returnValue(res)
@defer.inlineCallbacks
def update_group_role(self, group_id, requester_user_id, role_id, content):
"""Add/update a role in a group
"""
yield self.check_group_is_ours(
group_id,
requester_user_id,
and_exists=True,
and_is_admin=requester_user_id,
)
is_public = _parse_visibility_from_contents(content)
profile = content.get("profile")
yield self.store.upsert_group_role(
group_id=group_id,
role_id=role_id,
is_public=is_public,
profile=profile,
)
defer.returnValue({})
@defer.inlineCallbacks
def delete_group_role(self, group_id, requester_user_id, role_id):
"""Remove role from group
"""
yield self.check_group_is_ours(
group_id,
requester_user_id,
and_exists=True,
and_is_admin=requester_user_id,
)
yield self.store.remove_group_role(
group_id=group_id,
role_id=role_id,
)
defer.returnValue({})
@defer.inlineCallbacks
def update_group_summary_user(self, group_id, requester_user_id, user_id, role_id,
content):
"""Add/update a users entry in the group summary
"""
yield self.check_group_is_ours(
group_id, requester_user_id, and_exists=True, and_is_admin=requester_user_id,
)
order = content.get("order", None)
is_public = _parse_visibility_from_contents(content)
yield self.store.add_user_to_summary(
group_id=group_id,
user_id=user_id,
role_id=role_id,
order=order,
is_public=is_public,
)
defer.returnValue({})
@defer.inlineCallbacks
def delete_group_summary_user(self, group_id, requester_user_id, user_id, role_id):
"""Remove a user from the group summary
"""
yield self.check_group_is_ours(
group_id, requester_user_id, and_exists=True, and_is_admin=requester_user_id,
)
yield self.store.remove_user_from_summary(
group_id=group_id,
user_id=user_id,
role_id=role_id,
)
defer.returnValue({})
@defer.inlineCallbacks
def get_group_profile(self, group_id, requester_user_id):
"""Get the group profile as seen by requester_user_id
"""
yield self.check_group_is_ours(group_id, requester_user_id)
group_description = yield self.store.get_group(group_id)
if group_description:
defer.returnValue(group_description)
else:
raise SynapseError(404, "Unknown group")
@defer.inlineCallbacks
def update_group_profile(self, group_id, requester_user_id, content):
"""Update the group profile
"""
yield self.check_group_is_ours(
group_id, requester_user_id, and_exists=True, and_is_admin=requester_user_id,
)
profile = {}
for keyname in ("name", "avatar_url", "short_description",
"long_description"):
if keyname in content:
value = content[keyname]
if not isinstance(value, basestring):
raise SynapseError(400, "%r value is not a string" % (keyname,))
profile[keyname] = value
yield self.store.update_group_profile(group_id, profile)
@defer.inlineCallbacks
def get_users_in_group(self, group_id, requester_user_id):
"""Get the users in group as seen by requester_user_id.
The ordering is arbitrary at the moment
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
is_user_in_group = yield self.store.is_user_in_group(requester_user_id, group_id)
user_results = yield self.store.get_users_in_group(
group_id, include_private=is_user_in_group,
)
chunk = []
for user_result in user_results:
g_user_id = user_result["user_id"]
is_public = user_result["is_public"]
is_privileged = user_result["is_admin"]
entry = {"user_id": g_user_id}
profile = yield self.profile_handler.get_profile_from_cache(g_user_id)
entry.update(profile)
entry["is_public"] = bool(is_public)
entry["is_privileged"] = bool(is_privileged)
if not self.is_mine_id(g_user_id):
attestation = yield self.store.get_remote_attestation(group_id, g_user_id)
if not attestation:
continue
entry["attestation"] = attestation
else:
entry["attestation"] = self.attestations.create_attestation(
group_id, g_user_id,
)
chunk.append(entry)
# TODO: If admin add lists of users whose attestations have timed out
defer.returnValue({
"chunk": chunk,
"total_user_count_estimate": len(user_results),
})
@defer.inlineCallbacks
def get_invited_users_in_group(self, group_id, requester_user_id):
"""Get the users that have been invited to a group as seen by requester_user_id.
The ordering is arbitrary at the moment
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
is_user_in_group = yield self.store.is_user_in_group(requester_user_id, group_id)
if not is_user_in_group:
raise SynapseError(403, "User not in group")
invited_users = yield self.store.get_invited_users_in_group(group_id)
user_profiles = []
for user_id in invited_users:
user_profile = {
"user_id": user_id
}
try:
profile = yield self.profile_handler.get_profile_from_cache(user_id)
user_profile.update(profile)
except Exception as e:
logger.warn("Error getting profile for %s: %s", user_id, e)
user_profiles.append(user_profile)
defer.returnValue({
"chunk": user_profiles,
"total_user_count_estimate": len(invited_users),
})
@defer.inlineCallbacks
def get_rooms_in_group(self, group_id, requester_user_id):
"""Get the rooms in group as seen by requester_user_id
This returns rooms in order of decreasing number of joined users
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
is_user_in_group = yield self.store.is_user_in_group(requester_user_id, group_id)
room_results = yield self.store.get_rooms_in_group(
group_id, include_private=is_user_in_group,
)
chunk = []
for room_result in room_results:
room_id = room_result["room_id"]
joined_users = yield self.store.get_users_in_room(room_id)
entry = yield self.room_list_handler.generate_room_entry(
room_id, len(joined_users),
with_alias=False, allow_private=True,
)
if not entry:
continue
entry["is_public"] = bool(room_result["is_public"])
chunk.append(entry)
chunk.sort(key=lambda e: -e["num_joined_members"])
defer.returnValue({
"chunk": chunk,
"total_room_count_estimate": len(room_results),
})
@defer.inlineCallbacks
def add_room_to_group(self, group_id, requester_user_id, room_id, content):
"""Add room to group
"""
RoomID.from_string(room_id) # Ensure valid room id
yield self.check_group_is_ours(
group_id, requester_user_id, and_exists=True, and_is_admin=requester_user_id
)
is_public = _parse_visibility_from_contents(content)
yield self.store.add_room_to_group(group_id, room_id, is_public=is_public)
defer.returnValue({})
@defer.inlineCallbacks
def update_room_in_group(self, group_id, requester_user_id, room_id, config_key,
content):
"""Update room in group
"""
RoomID.from_string(room_id) # Ensure valid room id
yield self.check_group_is_ours(
group_id, requester_user_id, and_exists=True, and_is_admin=requester_user_id
)
if config_key == "m.visibility":
is_public = _parse_visibility_dict(content)
yield self.store.update_room_in_group_visibility(
group_id, room_id,
is_public=is_public,
)
else:
raise SynapseError(400, "Uknown config option")
defer.returnValue({})
@defer.inlineCallbacks
def remove_room_from_group(self, group_id, requester_user_id, room_id):
"""Remove room from group
"""
yield self.check_group_is_ours(
group_id, requester_user_id, and_exists=True, and_is_admin=requester_user_id
)
yield self.store.remove_room_from_group(group_id, room_id)
defer.returnValue({})
@defer.inlineCallbacks
def invite_to_group(self, group_id, user_id, requester_user_id, content):
"""Invite user to group
"""
group = yield self.check_group_is_ours(
group_id, requester_user_id, and_exists=True, and_is_admin=requester_user_id
)
# TODO: Check if user knocked
# TODO: Check if user is already invited
content = {
"profile": {
"name": group["name"],
"avatar_url": group["avatar_url"],
},
"inviter": requester_user_id,
}
if self.hs.is_mine_id(user_id):
groups_local = self.hs.get_groups_local_handler()
res = yield groups_local.on_invite(group_id, user_id, content)
local_attestation = None
else:
local_attestation = self.attestations.create_attestation(group_id, user_id)
content.update({
"attestation": local_attestation,
})
res = yield self.transport_client.invite_to_group_notification(
get_domain_from_id(user_id), group_id, user_id, content
)
user_profile = res.get("user_profile", {})
yield self.store.add_remote_profile_cache(
user_id,
displayname=user_profile.get("displayname"),
avatar_url=user_profile.get("avatar_url"),
)
if res["state"] == "join":
if not self.hs.is_mine_id(user_id):
remote_attestation = res["attestation"]
yield self.attestations.verify_attestation(
remote_attestation,
user_id=user_id,
group_id=group_id,
)
else:
remote_attestation = None
yield self.store.add_user_to_group(
group_id, user_id,
is_admin=False,
is_public=False, # TODO
local_attestation=local_attestation,
remote_attestation=remote_attestation,
)
elif res["state"] == "invite":
yield self.store.add_group_invite(
group_id, user_id,
)
defer.returnValue({
"state": "invite"
})
elif res["state"] == "reject":
defer.returnValue({
"state": "reject"
})
else:
raise SynapseError(502, "Unknown state returned by HS")
@defer.inlineCallbacks
def accept_invite(self, group_id, requester_user_id, content):
"""User tries to accept an invite to the group.
This is different from them asking to join, and so should error if no
invite exists (and they're not a member of the group)
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
is_invited = yield self.store.is_user_invited_to_local_group(
group_id, requester_user_id,
)
if not is_invited:
raise SynapseError(403, "User not invited to group")
if not self.hs.is_mine_id(requester_user_id):
local_attestation = self.attestations.create_attestation(
group_id, requester_user_id,
)
remote_attestation = content["attestation"]
yield self.attestations.verify_attestation(
remote_attestation,
user_id=requester_user_id,
group_id=group_id,
)
else:
local_attestation = None
remote_attestation = None
is_public = _parse_visibility_from_contents(content)
yield self.store.add_user_to_group(
group_id, requester_user_id,
is_admin=False,
is_public=is_public,
local_attestation=local_attestation,
remote_attestation=remote_attestation,
)
defer.returnValue({
"state": "join",
"attestation": local_attestation,
})
@defer.inlineCallbacks
def knock(self, group_id, requester_user_id, content):
"""A user requests becoming a member of the group
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
raise NotImplementedError()
@defer.inlineCallbacks
def accept_knock(self, group_id, requester_user_id, content):
"""Accept a users knock to the room.
Errors if the user hasn't knocked, rather than inviting them.
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
raise NotImplementedError()
@defer.inlineCallbacks
def remove_user_from_group(self, group_id, user_id, requester_user_id, content):
"""Remove a user from the group; either a user is leaving or an admin
kicked them.
"""
yield self.check_group_is_ours(group_id, requester_user_id, and_exists=True)
is_kick = False
if requester_user_id != user_id:
is_admin = yield self.store.is_user_admin_in_group(
group_id, requester_user_id
)
if not is_admin:
raise SynapseError(403, "User is not admin in group")
is_kick = True
yield self.store.remove_user_from_group(
group_id, user_id,
)
if is_kick:
if self.hs.is_mine_id(user_id):
groups_local = self.hs.get_groups_local_handler()
yield groups_local.user_removed_from_group(group_id, user_id, {})
else:
yield self.transport_client.remove_user_from_group_notification(
get_domain_from_id(user_id), group_id, user_id, {}
)
if not self.hs.is_mine_id(user_id):
yield self.store.maybe_delete_remote_profile_cache(user_id)
defer.returnValue({})
@defer.inlineCallbacks
def create_group(self, group_id, requester_user_id, content):
group = yield self.check_group_is_ours(group_id, requester_user_id)
logger.info("Attempting to create group with ID: %r", group_id)
# parsing the id into a GroupID validates it.
group_id_obj = GroupID.from_string(group_id)
if group:
raise SynapseError(400, "Group already exists")
is_admin = yield self.auth.is_server_admin(UserID.from_string(requester_user_id))
if not is_admin:
if not self.hs.config.enable_group_creation:
raise SynapseError(
403, "Only a server admin can create groups on this server",
)
localpart = group_id_obj.localpart
if not localpart.startswith(self.hs.config.group_creation_prefix):
raise SynapseError(
400,
"Can only create groups with prefix %r on this server" % (
self.hs.config.group_creation_prefix,
),
)
profile = content.get("profile", {})
name = profile.get("name")
avatar_url = profile.get("avatar_url")
short_description = profile.get("short_description")
long_description = profile.get("long_description")
user_profile = content.get("user_profile", {})
yield self.store.create_group(
group_id,
requester_user_id,
name=name,
avatar_url=avatar_url,
short_description=short_description,
long_description=long_description,
)
if not self.hs.is_mine_id(requester_user_id):
remote_attestation = content["attestation"]
yield self.attestations.verify_attestation(
remote_attestation,
user_id=requester_user_id,
group_id=group_id,
)
local_attestation = self.attestations.create_attestation(
group_id,
requester_user_id,
)
else:
local_attestation = None
remote_attestation = None
yield self.store.add_user_to_group(
group_id, requester_user_id,
is_admin=True,
is_public=True, # TODO
local_attestation=local_attestation,
remote_attestation=remote_attestation,
)
if not self.hs.is_mine_id(requester_user_id):
yield self.store.add_remote_profile_cache(
requester_user_id,
displayname=user_profile.get("displayname"),
avatar_url=user_profile.get("avatar_url"),
)
defer.returnValue({
"group_id": group_id,
})
def _parse_visibility_from_contents(content):
"""Given a content for a request parse out whether the entity should be
public or not
"""
visibility = content.get("m.visibility")
if visibility:
return _parse_visibility_dict(visibility)
else:
is_public = True
return is_public
def _parse_visibility_dict(visibility):
"""Given a dict for the "m.visibility" config return if the entity should
be public or not
"""
vis_type = visibility.get("type")
if not vis_type:
return True
if vis_type not in ("public", "private"):
raise SynapseError(
400, "Synapse only supports 'public'/'private' visibility"
)
return vis_type == "public"
-2
View File
@@ -20,7 +20,6 @@ from .room import (
from .room_member import RoomMemberHandler
from .message import MessageHandler
from .federation import FederationHandler
from .profile import ProfileHandler
from .directory import DirectoryHandler
from .admin import AdminHandler
from .identity import IdentityHandler
@@ -52,7 +51,6 @@ class Handlers(object):
self.room_creation_handler = RoomCreationHandler(hs)
self.room_member_handler = RoomMemberHandler(hs)
self.federation_handler = FederationHandler(hs)
self.profile_handler = ProfileHandler(hs)
self.directory_handler = DirectoryHandler(hs)
self.admin_handler = AdminHandler(hs)
self.identity_handler = IdentityHandler(hs)
+3 -7
View File
@@ -17,7 +17,7 @@ from twisted.internet import defer
from synapse.api.constants import EventTypes
from synapse.util.metrics import Measure
from synapse.util.logcontext import preserve_fn, preserve_context_over_deferred
from synapse.util.logcontext import make_deferred_yieldable, preserve_fn
import logging
@@ -70,11 +70,10 @@ class ApplicationServicesHandler(object):
with Measure(self.clock, "notify_interested_services"):
self.is_processing = True
try:
upper_bound = self.current_max
limit = 100
while True:
upper_bound, events = yield self.store.get_new_events_for_appservice(
upper_bound, limit
self.current_max, limit
)
if not events:
@@ -105,9 +104,6 @@ class ApplicationServicesHandler(object):
)
yield self.store.set_appservice_last_pos(upper_bound)
if len(events) < limit:
break
finally:
self.is_processing = False
@@ -163,7 +159,7 @@ class ApplicationServicesHandler(object):
def query_3pe(self, kind, protocol, fields):
services = yield self._get_services_for_3pn(protocol)
results = yield preserve_context_over_deferred(defer.DeferredList([
results = yield make_deferred_yieldable(defer.DeferredList([
preserve_fn(self.appservice_api.query_3pe)(service, kind, protocol, fields)
for service in services
], consumeErrors=True))
+306 -126
View File
@@ -13,14 +13,18 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from twisted.internet import defer
from ._base import BaseHandler
from synapse.api.constants import LoginType
from synapse.api.errors import (
AuthError, Codes, InteractiveAuthIncompleteError, LoginError, StoreError,
SynapseError,
)
from synapse.module_api import ModuleApi
from synapse.types import UserID
from synapse.api.errors import AuthError, LoginError, Codes, StoreError, SynapseError
from synapse.util.async import run_on_reactor
from synapse.util.caches.expiringcache import ExpiringCache
from twisted.web.client import PartialDownloadError
@@ -45,19 +49,23 @@ class AuthHandler(BaseHandler):
"""
super(AuthHandler, self).__init__(hs)
self.checkers = {
LoginType.PASSWORD: self._check_password_auth,
LoginType.RECAPTCHA: self._check_recaptcha,
LoginType.EMAIL_IDENTITY: self._check_email_identity,
LoginType.MSISDN: self._check_msisdn,
LoginType.DUMMY: self._check_dummy_auth,
}
self.bcrypt_rounds = hs.config.bcrypt_rounds
self.sessions = {}
account_handler = _AccountHandler(
hs, check_user_exists=self.check_user_exists
# This is not a cache per se, but a store of all current sessions that
# expire after N hours
self.sessions = ExpiringCache(
cache_name="register_sessions",
clock=hs.get_clock(),
expiry_ms=self.SESSION_EXPIRE_MS,
reset_expiry_on_get=True,
)
account_handler = ModuleApi(hs, self)
self.password_providers = [
module(config=config, account_handler=account_handler)
for module, config in hs.config.password_providers
@@ -66,39 +74,120 @@ class AuthHandler(BaseHandler):
logger.info("Extra password_providers: %r", self.password_providers)
self.hs = hs # FIXME better possibility to access registrationHandler later?
self.device_handler = hs.get_device_handler()
self.macaroon_gen = hs.get_macaroon_generator()
self._password_enabled = hs.config.password_enabled
# we keep this as a list despite the O(N^2) implication so that we can
# keep PASSWORD first and avoid confusing clients which pick the first
# type in the list. (NB that the spec doesn't require us to do so and
# clients which favour types that they don't understand over those that
# they do are technically broken)
login_types = []
if self._password_enabled:
login_types.append(LoginType.PASSWORD)
for provider in self.password_providers:
if hasattr(provider, "get_supported_login_types"):
for t in provider.get_supported_login_types().keys():
if t not in login_types:
login_types.append(t)
self._supported_login_types = login_types
@defer.inlineCallbacks
def validate_user_via_ui_auth(self, requester, request_body, clientip):
"""
Checks that the user is who they claim to be, via a UI auth.
This is used for things like device deletion and password reset where
the user already has a valid access token, but we want to double-check
that it isn't stolen by re-authenticating them.
Args:
requester (Requester): The user, as given by the access token
request_body (dict): The body of the request sent by the client
clientip (str): The IP address of the client.
Returns:
defer.Deferred[dict]: the parameters for this request (which may
have been given only in a previous call).
Raises:
InteractiveAuthIncompleteError if the client has not yet completed
any of the permitted login flows
AuthError if the client has completed a login flow, and it gives
a different user to `requester`
"""
# build a list of supported flows
flows = [
[login_type] for login_type in self._supported_login_types
]
result, params, _ = yield self.check_auth(
flows, request_body, clientip,
)
# find the completed login type
for login_type in self._supported_login_types:
if login_type not in result:
continue
user_id = result[login_type]
break
else:
# this can't happen
raise Exception(
"check_auth returned True but no successful login type",
)
# check that the UI auth matched the access token
if user_id != requester.user.to_string():
raise AuthError(403, "Invalid auth")
defer.returnValue(params)
@defer.inlineCallbacks
def check_auth(self, flows, clientdict, clientip):
"""
Takes a dictionary sent by the client in the login / registration
protocol and handles the login flow.
protocol and handles the User-Interactive Auth flow.
As a side effect, this function fills in the 'creds' key on the user's
session with a map, which maps each auth-type (str) to the relevant
identity authenticated by that auth-type (mostly str, but for captcha, bool).
If no auth flows have been completed successfully, raises an
InteractiveAuthIncompleteError. To handle this, you can use
synapse.rest.client.v2_alpha._base.interactive_auth_handler as a
decorator.
Args:
flows (list): A list of login flows. Each flow is an ordered list of
strings representing auth-types. At least one full
flow must be completed in order for auth to be successful.
clientdict: The dictionary from the client root level, not the
'auth' key: this method prompts for auth if none is sent.
clientip (str): The IP address of the client.
Returns:
A tuple of (authed, dict, dict, session_id) where authed is true if
the client has successfully completed an auth flow. If it is true
the first dict contains the authenticated credentials of each stage.
defer.Deferred[dict, dict, str]: a deferred tuple of
(creds, params, session_id).
If authed is false, the first dictionary is the server response to
the login request and should be passed back to the client.
'creds' contains the authenticated credentials of each stage.
In either case, the second dict contains the parameters for this
request (which may have been given only in a previous call).
'params' contains the parameters for this request (which may
have been given only in a previous call).
session_id is the ID of this session, either passed in by the client
or assigned by the call to check_auth
'session_id' is the ID of this session, either passed in by the
client or assigned by this call
Raises:
InteractiveAuthIncompleteError if the client has not yet completed
all the stages in any of the permitted flows.
"""
authdict = None
@@ -126,11 +215,8 @@ class AuthHandler(BaseHandler):
clientdict = session['clientdict']
if not authdict:
defer.returnValue(
(
False, self._auth_dict_for_flows(flows, session),
clientdict, session['id']
)
raise InteractiveAuthIncompleteError(
self._auth_dict_for_flows(flows, session),
)
if 'creds' not in session:
@@ -141,14 +227,12 @@ class AuthHandler(BaseHandler):
errordict = {}
if 'type' in authdict:
login_type = authdict['type']
if login_type not in self.checkers:
raise LoginError(400, "", Codes.UNRECOGNIZED)
try:
result = yield self.checkers[login_type](authdict, clientip)
result = yield self._check_auth_dict(authdict, clientip)
if result:
creds[login_type] = result
self._save_session(session)
except LoginError, e:
except LoginError as e:
if login_type == LoginType.EMAIL_IDENTITY:
# riot used to have a bug where it would request a new
# validation token (thus sending a new email) each time it
@@ -157,7 +241,7 @@ class AuthHandler(BaseHandler):
#
# Grandfather in the old behaviour for now to avoid
# breaking old riot deployments.
raise e
raise
# this step failed. Merge the error dict into the response
# so that the client can have another go.
@@ -174,12 +258,14 @@ class AuthHandler(BaseHandler):
"Auth completed with creds: %r. Client dict has keys: %r",
creds, clientdict.keys()
)
defer.returnValue((True, creds, clientdict, session['id']))
defer.returnValue((creds, clientdict, session['id']))
ret = self._auth_dict_for_flows(flows, session)
ret['completed'] = creds.keys()
ret.update(errordict)
defer.returnValue((False, ret, clientdict, session['id']))
raise InteractiveAuthIncompleteError(
ret,
)
@defer.inlineCallbacks
def add_oob_auth(self, stagetype, authdict, clientip):
@@ -251,16 +337,37 @@ class AuthHandler(BaseHandler):
sess = self._get_session_info(session_id)
return sess.setdefault('serverdict', {}).get(key, default)
def _check_password_auth(self, authdict, _):
if "user" not in authdict or "password" not in authdict:
raise LoginError(400, "", Codes.MISSING_PARAM)
@defer.inlineCallbacks
def _check_auth_dict(self, authdict, clientip):
"""Attempt to validate the auth dict provided by a client
user_id = authdict["user"]
password = authdict["password"]
if not user_id.startswith('@'):
user_id = UserID.create(user_id, self.hs.hostname).to_string()
Args:
authdict (object): auth dict provided by the client
clientip (str): IP address of the client
return self._check_password(user_id, password)
Returns:
Deferred: result of the stage verification.
Raises:
StoreError if there was a problem accessing the database
SynapseError if there was a problem with the request
LoginError if there was an authentication problem.
"""
login_type = authdict['type']
checker = self.checkers.get(login_type)
if checker is not None:
res = yield checker(authdict, clientip)
defer.returnValue(res)
# build a v1-login-style dict out of the authdict and fall back to the
# v1 code
user_id = authdict.get("user")
if user_id is None:
raise SynapseError(400, "", Codes.MISSING_PARAM)
(canonical_id, callback) = yield self.validate_login(user_id, authdict)
defer.returnValue(canonical_id)
@defer.inlineCallbacks
def _check_recaptcha(self, authdict, clientip):
@@ -389,26 +496,8 @@ class AuthHandler(BaseHandler):
return self.sessions[session_id]
def validate_password_login(self, user_id, password):
"""
Authenticates the user with their username and password.
Used only by the v1 login API.
Args:
user_id (str): complete @user:id
password (str): Password
Returns:
defer.Deferred: (str) canonical user id
Raises:
StoreError if there was a problem accessing the database
LoginError if there was an authentication problem.
"""
return self._check_password(user_id, password)
@defer.inlineCallbacks
def get_access_token_for_user_id(self, user_id, device_id=None,
initial_display_name=None):
def get_access_token_for_user_id(self, user_id, device_id=None):
"""
Creates a new access token for the user with the given user ID.
@@ -422,13 +511,10 @@ class AuthHandler(BaseHandler):
device_id (str|None): the device ID to associate with the tokens.
None to leave the tokens unassociated with a device (deprecated:
we should always have a device ID)
initial_display_name (str): display name to associate with the
device if it needs re-registering
Returns:
The access token for the user's session.
Raises:
StoreError if there was a problem storing the token.
LoginError if there was an authentication problem.
"""
logger.info("Logging in user %s on device %s", user_id, device_id)
access_token = yield self.issue_access_token(user_id, device_id)
@@ -438,9 +524,11 @@ class AuthHandler(BaseHandler):
# really don't want is active access_tokens without a record of the
# device, so we double-check it here.
if device_id is not None:
yield self.device_handler.check_device_registered(
user_id, device_id, initial_display_name
)
try:
yield self.store.get_device(user_id, device_id)
except StoreError:
yield self.store.delete_access_token(access_token)
raise StoreError(400, "Login raced against device deletion")
defer.returnValue(access_token)
@@ -492,29 +580,115 @@ class AuthHandler(BaseHandler):
)
defer.returnValue(result)
@defer.inlineCallbacks
def _check_password(self, user_id, password):
"""Authenticate a user against the LDAP and local databases.
def get_supported_login_types(self):
"""Get a the login types supported for the /login API
user_id is checked case insensitively against the local database, but
will throw if there are multiple inexact matches.
By default this is just 'm.login.password' (unless password_enabled is
False in the config file), but password auth providers can provide
other login types.
Returns:
Iterable[str]: login types
"""
return self._supported_login_types
@defer.inlineCallbacks
def validate_login(self, username, login_submission):
"""Authenticates the user for the /login API
Also used by the user-interactive auth flow to validate
m.login.password auth types.
Args:
user_id (str): complete @user:id
username (str): username supplied by the user
login_submission (dict): the whole of the login submission
(including 'type' and other relevant fields)
Returns:
(str) the canonical_user_id
Deferred[str, func]: canonical user id, and optional callback
to be called once the access token and device id are issued
Raises:
LoginError if login fails
StoreError if there was a problem accessing the database
SynapseError if there was a problem with the request
LoginError if there was an authentication problem.
"""
if username.startswith('@'):
qualified_user_id = username
else:
qualified_user_id = UserID(
username, self.hs.hostname
).to_string()
login_type = login_submission.get("type")
known_login_type = False
# special case to check for "password" for the check_password interface
# for the auth providers
password = login_submission.get("password")
if login_type == LoginType.PASSWORD:
if not self._password_enabled:
raise SynapseError(400, "Password login has been disabled.")
if not password:
raise SynapseError(400, "Missing parameter: password")
for provider in self.password_providers:
is_valid = yield provider.check_password(user_id, password)
if is_valid:
defer.returnValue(user_id)
if (hasattr(provider, "check_password")
and login_type == LoginType.PASSWORD):
known_login_type = True
is_valid = yield provider.check_password(
qualified_user_id, password,
)
if is_valid:
defer.returnValue((qualified_user_id, None))
canonical_user_id = yield self._check_local_password(user_id, password)
if (not hasattr(provider, "get_supported_login_types")
or not hasattr(provider, "check_auth")):
# this password provider doesn't understand custom login types
continue
if canonical_user_id:
defer.returnValue(canonical_user_id)
supported_login_types = provider.get_supported_login_types()
if login_type not in supported_login_types:
# this password provider doesn't understand this login type
continue
known_login_type = True
login_fields = supported_login_types[login_type]
missing_fields = []
login_dict = {}
for f in login_fields:
if f not in login_submission:
missing_fields.append(f)
else:
login_dict[f] = login_submission[f]
if missing_fields:
raise SynapseError(
400, "Missing parameters for login type %s: %s" % (
login_type,
missing_fields,
),
)
result = yield provider.check_auth(
username, login_type, login_dict,
)
if result:
if isinstance(result, str):
result = (result, None)
defer.returnValue(result)
if login_type == LoginType.PASSWORD:
known_login_type = True
canonical_user_id = yield self._check_local_password(
qualified_user_id, password,
)
if canonical_user_id:
defer.returnValue((canonical_user_id, None))
if not known_login_type:
raise SynapseError(400, "Unknown login type %s" % login_type)
# unknown username or invalid password. We raise a 403 here, but note
# that if we're doing user-interactive login, it turns all LoginErrors
@@ -564,22 +738,65 @@ class AuthHandler(BaseHandler):
raise AuthError(403, "Invalid token", errcode=Codes.FORBIDDEN)
@defer.inlineCallbacks
def set_password(self, user_id, newpassword, requester=None):
password_hash = self.hash(newpassword)
def delete_access_token(self, access_token):
"""Invalidate a single access token
except_access_token_id = requester.access_token_id if requester else None
Args:
access_token (str): access token to be deleted
try:
yield self.store.user_set_password_hash(user_id, password_hash)
except StoreError as e:
if e.code == 404:
raise SynapseError(404, "Unknown user", Codes.NOT_FOUND)
raise e
yield self.store.user_delete_access_tokens(
user_id, except_access_token_id
Returns:
Deferred
"""
user_info = yield self.auth.get_user_by_access_token(access_token)
yield self.store.delete_access_token(access_token)
# see if any of our auth providers want to know about this
for provider in self.password_providers:
if hasattr(provider, "on_logged_out"):
yield provider.on_logged_out(
user_id=str(user_info["user"]),
device_id=user_info["device_id"],
access_token=access_token,
)
# delete pushers associated with this access token
if user_info["token_id"] is not None:
yield self.hs.get_pusherpool().remove_pushers_by_access_token(
str(user_info["user"]), (user_info["token_id"], )
)
@defer.inlineCallbacks
def delete_access_tokens_for_user(self, user_id, except_token_id=None,
device_id=None):
"""Invalidate access tokens belonging to a user
Args:
user_id (str): ID of user the tokens belong to
except_token_id (str|None): access_token ID which should *not* be
deleted
device_id (str|None): ID of device the tokens are associated with.
If None, tokens associated with any device (or no device) will
be deleted
Returns:
Deferred
"""
tokens_and_devices = yield self.store.user_delete_access_tokens(
user_id, except_token_id=except_token_id, device_id=device_id,
)
yield self.hs.get_pusherpool().remove_pushers_by_user(
user_id, except_access_token_id
# see if any of our auth providers want to know about this
for provider in self.password_providers:
if hasattr(provider, "on_logged_out"):
for token, token_id, device_id in tokens_and_devices:
yield provider.on_logged_out(
user_id=user_id,
device_id=device_id,
access_token=token,
)
# delete pushers associated with the access tokens
yield self.hs.get_pusherpool().remove_pushers_by_access_token(
user_id, (token_id for _, token_id, _ in tokens_and_devices),
)
@defer.inlineCallbacks
@@ -617,16 +834,6 @@ class AuthHandler(BaseHandler):
logger.debug("Saving session %s", session)
session["last_used"] = self.hs.get_clock().time_msec()
self.sessions[session["id"]] = session
self._prune_sessions()
def _prune_sessions(self):
for sid, sess in self.sessions.items():
last_used = 0
if 'last_used' in sess:
last_used = sess['last_used']
now = self.hs.get_clock().time_msec()
if last_used < now - AuthHandler.SESSION_EXPIRE_MS:
del self.sessions[sid]
def hash(self, password):
"""Computes a secure hash of password.
@@ -697,30 +904,3 @@ class MacaroonGeneartor(object):
macaroon.add_first_party_caveat("gen = 1")
macaroon.add_first_party_caveat("user_id = %s" % (user_id,))
return macaroon
class _AccountHandler(object):
"""A proxy object that gets passed to password auth providers so they
can register new users etc if necessary.
"""
def __init__(self, hs, check_user_exists):
self.hs = hs
self._check_user_exists = check_user_exists
def check_user_exists(self, user_id):
"""Check if user exissts.
Returns:
Deferred(bool)
"""
return self._check_user_exists(user_id)
def register(self, localpart):
"""Registers a new user with given localpart
Returns:
Deferred: a 2-tuple of (user_id, access_token)
"""
reg = self.hs.get_handlers().registration_handler
return reg.register(localpart=localpart)
+52
View File
@@ -0,0 +1,52 @@
# -*- coding: utf-8 -*-
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from twisted.internet import defer
from ._base import BaseHandler
import logging
logger = logging.getLogger(__name__)
class DeactivateAccountHandler(BaseHandler):
"""Handler which deals with deactivating user accounts."""
def __init__(self, hs):
super(DeactivateAccountHandler, self).__init__(hs)
self._auth_handler = hs.get_auth_handler()
self._device_handler = hs.get_device_handler()
@defer.inlineCallbacks
def deactivate_account(self, user_id):
"""Deactivate a user's account
Args:
user_id (str): ID of user to be deactivated
Returns:
Deferred
"""
# FIXME: Theoretically there is a race here wherein user resets
# password using threepid.
# first delete any devices belonging to the user, which will also
# delete corresponding access tokens.
yield self._device_handler.delete_all_devices_for_user(user_id)
# then delete any remaining access tokens which weren't associated with
# a device.
yield self._auth_handler.delete_access_tokens_for_user(user_id)
yield self.store.user_delete_threepids(user_id)
yield self.store.user_set_password_hash(user_id, None)
+81 -18
View File
@@ -34,6 +34,7 @@ class DeviceHandler(BaseHandler):
self.hs = hs
self.state = hs.get_state_handler()
self._auth_handler = hs.get_auth_handler()
self.federation_sender = hs.get_federation_sender()
self.federation = hs.get_replication_layer()
@@ -106,7 +107,7 @@ class DeviceHandler(BaseHandler):
device_map = yield self.store.get_devices_by_user(user_id)
ips = yield self.store.get_last_client_ip_by_device(
devices=((user_id, device_id) for device_id in device_map.keys())
user_id, device_id=None
)
devices = device_map.values()
@@ -133,7 +134,7 @@ class DeviceHandler(BaseHandler):
except errors.StoreError:
raise errors.NotFoundError
ips = yield self.store.get_last_client_ip_by_device(
devices=((user_id, device_id),)
user_id, device_id,
)
_update_device_from_client_ips(device, ips)
defer.returnValue(device)
@@ -159,9 +160,8 @@ class DeviceHandler(BaseHandler):
else:
raise
yield self.store.user_delete_access_tokens(
yield self._auth_handler.delete_access_tokens_for_user(
user_id, device_id=device_id,
delete_refresh_tokens=True,
)
yield self.store.delete_e2e_keys_by_device(
@@ -170,13 +170,31 @@ class DeviceHandler(BaseHandler):
yield self.notify_device_update(user_id, [device_id])
@defer.inlineCallbacks
def delete_all_devices_for_user(self, user_id, except_device_id=None):
"""Delete all of the user's devices
Args:
user_id (str):
except_device_id (str|None): optional device id which should not
be deleted
Returns:
defer.Deferred:
"""
device_map = yield self.store.get_devices_by_user(user_id)
device_ids = device_map.keys()
if except_device_id is not None:
device_ids = [d for d in device_ids if d != except_device_id]
yield self.delete_devices(user_id, device_ids)
@defer.inlineCallbacks
def delete_devices(self, user_id, device_ids):
""" Delete several devices
Args:
user_id (str):
device_ids (str): The list of device IDs to delete
device_ids (List[str]): The list of device IDs to delete
Returns:
defer.Deferred:
@@ -194,9 +212,8 @@ class DeviceHandler(BaseHandler):
# Delete access tokens and e2e keys for each device. Not optimised as it is not
# considered as part of a critical path.
for device_id in device_ids:
yield self.store.user_delete_access_tokens(
yield self._auth_handler.delete_access_tokens_for_user(
user_id, device_id=device_id,
delete_refresh_tokens=True,
)
yield self.store.delete_e2e_keys_by_device(
user_id=user_id, device_id=device_id
@@ -270,6 +287,8 @@ class DeviceHandler(BaseHandler):
user_id (str)
from_token (StreamToken)
"""
now_token = yield self.hs.get_event_sources().get_current_token()
room_ids = yield self.store.get_rooms_for_user(user_id)
# First we check if any devices have changed
@@ -280,11 +299,30 @@ class DeviceHandler(BaseHandler):
# Then work out if any users have since joined
rooms_changed = self.store.get_rooms_that_changed(room_ids, from_token.room_key)
member_events = yield self.store.get_membership_changes_for_user(
user_id, from_token.room_key, now_token.room_key
)
rooms_changed.update(event.room_id for event in member_events)
stream_ordering = RoomStreamToken.parse_stream_token(
from_token.room_key).stream
from_token.room_key
).stream
possibly_changed = set(changed)
possibly_left = set()
for room_id in rooms_changed:
current_state_ids = yield self.store.get_current_state_ids(room_id)
# The user may have left the room
# TODO: Check if they actually did or if we were just invited.
if room_id not in room_ids:
for key, event_id in current_state_ids.iteritems():
etype, state_key = key
if etype != EventTypes.Member:
continue
possibly_left.add(state_key)
continue
# Fetch the current state at the time.
try:
event_ids = yield self.store.get_forward_extremeties_for_room(
@@ -295,8 +333,6 @@ class DeviceHandler(BaseHandler):
# ordering: treat it the same as a new room
event_ids = []
current_state_ids = yield self.store.get_current_state_ids(room_id)
# special-case for an empty prev state: include all members
# in the changed list
if not event_ids:
@@ -307,9 +343,25 @@ class DeviceHandler(BaseHandler):
possibly_changed.add(state_key)
continue
current_member_id = current_state_ids.get((EventTypes.Member, user_id))
if not current_member_id:
continue
# mapping from event_id -> state_dict
prev_state_ids = yield self.store.get_state_ids_for_events(event_ids)
# Check if we've joined the room? If so we just blindly add all the users to
# the "possibly changed" users.
for state_dict in prev_state_ids.itervalues():
member_event = state_dict.get((EventTypes.Member, user_id), None)
if not member_event or member_event != current_member_id:
for key, event_id in current_state_ids.iteritems():
etype, state_key = key
if etype != EventTypes.Member:
continue
possibly_changed.add(state_key)
break
# If there has been any change in membership, include them in the
# possibly changed list. We'll check if they are joined below,
# and we're not toooo worried about spuriously adding users.
@@ -320,19 +372,30 @@ class DeviceHandler(BaseHandler):
# check if this member has changed since any of the extremities
# at the stream_ordering, and add them to the list if so.
for state_dict in prev_state_ids.values():
for state_dict in prev_state_ids.itervalues():
prev_event_id = state_dict.get(key, None)
if not prev_event_id or prev_event_id != event_id:
possibly_changed.add(state_key)
if state_key != user_id:
possibly_changed.add(state_key)
break
users_who_share_room = yield self.store.get_users_who_share_room_with_user(
user_id
)
if possibly_changed or possibly_left:
users_who_share_room = yield self.store.get_users_who_share_room_with_user(
user_id
)
# Take the intersection of the users whose devices may have changed
# and those that actually still share a room with the user
defer.returnValue(users_who_share_room & possibly_changed)
# Take the intersection of the users whose devices may have changed
# and those that actually still share a room with the user
possibly_joined = possibly_changed & users_who_share_room
possibly_left = (possibly_changed | possibly_left) - users_who_share_room
else:
possibly_joined = []
possibly_left = []
defer.returnValue({
"changed": list(possibly_joined),
"left": list(possibly_left),
})
@defer.inlineCallbacks
def on_federation_query_user_devices(self, user_id):
+15
View File
@@ -40,6 +40,8 @@ class DirectoryHandler(BaseHandler):
"directory", self.on_directory_query
)
self.spam_checker = hs.get_spam_checker()
@defer.inlineCallbacks
def _create_association(self, room_alias, room_id, servers=None, creator=None):
# general association creation for both human users and app services
@@ -73,6 +75,11 @@ class DirectoryHandler(BaseHandler):
# association creation for human users
# TODO(erikj): Do user auth.
if not self.spam_checker.user_may_create_room_alias(user_id, room_alias):
raise SynapseError(
403, "This user is not permitted to create this alias",
)
can_create = yield self.can_modify_alias(
room_alias,
user_id=user_id
@@ -327,6 +334,14 @@ class DirectoryHandler(BaseHandler):
room_id (str)
visibility (str): "public" or "private"
"""
if not self.spam_checker.user_may_publish_room(
requester.user.to_string(), room_id
):
raise AuthError(
403,
"This user is not permitted to publish rooms to the room list"
)
if requester.is_guest:
raise AuthError(403, "Guests cannot edit the published room list")
+160 -62
View File
@@ -14,7 +14,6 @@
# limitations under the License.
"""Contains handlers for federation events."""
import synapse.util.logcontext
from signedjson.key import decode_verify_key_bytes
from signedjson.sign import verify_signed_json
from unpaddedbase64 import decode_base64
@@ -26,10 +25,7 @@ from synapse.api.errors import (
)
from synapse.api.constants import EventTypes, Membership, RejectedReason
from synapse.events.validator import EventValidator
from synapse.util import unwrapFirstError
from synapse.util.logcontext import (
preserve_fn, preserve_context_over_deferred
)
from synapse.util import unwrapFirstError, logcontext
from synapse.util.metrics import measure_func
from synapse.util.logutils import log_function
from synapse.util.async import run_on_reactor, Linearizer
@@ -75,6 +71,9 @@ class FederationHandler(BaseHandler):
self.server_name = hs.hostname
self.keyring = hs.get_keyring()
self.action_generator = hs.get_action_generator()
self.is_mine_id = hs.is_mine_id
self.pusher_pool = hs.get_pusherpool()
self.spam_checker = hs.get_spam_checker()
self.replication_layer.set_handler(self)
@@ -123,6 +122,28 @@ class FederationHandler(BaseHandler):
self.room_queues[pdu.room_id].append((pdu, origin))
return
# If we're no longer in the room just ditch the event entirely. This
# is probably an old server that has come back and thinks we're still
# in the room (or we've been rejoined to the room by a state reset).
#
# If we were never in the room then maybe our database got vaped and
# we should check if we *are* in fact in the room. If we are then we
# can magically rejoin the room.
is_in_room = yield self.auth.check_host_in_room(
pdu.room_id,
self.server_name
)
if not is_in_room:
was_in_room = yield self.store.was_host_joined(
pdu.room_id, self.server_name,
)
if was_in_room:
logger.info(
"Ignoring PDU %s for room %s from %s as we've left the room!",
pdu.event_id, pdu.room_id, origin,
)
return
state = None
auth_chain = []
@@ -206,7 +227,7 @@ class FederationHandler(BaseHandler):
state, auth_chain = yield self.replication_layer.get_state_for_room(
origin, pdu.room_id, pdu.event_id,
)
except:
except Exception:
logger.exception("Failed to get state for event: %s", pdu.event_id)
yield self._process_received_pdu(
@@ -440,7 +461,7 @@ class FederationHandler(BaseHandler):
def check_match(id):
try:
return server_name == get_domain_from_id(id)
except:
except Exception:
return False
# Parses mapping `event_id -> (type, state_key) -> state event_id`
@@ -478,7 +499,7 @@ class FederationHandler(BaseHandler):
continue
try:
domain = get_domain_from_id(ev.state_key)
except:
except Exception:
continue
if domain != server_name:
@@ -589,9 +610,9 @@ class FederationHandler(BaseHandler):
missing_auth - failed_to_fetch
)
results = yield preserve_context_over_deferred(defer.gatherResults(
results = yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(self.replication_layer.get_pdu)(
logcontext.preserve_fn(self.replication_layer.get_pdu)(
[dest],
event_id,
outlier=True,
@@ -717,7 +738,7 @@ class FederationHandler(BaseHandler):
joined_domains[dom] = min(d, old_d)
else:
joined_domains[dom] = d
except:
except Exception:
pass
return sorted(joined_domains.items(), key=lambda d: d[1])
@@ -783,10 +804,14 @@ class FederationHandler(BaseHandler):
event_ids = list(extremities.keys())
logger.debug("calling resolve_state_groups in _maybe_backfill")
states = yield preserve_context_over_deferred(defer.gatherResults([
preserve_fn(self.state_handler.resolve_state_groups)(room_id, [e])
for e in event_ids
]))
states = yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
logcontext.preserve_fn(self.state_handler.resolve_state_groups)(
room_id, [e]
)
for e in event_ids
], consumeErrors=True,
))
states = dict(zip(event_ids, [s.state for s in states]))
state_map = yield self.store.get_events(
@@ -832,7 +857,11 @@ class FederationHandler(BaseHandler):
@defer.inlineCallbacks
def on_event_auth(self, event_id):
auth = yield self.store.get_auth_chain([event_id])
event = yield self.store.get_event(event_id)
auth = yield self.store.get_auth_chain(
[auth_id for auth_id, _ in event.auth_events],
include_given=True
)
for event in auth:
event.signatures.update(
@@ -911,7 +940,7 @@ class FederationHandler(BaseHandler):
room_creator_user_id="",
is_public=False
)
except:
except Exception:
# FIXME
pass
@@ -935,9 +964,7 @@ class FederationHandler(BaseHandler):
# lots of requests for missing prev_events which we do actually
# have. Hence we fire off the deferred, but don't wait for it.
synapse.util.logcontext.preserve_fn(self._handle_queued_pdus)(
room_queue
)
logcontext.preserve_fn(self._handle_queued_pdus)(room_queue)
defer.returnValue(True)
@@ -1047,9 +1074,7 @@ class FederationHandler(BaseHandler):
yield user_joined_room(self.distributor, user, event.room_id)
state_ids = context.prev_state_ids.values()
auth_chain = yield self.store.get_auth_chain(set(
[event.event_id] + state_ids
))
auth_chain = yield self.store.get_auth_chain(state_ids)
state = yield self.store.get_events(context.prev_state_ids.values())
@@ -1066,6 +1091,34 @@ class FederationHandler(BaseHandler):
"""
event = pdu
if event.state_key is None:
raise SynapseError(400, "The invite event did not have a state key")
is_blocked = yield self.store.is_room_blocked(event.room_id)
if is_blocked:
raise SynapseError(403, "This room has been blocked on this server")
if self.hs.config.block_non_admin_invites:
raise SynapseError(403, "This server does not accept room invites")
if not self.spam_checker.user_may_invite(
event.sender, event.state_key, event.room_id,
):
raise SynapseError(
403, "This user is not permitted to send invites to this server/user"
)
membership = event.content.get("membership")
if event.type != EventTypes.Member or membership != Membership.INVITE:
raise SynapseError(400, "The event was not an m.room.member invite event")
sender_domain = get_domain_from_id(event.sender)
if sender_domain != origin:
raise SynapseError(400, "The invite event was not from the server sending it")
if not self.is_mine_id(event.state_key):
raise SynapseError(400, "The invite event must be for this server")
event.internal_metadata.outlier = True
event.internal_metadata.invite_from_remote = True
@@ -1100,6 +1153,9 @@ class FederationHandler(BaseHandler):
user_id,
"leave"
)
# Mark as outlier as we don't have any state for this event; we're not
# even in the room.
event.internal_metadata.outlier = True
event = self._sign_event(event)
# Try the host that we succesfully called /make_leave/ on first for
@@ -1271,7 +1327,7 @@ class FederationHandler(BaseHandler):
for event in res:
# We sign these again because there was a bug where we
# incorrectly signed things the first time round
if self.hs.is_mine_id(event.event_id):
if self.is_mine_id(event.event_id):
event.signatures.update(
compute_event_signature(
event,
@@ -1344,7 +1400,7 @@ class FederationHandler(BaseHandler):
)
if event:
if self.hs.is_mine_id(event.event_id):
if self.is_mine_id(event.event_id):
# FIXME: This is a temporary work around where we occasionally
# return events slightly differently than when they were
# originally signed
@@ -1388,7 +1444,7 @@ class FederationHandler(BaseHandler):
auth_events=auth_events,
)
if not event.internal_metadata.is_outlier():
if not event.internal_metadata.is_outlier() and not backfilled:
yield self.action_generator.handle_push_actions_for_event(
event, context
)
@@ -1402,7 +1458,7 @@ class FederationHandler(BaseHandler):
if not backfilled:
# this intentionally does not yield: we don't care about the result
# and don't need to wait for it.
preserve_fn(self.hs.get_pusherpool().on_new_notifications)(
logcontext.preserve_fn(self.pusher_pool.on_new_notifications)(
event_stream_id, max_stream_id
)
@@ -1415,16 +1471,16 @@ class FederationHandler(BaseHandler):
a bunch of outliers, but not a chunk of individual events that depend
on each other for state calculations.
"""
contexts = yield preserve_context_over_deferred(defer.gatherResults(
contexts = yield logcontext.make_deferred_yieldable(defer.gatherResults(
[
preserve_fn(self._prep_event)(
logcontext.preserve_fn(self._prep_event)(
origin,
ev_info["event"],
state=ev_info.get("state"),
auth_events=ev_info.get("auth_events"),
)
for ev_info in event_infos
]
], consumeErrors=True,
))
yield self.store.persist_events(
@@ -1581,7 +1637,7 @@ class FederationHandler(BaseHandler):
context.rejected = RejectedReason.AUTH_ERROR
if event.type == EventTypes.GuestAccess:
if event.type == EventTypes.GuestAccess and not context.rejected:
yield self.maybe_kick_guest_users(event)
defer.returnValue(context)
@@ -1598,7 +1654,11 @@ class FederationHandler(BaseHandler):
pass
# Now get the current auth_chain for the event.
local_auth_chain = yield self.store.get_auth_chain([event_id])
event = yield self.store.get_event(event_id)
local_auth_chain = yield self.store.get_auth_chain(
[auth_id for auth_id, _ in event.auth_events],
include_given=True
)
# TODO: Check if we would now reject event_id. If so we need to tell
# everyone.
@@ -1646,6 +1706,17 @@ class FederationHandler(BaseHandler):
@defer.inlineCallbacks
@log_function
def do_auth(self, origin, event, context, auth_events):
"""
Args:
origin (str):
event (synapse.events.FrozenEvent):
context (synapse.events.snapshot.EventContext):
auth_events (dict[(str, str)->str]):
Returns:
defer.Deferred[None]
"""
# Check if we have all the auth events.
current_state = set(e.event_id for e in auth_events.values())
event_auth_events = set(e_id for e_id, _ in event.auth_events)
@@ -1715,7 +1786,7 @@ class FederationHandler(BaseHandler):
[e_id for e_id, _ in event.auth_events]
)
seen_events = set(have_events.keys())
except:
except Exception:
# FIXME:
logger.exception("Failed to get auth chain")
@@ -1728,18 +1799,17 @@ class FederationHandler(BaseHandler):
# Do auth conflict res.
logger.info("Different auth: %s", different_auth)
different_events = yield preserve_context_over_deferred(defer.gatherResults(
[
preserve_fn(self.store.get_event)(
different_events = yield logcontext.make_deferred_yieldable(
defer.gatherResults([
logcontext.preserve_fn(self.store.get_event)(
d,
allow_none=True,
allow_rejected=False,
)
for d in different_auth
if d in have_events and not have_events[d]
],
consumeErrors=True
)).addErrback(unwrapFirstError)
], consumeErrors=True)
).addErrback(unwrapFirstError)
if different_events:
local_view = dict(auth_events)
@@ -1758,16 +1828,9 @@ class FederationHandler(BaseHandler):
current_state = set(e.event_id for e in auth_events.values())
different_auth = event_auth_events - current_state
context.current_state_ids = dict(context.current_state_ids)
context.current_state_ids.update({
k: a.event_id for k, a in auth_events.items()
if k != event_key
})
context.prev_state_ids = dict(context.prev_state_ids)
context.prev_state_ids.update({
k: a.event_id for k, a in auth_events.items()
})
context.state_group = self.store.get_next_state_group()
self._update_context_for_auth_events(
context, auth_events, event_key,
)
if different_auth and not event.internal_metadata.is_outlier():
logger.info("Different auth after resolution: %s", different_auth)
@@ -1791,7 +1854,9 @@ class FederationHandler(BaseHandler):
auth_ids = yield self.auth.compute_auth_events(
event, context.prev_state_ids
)
local_auth_chain = yield self.store.get_auth_chain(auth_ids)
local_auth_chain = yield self.store.get_auth_chain(
auth_ids, include_given=True
)
try:
# 2. Get remote difference.
@@ -1838,23 +1903,16 @@ class FederationHandler(BaseHandler):
except AuthError:
pass
except:
except Exception:
# FIXME:
logger.exception("Failed to query auth chain")
# 4. Look at rejects and their proofs.
# TODO.
context.current_state_ids = dict(context.current_state_ids)
context.current_state_ids.update({
k: a.event_id for k, a in auth_events.items()
if k != event_key
})
context.prev_state_ids = dict(context.prev_state_ids)
context.prev_state_ids.update({
k: a.event_id for k, a in auth_events.items()
})
context.state_group = self.store.get_next_state_group()
self._update_context_for_auth_events(
context, auth_events, event_key,
)
try:
self.auth.check(event, auth_events=auth_events)
@@ -1862,6 +1920,35 @@ class FederationHandler(BaseHandler):
logger.warn("Failed auth resolution for %r because %s", event, e)
raise e
def _update_context_for_auth_events(self, context, auth_events,
event_key):
"""Update the state_ids in an event context after auth event resolution
Args:
context (synapse.events.snapshot.EventContext): event context
to be updated
auth_events (dict[(str, str)->str]): Events to update in the event
context.
event_key ((str, str)): (type, state_key) for the current event.
this will not be included in the current_state in the context.
"""
state_updates = {
k: a.event_id for k, a in auth_events.iteritems()
if k != event_key
}
context.current_state_ids = dict(context.current_state_ids)
context.current_state_ids.update(state_updates)
if context.delta_ids is not None:
context.delta_ids = dict(context.delta_ids)
context.delta_ids.update(state_updates)
context.prev_state_ids = dict(context.prev_state_ids)
context.prev_state_ids.update({
k: a.event_id for k, a in auth_events.iteritems()
})
context.state_group = self.store.get_next_state_group()
@defer.inlineCallbacks
def construct_auth_difference(self, local_auth, remote_auth):
""" Given a local and remote auth chain, find the differences. This
@@ -1905,7 +1992,7 @@ class FederationHandler(BaseHandler):
def get_next(it, opt=None):
try:
return it.next()
except:
except Exception:
return opt
current_local = get_next(local_iter)
@@ -2059,6 +2146,14 @@ class FederationHandler(BaseHandler):
@defer.inlineCallbacks
@log_function
def on_exchange_third_party_invite_request(self, origin, room_id, event_dict):
"""Handle an exchange_third_party_invite request from a remote server
The remote server will call this when it wants to turn a 3pid invite
into a normal m.room.member invite.
Returns:
Deferred: resolves (to None)
"""
builder = self.event_builder_factory.new(event_dict)
message_handler = self.hs.get_handlers().message_handler
@@ -2077,9 +2172,12 @@ class FederationHandler(BaseHandler):
raise e
yield self._check_signature(event, context)
# XXX we send the invite here, but send_membership_event also sends it,
# so we end up making two requests. I think this is redundant.
returned_invite = yield self.send_invite(origin, event)
# TODO: Make sure the signatures actually are correct.
event.signatures.update(returned_invite.signatures)
member_handler = self.hs.get_handlers().room_member_handler
yield member_handler.send_membership_event(None, event, context)
+429
View File
@@ -0,0 +1,429 @@
# -*- coding: utf-8 -*-
# Copyright 2017 Vector Creations Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from twisted.internet import defer
from synapse.api.errors import SynapseError
from synapse.types import get_domain_from_id
import logging
logger = logging.getLogger(__name__)
def _create_rerouter(func_name):
"""Returns a function that looks at the group id and calls the function
on federation or the local group server if the group is local
"""
def f(self, group_id, *args, **kwargs):
if self.is_mine_id(group_id):
return getattr(self.groups_server_handler, func_name)(
group_id, *args, **kwargs
)
else:
destination = get_domain_from_id(group_id)
return getattr(self.transport_client, func_name)(
destination, group_id, *args, **kwargs
)
return f
class GroupsLocalHandler(object):
def __init__(self, hs):
self.hs = hs
self.store = hs.get_datastore()
self.room_list_handler = hs.get_room_list_handler()
self.groups_server_handler = hs.get_groups_server_handler()
self.transport_client = hs.get_federation_transport_client()
self.auth = hs.get_auth()
self.clock = hs.get_clock()
self.keyring = hs.get_keyring()
self.is_mine_id = hs.is_mine_id
self.signing_key = hs.config.signing_key[0]
self.server_name = hs.hostname
self.notifier = hs.get_notifier()
self.attestations = hs.get_groups_attestation_signing()
self.profile_handler = hs.get_profile_handler()
# Ensure attestations get renewed
hs.get_groups_attestation_renewer()
# The following functions merely route the query to the local groups server
# or federation depending on if the group is local or remote
get_group_profile = _create_rerouter("get_group_profile")
update_group_profile = _create_rerouter("update_group_profile")
get_rooms_in_group = _create_rerouter("get_rooms_in_group")
get_invited_users_in_group = _create_rerouter("get_invited_users_in_group")
add_room_to_group = _create_rerouter("add_room_to_group")
update_room_in_group = _create_rerouter("update_room_in_group")
remove_room_from_group = _create_rerouter("remove_room_from_group")
update_group_summary_room = _create_rerouter("update_group_summary_room")
delete_group_summary_room = _create_rerouter("delete_group_summary_room")
update_group_category = _create_rerouter("update_group_category")
delete_group_category = _create_rerouter("delete_group_category")
get_group_category = _create_rerouter("get_group_category")
get_group_categories = _create_rerouter("get_group_categories")
update_group_summary_user = _create_rerouter("update_group_summary_user")
delete_group_summary_user = _create_rerouter("delete_group_summary_user")
update_group_role = _create_rerouter("update_group_role")
delete_group_role = _create_rerouter("delete_group_role")
get_group_role = _create_rerouter("get_group_role")
get_group_roles = _create_rerouter("get_group_roles")
@defer.inlineCallbacks
def get_group_summary(self, group_id, requester_user_id):
"""Get the group summary for a group.
If the group is remote we check that the users have valid attestations.
"""
if self.is_mine_id(group_id):
res = yield self.groups_server_handler.get_group_summary(
group_id, requester_user_id
)
else:
res = yield self.transport_client.get_group_summary(
get_domain_from_id(group_id), group_id, requester_user_id,
)
group_server_name = get_domain_from_id(group_id)
# Loop through the users and validate the attestations.
chunk = res["users_section"]["users"]
valid_users = []
for entry in chunk:
g_user_id = entry["user_id"]
attestation = entry.pop("attestation", {})
try:
if get_domain_from_id(g_user_id) != group_server_name:
yield self.attestations.verify_attestation(
attestation,
group_id=group_id,
user_id=g_user_id,
server_name=get_domain_from_id(g_user_id),
)
valid_users.append(entry)
except Exception as e:
logger.info("Failed to verify user is in group: %s", e)
res["users_section"]["users"] = valid_users
res["users_section"]["users"].sort(key=lambda e: e.get("order", 0))
res["rooms_section"]["rooms"].sort(key=lambda e: e.get("order", 0))
# Add `is_publicised` flag to indicate whether the user has publicised their
# membership of the group on their profile
result = yield self.store.get_publicised_groups_for_user(requester_user_id)
is_publicised = group_id in result
res.setdefault("user", {})["is_publicised"] = is_publicised
defer.returnValue(res)
@defer.inlineCallbacks
def create_group(self, group_id, user_id, content):
"""Create a group
"""
logger.info("Asking to create group with ID: %r", group_id)
if self.is_mine_id(group_id):
res = yield self.groups_server_handler.create_group(
group_id, user_id, content
)
local_attestation = None
remote_attestation = None
else:
local_attestation = self.attestations.create_attestation(group_id, user_id)
content["attestation"] = local_attestation
content["user_profile"] = yield self.profile_handler.get_profile(user_id)
res = yield self.transport_client.create_group(
get_domain_from_id(group_id), group_id, user_id, content,
)
remote_attestation = res["attestation"]
yield self.attestations.verify_attestation(
remote_attestation,
group_id=group_id,
user_id=user_id,
server_name=get_domain_from_id(group_id),
)
is_publicised = content.get("publicise", False)
token = yield self.store.register_user_group_membership(
group_id, user_id,
membership="join",
is_admin=True,
local_attestation=local_attestation,
remote_attestation=remote_attestation,
is_publicised=is_publicised,
)
self.notifier.on_new_event(
"groups_key", token, users=[user_id],
)
defer.returnValue(res)
@defer.inlineCallbacks
def get_users_in_group(self, group_id, requester_user_id):
"""Get users in a group
"""
if self.is_mine_id(group_id):
res = yield self.groups_server_handler.get_users_in_group(
group_id, requester_user_id
)
defer.returnValue(res)
group_server_name = get_domain_from_id(group_id)
res = yield self.transport_client.get_users_in_group(
get_domain_from_id(group_id), group_id, requester_user_id,
)
chunk = res["chunk"]
valid_entries = []
for entry in chunk:
g_user_id = entry["user_id"]
attestation = entry.pop("attestation", {})
try:
if get_domain_from_id(g_user_id) != group_server_name:
yield self.attestations.verify_attestation(
attestation,
group_id=group_id,
user_id=g_user_id,
server_name=get_domain_from_id(g_user_id),
)
valid_entries.append(entry)
except Exception as e:
logger.info("Failed to verify user is in group: %s", e)
res["chunk"] = valid_entries
defer.returnValue(res)
@defer.inlineCallbacks
def join_group(self, group_id, user_id, content):
"""Request to join a group
"""
raise NotImplementedError() # TODO
@defer.inlineCallbacks
def accept_invite(self, group_id, user_id, content):
"""Accept an invite to a group
"""
if self.is_mine_id(group_id):
yield self.groups_server_handler.accept_invite(
group_id, user_id, content
)
local_attestation = None
remote_attestation = None
else:
local_attestation = self.attestations.create_attestation(group_id, user_id)
content["attestation"] = local_attestation
res = yield self.transport_client.accept_group_invite(
get_domain_from_id(group_id), group_id, user_id, content,
)
remote_attestation = res["attestation"]
yield self.attestations.verify_attestation(
remote_attestation,
group_id=group_id,
user_id=user_id,
server_name=get_domain_from_id(group_id),
)
# TODO: Check that the group is public and we're being added publically
is_publicised = content.get("publicise", False)
token = yield self.store.register_user_group_membership(
group_id, user_id,
membership="join",
is_admin=False,
local_attestation=local_attestation,
remote_attestation=remote_attestation,
is_publicised=is_publicised,
)
self.notifier.on_new_event(
"groups_key", token, users=[user_id],
)
defer.returnValue({})
@defer.inlineCallbacks
def invite(self, group_id, user_id, requester_user_id, config):
"""Invite a user to a group
"""
content = {
"requester_user_id": requester_user_id,
"config": config,
}
if self.is_mine_id(group_id):
res = yield self.groups_server_handler.invite_to_group(
group_id, user_id, requester_user_id, content,
)
else:
res = yield self.transport_client.invite_to_group(
get_domain_from_id(group_id), group_id, user_id, requester_user_id,
content,
)
defer.returnValue(res)
@defer.inlineCallbacks
def on_invite(self, group_id, user_id, content):
"""One of our users were invited to a group
"""
# TODO: Support auto join and rejection
if not self.is_mine_id(user_id):
raise SynapseError(400, "User not on this server")
local_profile = {}
if "profile" in content:
if "name" in content["profile"]:
local_profile["name"] = content["profile"]["name"]
if "avatar_url" in content["profile"]:
local_profile["avatar_url"] = content["profile"]["avatar_url"]
token = yield self.store.register_user_group_membership(
group_id, user_id,
membership="invite",
content={"profile": local_profile, "inviter": content["inviter"]},
)
self.notifier.on_new_event(
"groups_key", token, users=[user_id],
)
try:
user_profile = yield self.profile_handler.get_profile(user_id)
except Exception as e:
logger.warn("No profile for user %s: %s", user_id, e)
user_profile = {}
defer.returnValue({"state": "invite", "user_profile": user_profile})
@defer.inlineCallbacks
def remove_user_from_group(self, group_id, user_id, requester_user_id, content):
"""Remove a user from a group
"""
if user_id == requester_user_id:
token = yield self.store.register_user_group_membership(
group_id, user_id,
membership="leave",
)
self.notifier.on_new_event(
"groups_key", token, users=[user_id],
)
# TODO: Should probably remember that we tried to leave so that we can
# retry if the group server is currently down.
if self.is_mine_id(group_id):
res = yield self.groups_server_handler.remove_user_from_group(
group_id, user_id, requester_user_id, content,
)
else:
content["requester_user_id"] = requester_user_id
res = yield self.transport_client.remove_user_from_group(
get_domain_from_id(group_id), group_id, requester_user_id,
user_id, content,
)
defer.returnValue(res)
@defer.inlineCallbacks
def user_removed_from_group(self, group_id, user_id, content):
"""One of our users was removed/kicked from a group
"""
# TODO: Check if user in group
token = yield self.store.register_user_group_membership(
group_id, user_id,
membership="leave",
)
self.notifier.on_new_event(
"groups_key", token, users=[user_id],
)
@defer.inlineCallbacks
def get_joined_groups(self, user_id):
group_ids = yield self.store.get_joined_groups(user_id)
defer.returnValue({"groups": group_ids})
@defer.inlineCallbacks
def get_publicised_groups_for_user(self, user_id):
if self.hs.is_mine_id(user_id):
result = yield self.store.get_publicised_groups_for_user(user_id)
# Check AS associated groups for this user - this depends on the
# RegExps in the AS registration file (under `users`)
for app_service in self.store.get_app_services():
result.extend(app_service.get_groups_for_user(user_id))
defer.returnValue({"groups": result})
else:
result = yield self.transport_client.get_publicised_groups_for_user(
get_domain_from_id(user_id), user_id
)
# TODO: Verify attestations
defer.returnValue(result)
@defer.inlineCallbacks
def bulk_get_publicised_groups(self, user_ids, proxy=True):
destinations = {}
local_users = set()
for user_id in user_ids:
if self.hs.is_mine_id(user_id):
local_users.add(user_id)
else:
destinations.setdefault(
get_domain_from_id(user_id), set()
).add(user_id)
if not proxy and destinations:
raise SynapseError(400, "Some user_ids are not local")
results = {}
failed_results = []
for destination, dest_user_ids in destinations.iteritems():
try:
r = yield self.transport_client.bulk_get_publicised_groups(
destination, list(dest_user_ids),
)
results.update(r["users"])
except Exception:
failed_results.extend(dest_user_ids)
for uid in local_users:
results[uid] = yield self.store.get_publicised_groups_for_user(
uid
)
# Check AS associated groups for this user - this depends on the
# RegExps in the AS registration file (under `users`)
for app_service in self.store.get_app_services():
results[uid].extend(app_service.get_groups_for_user(uid))
defer.returnValue({"users": results})
+3 -3
View File
@@ -27,7 +27,7 @@ from synapse.types import (
from synapse.util import unwrapFirstError
from synapse.util.async import concurrently_execute
from synapse.util.caches.snapshot_cache import SnapshotCache
from synapse.util.logcontext import preserve_fn, preserve_context_over_deferred
from synapse.util.logcontext import make_deferred_yieldable, preserve_fn
from synapse.visibility import filter_events_for_client
from ._base import BaseHandler
@@ -163,7 +163,7 @@ class InitialSyncHandler(BaseHandler):
lambda states: states[event.event_id]
)
(messages, token), current_state = yield preserve_context_over_deferred(
(messages, token), current_state = yield make_deferred_yieldable(
defer.gatherResults(
[
preserve_fn(self.store.get_recent_events_for_room)(
@@ -214,7 +214,7 @@ class InitialSyncHandler(BaseHandler):
})
d["account_data"] = account_data_events
except:
except Exception:
logger.exception("Failed to get snapshot")
yield concurrently_execute(handle_room, room_list, 10)
+72 -3
View File
@@ -1,5 +1,6 @@
# -*- coding: utf-8 -*-
# Copyright 2014 - 2016 OpenMarket Ltd
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
@@ -12,7 +13,6 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from twisted.internet import defer
from synapse.api.constants import EventTypes, Membership
@@ -26,6 +26,7 @@ from synapse.types import (
from synapse.util.async import run_on_reactor, ReadWriteLock, Limiter
from synapse.util.logcontext import preserve_fn
from synapse.util.metrics import measure_func
from synapse.util.frozenutils import unfreeze
from synapse.visibility import filter_events_for_client
from ._base import BaseHandler
@@ -34,6 +35,7 @@ from canonicaljson import encode_canonical_json
import logging
import random
import ujson
logger = logging.getLogger(__name__)
@@ -46,15 +48,20 @@ class MessageHandler(BaseHandler):
self.state = hs.get_state_handler()
self.clock = hs.get_clock()
self.validator = EventValidator()
self.profile_handler = hs.get_profile_handler()
self.pagination_lock = ReadWriteLock()
self.pusher_pool = hs.get_pusherpool()
# We arbitrarily limit concurrent event creation for a room to 5.
# This is to stop us from diverging history *too* much.
self.limiter = Limiter(max_count=5)
self.action_generator = hs.get_action_generator()
self.spam_checker = hs.get_spam_checker()
@defer.inlineCallbacks
def purge_history(self, room_id, event_id):
event = yield self.store.get_event(event_id)
@@ -207,7 +214,7 @@ class MessageHandler(BaseHandler):
if membership in {Membership.JOIN, Membership.INVITE}:
# If event doesn't include a display name, add one.
profile = self.hs.get_handlers().profile_handler
profile = self.profile_handler
content = builder.content
try:
@@ -318,6 +325,15 @@ class MessageHandler(BaseHandler):
token_id=requester.access_token_id,
txn_id=txn_id
)
spam_error = self.spam_checker.check_event_for_spam(event)
if spam_error:
if not isinstance(spam_error, basestring):
spam_error = "Spam is not permitted here"
raise SynapseError(
403, spam_error, Codes.FORBIDDEN
)
yield self.send_nonmember_event(
requester,
event,
@@ -409,6 +425,51 @@ class MessageHandler(BaseHandler):
[serialize_event(c, now) for c in room_state.values()]
)
@defer.inlineCallbacks
def get_joined_members(self, requester, room_id):
"""Get all the joined members in the room and their profile information.
If the user has left the room return the state events from when they left.
Args:
requester(Requester): The user requesting state events.
room_id(str): The room ID to get all state events from.
Returns:
A dict of user_id to profile info
"""
user_id = requester.user.to_string()
if not requester.app_service:
# We check AS auth after fetching the room membership, as it
# requires us to pull out all joined members anyway.
membership, _ = yield self._check_in_room_or_world_readable(
room_id, user_id
)
if membership != Membership.JOIN:
raise NotImplementedError(
"Getting joined members after leaving is not implemented"
)
users_with_profile = yield self.state.get_current_user_in_room(room_id)
# If this is an AS, double check that they are allowed to see the members.
# This can either be because the AS user is in the room or becuase there
# is a user in the room that the AS is "interested in"
if requester.app_service and user_id not in users_with_profile:
for uid in users_with_profile:
if requester.app_service.is_interested_in_user(uid):
break
else:
# Loop fell through, AS has no interested users in room
raise AuthError(403, "Appservice not in room")
defer.returnValue({
user_id: {
"avatar_url": profile.avatar_url,
"display_name": profile.display_name,
}
for user_id, profile in users_with_profile.iteritems()
})
@measure_func("_create_new_client_event")
@defer.inlineCallbacks
def _create_new_client_event(self, builder, requester=None, prev_event_ids=None):
@@ -498,6 +559,14 @@ class MessageHandler(BaseHandler):
logger.warn("Denying new event %r because %s", event, err)
raise err
# Ensure that we can round trip before trying to persist in db
try:
dump = ujson.dumps(unfreeze(event.content))
ujson.loads(dump)
except Exception:
logger.exception("Failed to encode content: %r", event.content)
raise
yield self.maybe_kick_guest_users(event, context)
if event.type == EventTypes.CanonicalAlias:
@@ -601,7 +670,7 @@ class MessageHandler(BaseHandler):
# this intentionally does not yield: we don't care about the result
# and don't need to wait for it.
preserve_fn(self.hs.get_pusherpool().on_new_notifications)(
preserve_fn(self.pusher_pool.on_new_notifications)(
event_stream_id, max_stream_id
)
+2 -2
View File
@@ -364,7 +364,7 @@ class PresenceHandler(object):
)
preserve_fn(self._update_states)(changes)
except:
except Exception:
logger.exception("Exception in _handle_timeouts loop")
@defer.inlineCallbacks
@@ -1199,7 +1199,7 @@ def handle_timeout(state, is_mine, syncing_user_ids, now):
)
changed = True
else:
# We expect to be poked occaisonally by the other side.
# We expect to be poked occasionally by the other side.
# This is to protect against forgetful/buggy servers, so that
# no one gets stuck online forever.
if now - state.last_federation_update_ts > FEDERATION_TIMEOUT:
+125 -17
View File
@@ -17,16 +17,16 @@ import logging
from twisted.internet import defer
import synapse.types
from synapse.api.errors import SynapseError, AuthError, CodeMessageException
from synapse.types import UserID
from synapse.types import UserID, get_domain_from_id
from ._base import BaseHandler
logger = logging.getLogger(__name__)
class ProfileHandler(BaseHandler):
PROFILE_UPDATE_MS = 60 * 1000
PROFILE_UPDATE_EVERY_MS = 24 * 60 * 60 * 1000
def __init__(self, hs):
super(ProfileHandler, self).__init__(hs)
@@ -36,6 +36,65 @@ class ProfileHandler(BaseHandler):
"profile", self.on_profile_query
)
self.user_directory_handler = hs.get_user_directory_handler()
self.clock.looping_call(self._update_remote_profile_cache, self.PROFILE_UPDATE_MS)
@defer.inlineCallbacks
def get_profile(self, user_id):
target_user = UserID.from_string(user_id)
if self.hs.is_mine(target_user):
displayname = yield self.store.get_profile_displayname(
target_user.localpart
)
avatar_url = yield self.store.get_profile_avatar_url(
target_user.localpart
)
defer.returnValue({
"displayname": displayname,
"avatar_url": avatar_url,
})
else:
try:
result = yield self.federation.make_query(
destination=target_user.domain,
query_type="profile",
args={
"user_id": user_id,
},
ignore_backoff=True,
)
defer.returnValue(result)
except CodeMessageException as e:
if e.code != 404:
logger.exception("Failed to get displayname")
raise
@defer.inlineCallbacks
def get_profile_from_cache(self, user_id):
"""Get the profile information from our local cache. If the user is
ours then the profile information will always be corect. Otherwise,
it may be out of date/missing.
"""
target_user = UserID.from_string(user_id)
if self.hs.is_mine(target_user):
displayname = yield self.store.get_profile_displayname(
target_user.localpart
)
avatar_url = yield self.store.get_profile_avatar_url(
target_user.localpart
)
defer.returnValue({
"displayname": displayname,
"avatar_url": avatar_url,
})
else:
profile = yield self.store.get_from_remote_profile_cache(user_id)
defer.returnValue(profile or {})
@defer.inlineCallbacks
def get_displayname(self, target_user):
if self.hs.is_mine(target_user):
@@ -60,7 +119,7 @@ class ProfileHandler(BaseHandler):
logger.exception("Failed to get displayname")
raise
except:
except Exception:
logger.exception("Failed to get displayname")
else:
defer.returnValue(result["displayname"])
@@ -82,7 +141,13 @@ class ProfileHandler(BaseHandler):
target_user.localpart, new_displayname
)
yield self._update_join_states(requester)
if self.hs.config.user_directory_search_all_users:
profile = yield self.store.get_profileinfo(target_user.localpart)
yield self.user_directory_handler.handle_local_profile_change(
target_user.to_string(), profile
)
yield self._update_join_states(requester, target_user)
@defer.inlineCallbacks
def get_avatar_url(self, target_user):
@@ -107,7 +172,7 @@ class ProfileHandler(BaseHandler):
if e.code != 404:
logger.exception("Failed to get avatar_url")
raise
except:
except Exception:
logger.exception("Failed to get avatar_url")
defer.returnValue(result["avatar_url"])
@@ -126,7 +191,13 @@ class ProfileHandler(BaseHandler):
target_user.localpart, new_avatar_url
)
yield self._update_join_states(requester)
if self.hs.config.user_directory_search_all_users:
profile = yield self.store.get_profileinfo(target_user.localpart)
yield self.user_directory_handler.handle_local_profile_change(
target_user.to_string(), profile
)
yield self._update_join_states(requester, target_user)
@defer.inlineCallbacks
def on_profile_query(self, args):
@@ -151,28 +222,24 @@ class ProfileHandler(BaseHandler):
defer.returnValue(response)
@defer.inlineCallbacks
def _update_join_states(self, requester):
user = requester.user
if not self.hs.is_mine(user):
def _update_join_states(self, requester, target_user):
if not self.hs.is_mine(target_user):
return
yield self.ratelimit(requester)
room_ids = yield self.store.get_rooms_for_user(
user.to_string(),
target_user.to_string(),
)
for room_id in room_ids:
handler = self.hs.get_handlers().room_member_handler
try:
# Assume the user isn't a guest because we don't let guests set
# profile or avatar data.
# XXX why are we recreating `requester` here for each room?
# what was wrong with the `requester` we were passed?
requester = synapse.types.create_requester(user)
# Assume the target_user isn't a guest,
# because we don't let guests set profile or avatar data.
yield handler.update_membership(
requester,
user,
target_user,
room_id,
"join", # We treat a profile update like a join.
ratelimit=False, # Try to hide that these events aren't atomic.
@@ -182,3 +249,44 @@ class ProfileHandler(BaseHandler):
"Failed to update join event for room %s - %s",
room_id, str(e.message)
)
def _update_remote_profile_cache(self):
"""Called periodically to check profiles of remote users we haven't
checked in a while.
"""
entries = yield self.store.get_remote_profile_cache_entries_that_expire(
last_checked=self.clock.time_msec() - self.PROFILE_UPDATE_EVERY_MS
)
for user_id, displayname, avatar_url in entries:
is_subscribed = yield self.store.is_subscribed_remote_profile_for_user(
user_id,
)
if not is_subscribed:
yield self.store.maybe_delete_remote_profile_cache(user_id)
continue
try:
profile = yield self.federation.make_query(
destination=get_domain_from_id(user_id),
query_type="profile",
args={
"user_id": user_id,
},
ignore_backoff=True,
)
except Exception:
logger.exception("Failed to get avatar_url")
yield self.store.update_remote_profile_cache(
user_id, displayname, avatar_url
)
continue
new_name = profile.get("displayname")
new_avatar = profile.get("avatar_url")
# We always hit update to update the last_check timestamp
yield self.store.update_remote_profile_cache(
user_id, new_name, new_avatar
)
+4
View File
@@ -12,6 +12,7 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from synapse.util import logcontext
from ._base import BaseHandler
@@ -59,6 +60,8 @@ class ReceiptsHandler(BaseHandler):
is_new = yield self._handle_new_receipts([receipt])
if is_new:
# fire off a process in the background to send the receipt to
# remote servers
self._push_remotes([receipt])
@defer.inlineCallbacks
@@ -126,6 +129,7 @@ class ReceiptsHandler(BaseHandler):
defer.returnValue(True)
@logcontext.preserve_fn # caller should not yield on this
@defer.inlineCallbacks
def _push_remotes(self, receipts):
"""Given a list of receipts, works out which remote servers should be
+19 -13
View File
@@ -15,7 +15,6 @@
"""Contains functions for registering clients."""
import logging
import urllib
from twisted.internet import defer
@@ -23,6 +22,7 @@ from synapse.api.errors import (
AuthError, Codes, SynapseError, RegistrationError, InvalidCaptchaError
)
from synapse.http.client import CaptchaServerHttpClient
from synapse import types
from synapse.types import UserID
from synapse.util.async import run_on_reactor
from ._base import BaseHandler
@@ -36,6 +36,9 @@ class RegistrationHandler(BaseHandler):
super(RegistrationHandler, self).__init__(hs)
self.auth = hs.get_auth()
self._auth_handler = hs.get_auth_handler()
self.profile_handler = hs.get_profile_handler()
self.user_directory_handler = hs.get_user_directory_handler()
self.captcha_client = CaptchaServerHttpClient(hs)
self._next_generated_user_id = None
@@ -45,12 +48,10 @@ class RegistrationHandler(BaseHandler):
@defer.inlineCallbacks
def check_username(self, localpart, guest_access_token=None,
assigned_user_id=None):
yield run_on_reactor()
if urllib.quote(localpart.encode('utf-8')) != localpart:
if types.contains_invalid_mxid_characters(localpart):
raise SynapseError(
400,
"User ID can only contain characters a-z, 0-9, or '_-./'",
"User ID can only contain characters a-z, 0-9, or '=_-./'",
Codes.INVALID_USERNAME
)
@@ -80,7 +81,7 @@ class RegistrationHandler(BaseHandler):
"A different user ID has already been registered for this session",
)
yield self.check_user_id_not_appservice_exclusive(user_id)
self.check_user_id_not_appservice_exclusive(user_id)
users = yield self.store.get_users_by_id_case_insensitive(user_id)
if users:
@@ -165,6 +166,13 @@ class RegistrationHandler(BaseHandler):
),
admin=admin,
)
if self.hs.config.user_directory_search_all_users:
profile = yield self.store.get_profileinfo(localpart)
yield self.user_directory_handler.handle_local_profile_change(
user_id, profile
)
else:
# autogen a sequential user ID
attempts = 0
@@ -253,11 +261,10 @@ class RegistrationHandler(BaseHandler):
"""
Registers email_id as SAML2 Based Auth.
"""
if urllib.quote(localpart) != localpart:
if types.contains_invalid_mxid_characters(localpart):
raise SynapseError(
400,
"User ID must only contain characters which do not"
" require URL encoding."
"User ID can only contain characters a-z, 0-9, or '=_-./'",
)
user = UserID(localpart, self.hs.hostname)
user_id = user.to_string()
@@ -291,7 +298,7 @@ class RegistrationHandler(BaseHandler):
try:
identity_handler = self.hs.get_handlers().identity_handler
threepid = yield identity_handler.threepid_from_creds(c)
except:
except Exception:
logger.exception("Couldn't validate 3pid")
raise RegistrationError(400, "Couldn't validate 3pid")
@@ -418,13 +425,12 @@ class RegistrationHandler(BaseHandler):
create_profile_with_localpart=user.localpart,
)
else:
yield self.store.user_delete_access_tokens(user_id=user_id)
yield self._auth_handler.delete_access_tokens_for_user(user_id)
yield self.store.add_access_token_to_user(user_id=user_id, token=token)
if displayname is not None:
logger.info("setting user display name: %s -> %s", user_id, displayname)
profile_handler = self.hs.get_handlers().profile_handler
yield profile_handler.set_displayname(
yield self.profile_handler.set_displayname(
user, requester, displayname, by_admin=True,
)
+33 -12
View File
@@ -60,8 +60,13 @@ class RoomCreationHandler(BaseHandler):
},
}
def __init__(self, hs):
super(RoomCreationHandler, self).__init__(hs)
self.spam_checker = hs.get_spam_checker()
@defer.inlineCallbacks
def create_room(self, requester, config):
def create_room(self, requester, config, ratelimit=True):
""" Creates a new room.
Args:
@@ -75,14 +80,18 @@ class RoomCreationHandler(BaseHandler):
"""
user_id = requester.user.to_string()
yield self.ratelimit(requester)
if not self.spam_checker.user_may_create_room(user_id):
raise SynapseError(403, "You are not permitted to create rooms")
if ratelimit:
yield self.ratelimit(requester)
if "room_alias_name" in config:
for wchar in string.whitespace:
if wchar in config["room_alias_name"]:
raise SynapseError(400, "Invalid characters in room alias")
room_alias = RoomAlias.create(
room_alias = RoomAlias(
config["room_alias_name"],
self.hs.hostname,
)
@@ -99,7 +108,7 @@ class RoomCreationHandler(BaseHandler):
for i in invite_list:
try:
UserID.from_string(i)
except:
except Exception:
raise SynapseError(400, "Invalid user_id: %s" % (i,))
invite_3pid_list = config.get("invite_3pid", [])
@@ -114,7 +123,7 @@ class RoomCreationHandler(BaseHandler):
while attempts < 5:
try:
random_string = stringutils.random_string(18)
gen_room_id = RoomID.create(
gen_room_id = RoomID(
random_string,
self.hs.hostname,
)
@@ -167,6 +176,7 @@ class RoomCreationHandler(BaseHandler):
initial_state=initial_state,
creation_content=creation_content,
room_alias=room_alias,
power_level_content_override=config.get("power_level_content_override", {})
)
if "name" in config:
@@ -195,12 +205,12 @@ class RoomCreationHandler(BaseHandler):
},
ratelimit=False)
content = {}
is_direct = config.get("is_direct", None)
if is_direct:
content["is_direct"] = is_direct
for invitee in invite_list:
content = {}
is_direct = config.get("is_direct", None)
if is_direct:
content["is_direct"] = is_direct
yield room_member_handler.update_membership(
requester,
UserID.from_string(invitee),
@@ -245,7 +255,8 @@ class RoomCreationHandler(BaseHandler):
invite_list,
initial_state,
creation_content,
room_alias
room_alias,
power_level_content_override,
):
def create(etype, content, **kwargs):
e = {
@@ -291,7 +302,15 @@ class RoomCreationHandler(BaseHandler):
ratelimit=False,
)
if (EventTypes.PowerLevels, '') not in initial_state:
# We treat the power levels override specially as this needs to be one
# of the first events that get sent into a room.
pl_content = initial_state.pop((EventTypes.PowerLevels, ''), None)
if pl_content is not None:
yield send(
etype=EventTypes.PowerLevels,
content=pl_content,
)
else:
power_level_content = {
"users": {
creator_id: 100,
@@ -316,6 +335,8 @@ class RoomCreationHandler(BaseHandler):
for invitee in invite_list:
power_level_content["users"][invitee] = 100
power_level_content.update(power_level_content_override)
yield send(
etype=EventTypes.PowerLevels,
content=power_level_content,
+47 -30
View File
@@ -20,6 +20,7 @@ from ._base import BaseHandler
from synapse.api.constants import (
EventTypes, JoinRules,
)
from synapse.util.logcontext import make_deferred_yieldable, preserve_fn
from synapse.util.async import concurrently_execute
from synapse.util.caches.descriptors import cachedInlineCallbacks
from synapse.util.caches.response_cache import ResponseCache
@@ -70,6 +71,7 @@ class RoomListHandler(BaseHandler):
if search_filter:
# We explicitly don't bother caching searches or requests for
# appservice specific lists.
logger.info("Bypassing cache as search request.")
return self._get_public_room_list(
limit, since_token, search_filter, network_tuple=network_tuple,
)
@@ -77,13 +79,16 @@ class RoomListHandler(BaseHandler):
key = (limit, since_token, network_tuple)
result = self.response_cache.get(key)
if not result:
logger.info("No cached result, calculating one.")
result = self.response_cache.set(
key,
self._get_public_room_list(
preserve_fn(self._get_public_room_list)(
limit, since_token, network_tuple=network_tuple
)
)
return result
else:
logger.info("Using cached deferred result.")
return make_deferred_yieldable(result)
@defer.inlineCallbacks
def _get_public_room_list(self, limit=None, since_token=None,
@@ -149,6 +154,8 @@ class RoomListHandler(BaseHandler):
# We want larger rooms to be first, hence negating num_joined_users
rooms_to_order_value[room_id] = (-num_joined_users, room_id)
logger.info("Getting ordering for %i rooms since %s",
len(room_ids), stream_token)
yield concurrently_execute(get_order_for_room, room_ids, 10)
sorted_entries = sorted(rooms_to_order_value.items(), key=lambda e: e[1])
@@ -176,34 +183,42 @@ class RoomListHandler(BaseHandler):
rooms_to_scan = rooms_to_scan[:since_token.current_limit]
rooms_to_scan.reverse()
# Actually generate the entries. _append_room_entry_to_chunk will append to
# chunk but will stop if len(chunk) > limit
chunk = []
if limit and not search_filter:
logger.info("After sorting and filtering, %i rooms remain",
len(rooms_to_scan))
# _append_room_entry_to_chunk will append to chunk but will stop if
# len(chunk) > limit
#
# Normally we will generate enough results on the first iteration here,
# but if there is a search filter, _append_room_entry_to_chunk may
# filter some results out, in which case we loop again.
#
# We don't want to scan over the entire range either as that
# would potentially waste a lot of work.
#
# XXX if there is no limit, we may end up DoSing the server with
# calls to get_current_state_ids for every single room on the
# server. Surely we should cap this somehow?
#
if limit:
step = limit + 1
for i in xrange(0, len(rooms_to_scan), step):
# We iterate here because the vast majority of cases we'll stop
# at first iteration, but occaisonally _append_room_entry_to_chunk
# won't append to the chunk and so we need to loop again.
# We don't want to scan over the entire range either as that
# would potentially waste a lot of work.
yield concurrently_execute(
lambda r: self._append_room_entry_to_chunk(
r, rooms_to_num_joined[r],
chunk, limit, search_filter
),
rooms_to_scan[i:i + step], 10
)
if len(chunk) >= limit + 1:
break
else:
step = len(rooms_to_scan)
chunk = []
for i in xrange(0, len(rooms_to_scan), step):
batch = rooms_to_scan[i:i + step]
logger.info("Processing %i rooms for result", len(batch))
yield concurrently_execute(
lambda r: self._append_room_entry_to_chunk(
r, rooms_to_num_joined[r],
chunk, limit, search_filter
),
rooms_to_scan, 5
batch, 5,
)
logger.info("Now %i rooms in result", len(chunk))
if len(chunk) >= limit + 1:
break
chunk.sort(key=lambda e: (-e["num_joined_members"], e["room_id"]))
@@ -276,13 +291,14 @@ class RoomListHandler(BaseHandler):
# We've already got enough, so lets just drop it.
return
result = yield self._generate_room_entry(room_id, num_joined_users)
result = yield self.generate_room_entry(room_id, num_joined_users)
if result and _matches_room_entry(result, search_filter):
chunk.append(result)
@cachedInlineCallbacks(num_args=1, cache_context=True)
def _generate_room_entry(self, room_id, num_joined_users, cache_context):
def generate_room_entry(self, room_id, num_joined_users, cache_context,
with_alias=True, allow_private=False):
"""Returns the entry for a room
"""
result = {
@@ -316,14 +332,15 @@ class RoomListHandler(BaseHandler):
join_rules_event = current_state.get((EventTypes.JoinRules, ""))
if join_rules_event:
join_rule = join_rules_event.content.get("join_rule", None)
if join_rule and join_rule != JoinRules.PUBLIC:
if not allow_private and join_rule and join_rule != JoinRules.PUBLIC:
defer.returnValue(None)
aliases = yield self.store.get_aliases_for_room(
room_id, on_invalidate=cache_context.invalidate
)
if aliases:
result["aliases"] = aliases
if with_alias:
aliases = yield self.store.get_aliases_for_room(
room_id, on_invalidate=cache_context.invalidate
)
if aliases:
result["aliases"] = aliases
name_event = yield current_state.get((EventTypes.Name, ""))
if name_event:
+54 -1
View File
@@ -45,9 +45,12 @@ class RoomMemberHandler(BaseHandler):
def __init__(self, hs):
super(RoomMemberHandler, self).__init__(hs)
self.profile_handler = hs.get_profile_handler()
self.member_linearizer = Linearizer(name="member")
self.clock = hs.get_clock()
self.spam_checker = hs.get_spam_checker()
self.distributor = hs.get_distributor()
self.distributor.declare("user_joined_room")
@@ -186,11 +189,17 @@ class RoomMemberHandler(BaseHandler):
content_specified = bool(content)
if content is None:
content = {}
else:
# We do a copy here as we potentially change some keys
# later on.
content = dict(content)
effective_membership_state = action
if action in ["kick", "unban"]:
effective_membership_state = "leave"
# if this is a join with a 3pid signature, we may need to turn a 3pid
# invite into a normal invite before we can handle the join.
if third_party_signed is not None:
replication = self.hs.get_replication_layer()
yield replication.exchange_third_party_invite(
@@ -203,6 +212,35 @@ class RoomMemberHandler(BaseHandler):
if not remote_room_hosts:
remote_room_hosts = []
if effective_membership_state not in ("leave", "ban",):
is_blocked = yield self.store.is_room_blocked(room_id)
if is_blocked:
raise SynapseError(403, "This room has been blocked on this server")
if effective_membership_state == "invite":
block_invite = False
is_requester_admin = yield self.auth.is_server_admin(
requester.user,
)
if not is_requester_admin:
if self.hs.config.block_non_admin_invites:
logger.info(
"Blocking invite: user is not admin and non-admin "
"invites disabled"
)
block_invite = True
if not self.spam_checker.user_may_invite(
requester.user.to_string(), target.to_string(), room_id,
):
logger.info("Blocking invite due to spam checker")
block_invite = True
if block_invite:
raise SynapseError(
403, "Invites have been disabled on this server",
)
latest_event_ids = yield self.store.get_latest_event_ids_in_room(room_id)
current_state_ids = yield self.state_handler.get_current_state_ids(
room_id, latest_event_ids=latest_event_ids,
@@ -250,7 +288,7 @@ class RoomMemberHandler(BaseHandler):
content["membership"] = Membership.JOIN
profile = self.hs.get_handlers().profile_handler
profile = self.profile_handler
if not content_specified:
content["displayname"] = yield profile.get_displayname(target)
content["avatar_url"] = yield profile.get_avatar_url(target)
@@ -369,6 +407,11 @@ class RoomMemberHandler(BaseHandler):
# so don't really fit into the general auth process.
raise AuthError(403, "Guest access not allowed")
if event.membership not in (Membership.LEAVE, Membership.BAN):
is_blocked = yield self.store.is_room_blocked(room_id)
if is_blocked:
raise SynapseError(403, "This room has been blocked on this server")
yield message_handler.handle_new_client_event(
requester,
event,
@@ -461,6 +504,16 @@ class RoomMemberHandler(BaseHandler):
requester,
txn_id
):
if self.hs.config.block_non_admin_invites:
is_requester_admin = yield self.auth.is_server_admin(
requester.user,
)
if not is_requester_admin:
raise SynapseError(
403, "Invites have been disabled on this server",
Codes.FORBIDDEN,
)
invitee = yield self._lookup_3pid(
id_server, medium, address
)
+1 -1
View File
@@ -61,7 +61,7 @@ class SearchHandler(BaseHandler):
assert batch_group is not None
assert batch_group_key is not None
assert batch_token is not None
except:
except Exception:
raise SynapseError(400, "Invalid batch")
try:
+56
View File
@@ -0,0 +1,56 @@
# -*- coding: utf-8 -*-
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
from twisted.internet import defer
from synapse.api.errors import Codes, StoreError, SynapseError
from ._base import BaseHandler
logger = logging.getLogger(__name__)
class SetPasswordHandler(BaseHandler):
"""Handler which deals with changing user account passwords"""
def __init__(self, hs):
super(SetPasswordHandler, self).__init__(hs)
self._auth_handler = hs.get_auth_handler()
self._device_handler = hs.get_device_handler()
@defer.inlineCallbacks
def set_password(self, user_id, newpassword, requester=None):
password_hash = self._auth_handler.hash(newpassword)
except_device_id = requester.device_id if requester else None
except_access_token_id = requester.access_token_id if requester else None
try:
yield self.store.user_set_password_hash(user_id, password_hash)
except StoreError as e:
if e.code == 404:
raise SynapseError(404, "Unknown user", Codes.NOT_FOUND)
raise e
# we want to log out all of the user's other sessions. First delete
# all his other devices.
yield self._device_handler.delete_all_devices_for_user(
user_id, except_device_id=except_device_id,
)
# and now delete any access tokens which weren't associated with
# devices (or were associated with this device).
yield self._auth_handler.delete_access_tokens_for_user(
user_id, except_token_id=except_access_token_id,
)
+209 -26
View File
@@ -15,7 +15,7 @@
from synapse.api.constants import Membership, EventTypes
from synapse.util.async import concurrently_execute
from synapse.util.logcontext import LoggingContext
from synapse.util.logcontext import LoggingContext, make_deferred_yieldable, preserve_fn
from synapse.util.metrics import Measure, measure_func
from synapse.util.caches.response_cache import ResponseCache
from synapse.push.clientformat import format_push_rules_for_user
@@ -108,6 +108,27 @@ class InvitedSyncResult(collections.namedtuple("InvitedSyncResult", [
return True
class GroupsSyncResult(collections.namedtuple("GroupsSyncResult", [
"join",
"invite",
"leave",
])):
__slots__ = []
def __nonzero__(self):
return bool(self.join or self.invite or self.leave)
class DeviceLists(collections.namedtuple("DeviceLists", [
"changed", # list of user_ids whose devices may have changed
"left", # list of user_ids whose devices we no longer track
])):
__slots__ = []
def __nonzero__(self):
return bool(self.changed or self.left)
class SyncResult(collections.namedtuple("SyncResult", [
"next_batch", # Token for the next sync
"presence", # List of presence events for the user.
@@ -117,6 +138,9 @@ class SyncResult(collections.namedtuple("SyncResult", [
"archived", # ArchivedSyncResult for each archived room.
"to_device", # List of direct messages for the device.
"device_lists", # List of user_ids whose devices have chanegd
"device_one_time_keys_count", # Dict of algorithm to count for one time keys
# for this device
"groups",
])):
__slots__ = []
@@ -132,7 +156,8 @@ class SyncResult(collections.namedtuple("SyncResult", [
self.archived or
self.account_data or
self.to_device or
self.device_lists
self.device_lists or
self.groups
)
@@ -159,11 +184,11 @@ class SyncHandler(object):
if not result:
result = self.response_cache.set(
sync_config.request_key,
self._wait_for_sync_for_user(
preserve_fn(self._wait_for_sync_for_user)(
sync_config, since_token, timeout, full_state
)
)
return result
return make_deferred_yieldable(result)
@defer.inlineCallbacks
def _wait_for_sync_for_user(self, sync_config, since_token, timeout,
@@ -288,10 +313,20 @@ class SyncHandler(object):
if recents:
recents = sync_config.filter_collection.filter_room_timeline(recents)
# We check if there are any state events, if there are then we pass
# all current state events to the filter_events function. This is to
# ensure that we always include current state in the timeline
current_state_ids = frozenset()
if any(e.is_state() for e in recents):
current_state_ids = yield self.state.get_current_state_ids(room_id)
current_state_ids = frozenset(current_state_ids.itervalues())
recents = yield filter_events_for_client(
self.store,
sync_config.user.to_string(),
recents,
always_include_ids=current_state_ids,
)
else:
recents = []
@@ -323,10 +358,20 @@ class SyncHandler(object):
loaded_recents = sync_config.filter_collection.filter_room_timeline(
events
)
# We check if there are any state events, if there are then we pass
# all current state events to the filter_events function. This is to
# ensure that we always include current state in the timeline
current_state_ids = frozenset()
if any(e.is_state() for e in loaded_recents):
current_state_ids = yield self.state.get_current_state_ids(room_id)
current_state_ids = frozenset(current_state_ids.itervalues())
loaded_recents = yield filter_events_for_client(
self.store,
sync_config.user.to_string(),
loaded_recents,
always_include_ids=current_state_ids,
)
loaded_recents.extend(recents)
recents = loaded_recents
@@ -533,7 +578,8 @@ class SyncHandler(object):
res = yield self._generate_sync_entry_for_rooms(
sync_result_builder, account_data_by_room
)
newly_joined_rooms, newly_joined_users = res
newly_joined_rooms, newly_joined_users, _, _ = res
_, _, newly_left_rooms, newly_left_users = res
block_all_presence_data = (
since_token is None and
@@ -547,9 +593,23 @@ class SyncHandler(object):
yield self._generate_sync_entry_for_to_device(sync_result_builder)
device_lists = yield self._generate_sync_entry_for_device_list(
sync_result_builder
sync_result_builder,
newly_joined_rooms=newly_joined_rooms,
newly_joined_users=newly_joined_users,
newly_left_rooms=newly_left_rooms,
newly_left_users=newly_left_users,
)
device_id = sync_config.device_id
one_time_key_counts = {}
if device_id:
user_id = sync_config.user.to_string()
one_time_key_counts = yield self.store.count_e2e_one_time_keys(
user_id, device_id
)
yield self._generate_sync_entry_for_groups(sync_result_builder)
defer.returnValue(SyncResult(
presence=sync_result_builder.presence,
account_data=sync_result_builder.account_data,
@@ -558,30 +618,103 @@ class SyncHandler(object):
archived=sync_result_builder.archived,
to_device=sync_result_builder.to_device,
device_lists=device_lists,
groups=sync_result_builder.groups,
device_one_time_keys_count=one_time_key_counts,
next_batch=sync_result_builder.now_token,
))
@measure_func("_generate_sync_entry_for_groups")
@defer.inlineCallbacks
def _generate_sync_entry_for_groups(self, sync_result_builder):
user_id = sync_result_builder.sync_config.user.to_string()
since_token = sync_result_builder.since_token
now_token = sync_result_builder.now_token
if since_token and since_token.groups_key:
results = yield self.store.get_groups_changes_for_user(
user_id, since_token.groups_key, now_token.groups_key,
)
else:
results = yield self.store.get_all_groups_for_user(
user_id, now_token.groups_key,
)
invited = {}
joined = {}
left = {}
for result in results:
membership = result["membership"]
group_id = result["group_id"]
gtype = result["type"]
content = result["content"]
if membership == "join":
if gtype == "membership":
# TODO: Add profile
content.pop("membership", None)
joined[group_id] = content["content"]
else:
joined.setdefault(group_id, {})[gtype] = content
elif membership == "invite":
if gtype == "membership":
content.pop("membership", None)
invited[group_id] = content["content"]
else:
if gtype == "membership":
left[group_id] = content["content"]
sync_result_builder.groups = GroupsSyncResult(
join=joined,
invite=invited,
leave=left,
)
@measure_func("_generate_sync_entry_for_device_list")
@defer.inlineCallbacks
def _generate_sync_entry_for_device_list(self, sync_result_builder):
def _generate_sync_entry_for_device_list(self, sync_result_builder,
newly_joined_rooms, newly_joined_users,
newly_left_rooms, newly_left_users):
user_id = sync_result_builder.sync_config.user.to_string()
since_token = sync_result_builder.since_token
if since_token and since_token.device_list_key:
room_ids = yield self.store.get_rooms_for_user(user_id)
user_ids_changed = set()
changed = yield self.store.get_user_whose_devices_changed(
since_token.device_list_key
)
for other_user_id in changed:
other_room_ids = yield self.store.get_rooms_for_user(other_user_id)
if room_ids.intersection(other_room_ids):
user_ids_changed.add(other_user_id)
defer.returnValue(user_ids_changed)
# TODO: Be more clever than this, i.e. remove users who we already
# share a room with?
for room_id in newly_joined_rooms:
joined_users = yield self.state.get_current_user_in_room(room_id)
newly_joined_users.update(joined_users)
for room_id in newly_left_rooms:
left_users = yield self.state.get_current_user_in_room(room_id)
newly_left_users.update(left_users)
# TODO: Check that these users are actually new, i.e. either they
# weren't in the previous sync *or* they left and rejoined.
changed.update(newly_joined_users)
if not changed and not newly_left_users:
defer.returnValue(DeviceLists(
changed=[],
left=newly_left_users,
))
users_who_share_room = yield self.store.get_users_who_share_room_with_user(
user_id
)
defer.returnValue(DeviceLists(
changed=users_who_share_room & changed,
left=set(newly_left_users) - users_who_share_room,
))
else:
defer.returnValue([])
defer.returnValue(DeviceLists(
changed=[],
left=[],
))
@defer.inlineCallbacks
def _generate_sync_entry_for_to_device(self, sync_result_builder):
@@ -745,8 +878,8 @@ class SyncHandler(object):
account_data_by_room(dict): Dictionary of per room account data
Returns:
Deferred(tuple): Returns a 2-tuple of
`(newly_joined_rooms, newly_joined_users)`
Deferred(tuple): Returns a 4-tuple of
`(newly_joined_rooms, newly_joined_users, newly_left_rooms, newly_left_users)`
"""
user_id = sync_result_builder.sync_config.user.to_string()
block_all_room_ephemeral = (
@@ -777,7 +910,7 @@ class SyncHandler(object):
)
if not tags_by_room:
logger.debug("no-oping sync")
defer.returnValue(([], []))
defer.returnValue(([], [], [], []))
ignored_account_data = yield self.store.get_global_account_data_by_type_for_user(
"m.ignored_user_list", user_id=user_id,
@@ -790,7 +923,7 @@ class SyncHandler(object):
if since_token:
res = yield self._get_rooms_changed(sync_result_builder, ignored_users)
room_entries, invited, newly_joined_rooms = res
room_entries, invited, newly_joined_rooms, newly_left_rooms = res
tags_by_room = yield self.store.get_updated_tags(
user_id, since_token.account_data_key,
@@ -798,6 +931,7 @@ class SyncHandler(object):
else:
res = yield self._get_all_rooms(sync_result_builder, ignored_users)
room_entries, invited, newly_joined_rooms = res
newly_left_rooms = []
tags_by_room = yield self.store.get_tags_for_user(user_id)
@@ -818,17 +952,30 @@ class SyncHandler(object):
# Now we want to get any newly joined users
newly_joined_users = set()
newly_left_users = set()
if since_token:
for joined_sync in sync_result_builder.joined:
it = itertools.chain(
joined_sync.timeline.events, joined_sync.state.values()
joined_sync.timeline.events, joined_sync.state.itervalues()
)
for event in it:
if event.type == EventTypes.Member:
if event.membership == Membership.JOIN:
newly_joined_users.add(event.state_key)
else:
prev_content = event.unsigned.get("prev_content", {})
prev_membership = prev_content.get("membership", None)
if prev_membership == Membership.JOIN:
newly_left_users.add(event.state_key)
defer.returnValue((newly_joined_rooms, newly_joined_users))
newly_left_users -= newly_joined_users
defer.returnValue((
newly_joined_rooms,
newly_joined_users,
newly_left_rooms,
newly_left_users,
))
@defer.inlineCallbacks
def _have_rooms_changed(self, sync_result_builder):
@@ -898,15 +1045,28 @@ class SyncHandler(object):
mem_change_events_by_room_id.setdefault(event.room_id, []).append(event)
newly_joined_rooms = []
newly_left_rooms = []
room_entries = []
invited = []
for room_id, events in mem_change_events_by_room_id.items():
for room_id, events in mem_change_events_by_room_id.iteritems():
non_joins = [e for e in events if e.membership != Membership.JOIN]
has_join = len(non_joins) != len(events)
# We want to figure out if we joined the room at some point since
# the last sync (even if we have since left). This is to make sure
# we do send down the room, and with full state, where necessary
old_state_ids = None
if room_id in joined_room_ids and non_joins:
# Always include if the user (re)joined the room, especially
# important so that device list changes are calculated correctly.
# If there are non join member events, but we are still in the room,
# then the user must have left and joined
newly_joined_rooms.append(room_id)
# User is in the room so we don't need to do the invite/leave checks
continue
if room_id in joined_room_ids or has_join:
old_state_ids = yield self.get_state_at(room_id, since_token)
old_mem_ev_id = old_state_ids.get((EventTypes.Member, user_id), None)
@@ -918,12 +1078,33 @@ class SyncHandler(object):
if not old_mem_ev or old_mem_ev.membership != Membership.JOIN:
newly_joined_rooms.append(room_id)
if room_id in joined_room_ids:
continue
# If user is in the room then we don't need to do the invite/leave checks
if room_id in joined_room_ids:
continue
if not non_joins:
continue
# Check if we have left the room. This can either be because we were
# joined before *or* that we since joined and then left.
if events[-1].membership != Membership.JOIN:
if has_join:
newly_left_rooms.append(room_id)
else:
if not old_state_ids:
old_state_ids = yield self.get_state_at(room_id, since_token)
old_mem_ev_id = old_state_ids.get(
(EventTypes.Member, user_id),
None,
)
old_mem_ev = None
if old_mem_ev_id:
old_mem_ev = yield self.store.get_event(
old_mem_ev_id, allow_none=True
)
if old_mem_ev and old_mem_ev.membership == Membership.JOIN:
newly_left_rooms.append(room_id)
# Only bother if we're still currently invited
should_invite = non_joins[-1].membership == Membership.INVITE
if should_invite:
@@ -1001,7 +1182,7 @@ class SyncHandler(object):
upto_token=since_token,
))
defer.returnValue((room_entries, invited, newly_joined_rooms))
defer.returnValue((room_entries, invited, newly_joined_rooms, newly_left_rooms))
@defer.inlineCallbacks
def _get_all_rooms(self, sync_result_builder, ignored_users):
@@ -1249,6 +1430,8 @@ class SyncResultBuilder(object):
self.invited = []
self.archived = []
self.device = []
self.groups = None
self.to_device = []
class RoomSyncResultBuilder(object):
+5 -7
View File
@@ -89,7 +89,7 @@ class TypingHandler(object):
until = self._member_typing_until.get(member, None)
if not until or until <= now:
logger.info("Timing out typing for: %s", member.user_id)
preserve_fn(self._stopped_typing)(member)
self._stopped_typing(member)
continue
# Check if we need to resend a keep alive over federation for this
@@ -147,7 +147,7 @@ class TypingHandler(object):
# No point sending another notification
defer.returnValue(None)
yield self._push_update(
self._push_update(
member=member,
typing=True,
)
@@ -171,7 +171,7 @@ class TypingHandler(object):
member = RoomMember(room_id=room_id, user_id=target_user_id)
yield self._stopped_typing(member)
self._stopped_typing(member)
@defer.inlineCallbacks
def user_left_room(self, user, room_id):
@@ -180,7 +180,6 @@ class TypingHandler(object):
member = RoomMember(room_id=room_id, user_id=user_id)
yield self._stopped_typing(member)
@defer.inlineCallbacks
def _stopped_typing(self, member):
if member.user_id not in self._room_typing.get(member.room_id, set()):
# No point
@@ -189,16 +188,15 @@ class TypingHandler(object):
self._member_typing_until.pop(member, None)
self._member_last_federation_poke.pop(member, None)
yield self._push_update(
self._push_update(
member=member,
typing=False,
)
@defer.inlineCallbacks
def _push_update(self, member, typing):
if self.hs.is_mine_id(member.user_id):
# Only send updates for changes to our own users.
yield self._push_remote(member, typing)
preserve_fn(self._push_remote)(member, typing)
self._push_update_local(
member=member,
+681
View File
@@ -0,0 +1,681 @@
# -*- coding: utf-8 -*-
# Copyright 2017 Vector Creations Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
from twisted.internet import defer
from synapse.api.constants import EventTypes, JoinRules, Membership
from synapse.storage.roommember import ProfileInfo
from synapse.util.metrics import Measure
from synapse.util.async import sleep
from synapse.types import get_localpart_from_id
logger = logging.getLogger(__name__)
class UserDirectoryHandler(object):
"""Handles querying of and keeping updated the user_directory.
N.B.: ASSUMES IT IS THE ONLY THING THAT MODIFIES THE USER DIRECTORY
The user directory is filled with users who this server can see are joined to a
world_readable or publically joinable room. We keep a database table up to date
by streaming changes of the current state and recalculating whether users should
be in the directory or not when necessary.
For each user in the directory we also store a room_id which is public and that the
user is joined to. This allows us to ignore history_visibility and join_rules changes
for that user in all other public rooms, as we know they'll still be in at least
one public room.
"""
INITIAL_ROOM_SLEEP_MS = 50
INITIAL_ROOM_SLEEP_COUNT = 100
INITIAL_ROOM_BATCH_SIZE = 100
INITIAL_USER_SLEEP_MS = 10
def __init__(self, hs):
self.store = hs.get_datastore()
self.state = hs.get_state_handler()
self.server_name = hs.hostname
self.clock = hs.get_clock()
self.notifier = hs.get_notifier()
self.is_mine_id = hs.is_mine_id
self.update_user_directory = hs.config.update_user_directory
self.search_all_users = hs.config.user_directory_search_all_users
# When start up for the first time we need to populate the user_directory.
# This is a set of user_id's we've inserted already
self.initially_handled_users = set()
self.initially_handled_users_in_public = set()
self.initially_handled_users_share = set()
self.initially_handled_users_share_private_room = set()
# The current position in the current_state_delta stream
self.pos = None
# Guard to ensure we only process deltas one at a time
self._is_processing = False
if self.update_user_directory:
self.notifier.add_replication_callback(self.notify_new_event)
# We kick this off so that we don't have to wait for a change before
# we start populating the user directory
self.clock.call_later(0, self.notify_new_event)
def search_users(self, user_id, search_term, limit):
"""Searches for users in directory
Returns:
dict of the form::
{
"limited": <bool>, # whether there were more results or not
"results": [ # Ordered by best match first
{
"user_id": <user_id>,
"display_name": <display_name>,
"avatar_url": <avatar_url>
}
]
}
"""
return self.store.search_user_dir(user_id, search_term, limit)
@defer.inlineCallbacks
def notify_new_event(self):
"""Called when there may be more deltas to process
"""
if not self.update_user_directory:
return
if self._is_processing:
return
self._is_processing = True
try:
yield self._unsafe_process()
finally:
self._is_processing = False
@defer.inlineCallbacks
def handle_local_profile_change(self, user_id, profile):
"""Called to update index of our local user profiles when they change
irrespective of any rooms the user may be in.
"""
yield self.store.update_profile_in_user_dir(
user_id, profile.display_name, profile.avatar_url, None,
)
@defer.inlineCallbacks
def _unsafe_process(self):
# If self.pos is None then means we haven't fetched it from DB
if self.pos is None:
self.pos = yield self.store.get_user_directory_stream_pos()
# If still None then we need to do the initial fill of directory
if self.pos is None:
yield self._do_initial_spam()
self.pos = yield self.store.get_user_directory_stream_pos()
# Loop round handling deltas until we're up to date
while True:
with Measure(self.clock, "user_dir_delta"):
deltas = yield self.store.get_current_state_deltas(self.pos)
if not deltas:
return
logger.info("Handling %d state deltas", len(deltas))
yield self._handle_deltas(deltas)
self.pos = deltas[-1]["stream_id"]
yield self.store.update_user_directory_stream_pos(self.pos)
@defer.inlineCallbacks
def _do_initial_spam(self):
"""Populates the user_directory from the current state of the DB, used
when synapse first starts with user_directory support
"""
new_pos = yield self.store.get_max_stream_id_in_current_state_deltas()
# Delete any existing entries just in case there are any
yield self.store.delete_all_from_user_dir()
# We process by going through each existing room at a time.
room_ids = yield self.store.get_all_rooms()
logger.info("Doing initial update of user directory. %d rooms", len(room_ids))
num_processed_rooms = 0
for room_id in room_ids:
logger.info("Handling room %d/%d", num_processed_rooms + 1, len(room_ids))
yield self._handle_initial_room(room_id)
num_processed_rooms += 1
yield sleep(self.INITIAL_ROOM_SLEEP_MS / 1000.)
logger.info("Processed all rooms.")
if self.search_all_users:
num_processed_users = 0
user_ids = yield self.store.get_all_local_users()
logger.info("Doing initial update of user directory. %d users", len(user_ids))
for user_id in user_ids:
# We add profiles for all users even if they don't match the
# include pattern, just in case we want to change it in future
logger.info("Handling user %d/%d", num_processed_users + 1, len(user_ids))
yield self._handle_local_user(user_id)
num_processed_users += 1
yield sleep(self.INITIAL_USER_SLEEP_MS / 1000.)
logger.info("Processed all users")
self.initially_handled_users = None
self.initially_handled_users_in_public = None
self.initially_handled_users_share = None
self.initially_handled_users_share_private_room = None
yield self.store.update_user_directory_stream_pos(new_pos)
@defer.inlineCallbacks
def _handle_initial_room(self, room_id):
"""Called when we initially fill out user_directory one room at a time
"""
is_in_room = yield self.store.is_host_joined(room_id, self.server_name)
if not is_in_room:
return
is_public = yield self.store.is_room_world_readable_or_publicly_joinable(room_id)
users_with_profile = yield self.state.get_current_user_in_room(room_id)
user_ids = set(users_with_profile)
unhandled_users = user_ids - self.initially_handled_users
yield self.store.add_profiles_to_user_dir(
room_id, {
user_id: users_with_profile[user_id] for user_id in unhandled_users
}
)
self.initially_handled_users |= unhandled_users
if is_public:
yield self.store.add_users_to_public_room(
room_id,
user_ids=user_ids - self.initially_handled_users_in_public
)
self.initially_handled_users_in_public |= user_ids
# We now go and figure out the new users who share rooms with user entries
# We sleep aggressively here as otherwise it can starve resources.
# We also batch up inserts/updates, but try to avoid too many at once.
to_insert = set()
to_update = set()
count = 0
for user_id in user_ids:
if count % self.INITIAL_ROOM_SLEEP_COUNT == 0:
yield sleep(self.INITIAL_ROOM_SLEEP_MS / 1000.)
if not self.is_mine_id(user_id):
count += 1
continue
if self.store.get_if_app_services_interested_in_user(user_id):
count += 1
continue
for other_user_id in user_ids:
if user_id == other_user_id:
continue
if count % self.INITIAL_ROOM_SLEEP_COUNT == 0:
yield sleep(self.INITIAL_ROOM_SLEEP_MS / 1000.)
count += 1
user_set = (user_id, other_user_id)
if user_set in self.initially_handled_users_share_private_room:
continue
if user_set in self.initially_handled_users_share:
if is_public:
continue
to_update.add(user_set)
else:
to_insert.add(user_set)
if is_public:
self.initially_handled_users_share.add(user_set)
else:
self.initially_handled_users_share_private_room.add(user_set)
if len(to_insert) > self.INITIAL_ROOM_BATCH_SIZE:
yield self.store.add_users_who_share_room(
room_id, not is_public, to_insert,
)
to_insert.clear()
if len(to_update) > self.INITIAL_ROOM_BATCH_SIZE:
yield self.store.update_users_who_share_room(
room_id, not is_public, to_update,
)
to_update.clear()
if to_insert:
yield self.store.add_users_who_share_room(
room_id, not is_public, to_insert,
)
to_insert.clear()
if to_update:
yield self.store.update_users_who_share_room(
room_id, not is_public, to_update,
)
to_update.clear()
@defer.inlineCallbacks
def _handle_deltas(self, deltas):
"""Called with the state deltas to process
"""
for delta in deltas:
typ = delta["type"]
state_key = delta["state_key"]
room_id = delta["room_id"]
event_id = delta["event_id"]
prev_event_id = delta["prev_event_id"]
logger.debug("Handling: %r %r, %s", typ, state_key, event_id)
# For join rule and visibility changes we need to check if the room
# may have become public or not and add/remove the users in said room
if typ in (EventTypes.RoomHistoryVisibility, EventTypes.JoinRules):
yield self._handle_room_publicity_change(
room_id, prev_event_id, event_id, typ,
)
elif typ == EventTypes.Member:
change = yield self._get_key_change(
prev_event_id, event_id,
key_name="membership",
public_value=Membership.JOIN,
)
if change is None:
# Handle any profile changes
yield self._handle_profile_change(
state_key, room_id, prev_event_id, event_id,
)
continue
if not change:
# Need to check if the server left the room entirely, if so
# we might need to remove all the users in that room
is_in_room = yield self.store.is_host_joined(
room_id, self.server_name,
)
if not is_in_room:
logger.info("Server left room: %r", room_id)
# Fetch all the users that we marked as being in user
# directory due to being in the room and then check if
# need to remove those users or not
user_ids = yield self.store.get_users_in_dir_due_to_room(room_id)
for user_id in user_ids:
yield self._handle_remove_user(room_id, user_id)
return
else:
logger.debug("Server is still in room: %r", room_id)
if change: # The user joined
event = yield self.store.get_event(event_id, allow_none=True)
profile = ProfileInfo(
avatar_url=event.content.get("avatar_url"),
display_name=event.content.get("displayname"),
)
yield self._handle_new_user(room_id, state_key, profile)
else: # The user left
yield self._handle_remove_user(room_id, state_key)
else:
logger.debug("Ignoring irrelevant type: %r", typ)
@defer.inlineCallbacks
def _handle_room_publicity_change(self, room_id, prev_event_id, event_id, typ):
"""Handle a room having potentially changed from/to world_readable/publically
joinable.
Args:
room_id (str)
prev_event_id (str|None): The previous event before the state change
event_id (str|None): The new event after the state change
typ (str): Type of the event
"""
logger.debug("Handling change for %s: %s", typ, room_id)
if typ == EventTypes.RoomHistoryVisibility:
change = yield self._get_key_change(
prev_event_id, event_id,
key_name="history_visibility",
public_value="world_readable",
)
elif typ == EventTypes.JoinRules:
change = yield self._get_key_change(
prev_event_id, event_id,
key_name="join_rule",
public_value=JoinRules.PUBLIC,
)
else:
raise Exception("Invalid event type")
# If change is None, no change. True => become world_readable/public,
# False => was world_readable/public
if change is None:
logger.debug("No change")
return
# There's been a change to or from being world readable.
is_public = yield self.store.is_room_world_readable_or_publicly_joinable(
room_id
)
logger.debug("Change: %r, is_public: %r", change, is_public)
if change and not is_public:
# If we became world readable but room isn't currently public then
# we ignore the change
return
elif not change and is_public:
# If we stopped being world readable but are still public,
# ignore the change
return
if change:
users_with_profile = yield self.state.get_current_user_in_room(room_id)
for user_id, profile in users_with_profile.iteritems():
yield self._handle_new_user(room_id, user_id, profile)
else:
users = yield self.store.get_users_in_public_due_to_room(room_id)
for user_id in users:
yield self._handle_remove_user(room_id, user_id)
@defer.inlineCallbacks
def _handle_local_user(self, user_id):
"""Adds a new local roomless user into the user_directory_search table.
Used to populate up the user index when we have an
user_directory_search_all_users specified.
"""
logger.debug("Adding new local user to dir, %r", user_id)
profile = yield self.store.get_profileinfo(get_localpart_from_id(user_id))
row = yield self.store.get_user_in_directory(user_id)
if not row:
yield self.store.add_profiles_to_user_dir(None, {user_id: profile})
@defer.inlineCallbacks
def _handle_new_user(self, room_id, user_id, profile):
"""Called when we might need to add user to directory
Args:
room_id (str): room_id that user joined or started being public
user_id (str)
"""
logger.debug("Adding new user to dir, %r", user_id)
row = yield self.store.get_user_in_directory(user_id)
if not row:
yield self.store.add_profiles_to_user_dir(room_id, {user_id: profile})
is_public = yield self.store.is_room_world_readable_or_publicly_joinable(
room_id
)
if is_public:
row = yield self.store.get_user_in_public_room(user_id)
if not row:
yield self.store.add_users_to_public_room(room_id, [user_id])
else:
logger.debug("Not adding new user to public dir, %r", user_id)
# Now we update users who share rooms with users. We do this by getting
# all the current users in the room and seeing which aren't already
# marked in the database as sharing with `user_id`
users_with_profile = yield self.state.get_current_user_in_room(room_id)
to_insert = set()
to_update = set()
is_appservice = self.store.get_if_app_services_interested_in_user(user_id)
# First, if they're our user then we need to update for every user
if self.is_mine_id(user_id) and not is_appservice:
# Returns a map of other_user_id -> shared_private. We only need
# to update mappings if for users that either don't share a room
# already (aren't in the map) or, if the room is private, those that
# only share a public room.
user_ids_shared = yield self.store.get_users_who_share_room_from_dir(
user_id
)
for other_user_id in users_with_profile:
if user_id == other_user_id:
continue
shared_is_private = user_ids_shared.get(other_user_id)
if shared_is_private is True:
# We've already marked in the database they share a private room
continue
elif shared_is_private is False:
# They already share a public room, so only update if this is
# a private room
if not is_public:
to_update.add((user_id, other_user_id))
elif shared_is_private is None:
# This is the first time they both share a room
to_insert.add((user_id, other_user_id))
# Next we need to update for every local user in the room
for other_user_id in users_with_profile:
if user_id == other_user_id:
continue
is_appservice = self.store.get_if_app_services_interested_in_user(
other_user_id
)
if self.is_mine_id(other_user_id) and not is_appservice:
shared_is_private = yield self.store.get_if_users_share_a_room(
other_user_id, user_id,
)
if shared_is_private is True:
# We've already marked in the database they share a private room
continue
elif shared_is_private is False:
# They already share a public room, so only update if this is
# a private room
if not is_public:
to_update.add((other_user_id, user_id))
elif shared_is_private is None:
# This is the first time they both share a room
to_insert.add((other_user_id, user_id))
if to_insert:
yield self.store.add_users_who_share_room(
room_id, not is_public, to_insert,
)
if to_update:
yield self.store.update_users_who_share_room(
room_id, not is_public, to_update,
)
@defer.inlineCallbacks
def _handle_remove_user(self, room_id, user_id):
"""Called when we might need to remove user to directory
Args:
room_id (str): room_id that user left or stopped being public that
user_id (str)
"""
logger.debug("Maybe removing user %r", user_id)
row = yield self.store.get_user_in_directory(user_id)
update_user_dir = row and row["room_id"] == room_id
row = yield self.store.get_user_in_public_room(user_id)
update_user_in_public = row and row["room_id"] == room_id
if (update_user_in_public or update_user_dir):
# XXX: Make this faster?
rooms = yield self.store.get_rooms_for_user(user_id)
for j_room_id in rooms:
if (not update_user_in_public and not update_user_dir):
break
is_in_room = yield self.store.is_host_joined(
j_room_id, self.server_name,
)
if not is_in_room:
continue
if update_user_dir:
update_user_dir = False
yield self.store.update_user_in_user_dir(user_id, j_room_id)
is_public = yield self.store.is_room_world_readable_or_publicly_joinable(
j_room_id
)
if update_user_in_public and is_public:
yield self.store.update_user_in_public_user_list(user_id, j_room_id)
update_user_in_public = False
if update_user_dir:
yield self.store.remove_from_user_dir(user_id)
elif update_user_in_public:
yield self.store.remove_from_user_in_public_room(user_id)
# Now handle users_who_share_rooms.
# Get a list of user tuples that were in the DB due to this room and
# users (this includes tuples where the other user matches `user_id`)
user_tuples = yield self.store.get_users_in_share_dir_with_room_id(
user_id, room_id,
)
for user_id, other_user_id in user_tuples:
# For each user tuple get a list of rooms that they still share,
# trying to find a private room, and update the entry in the DB
rooms = yield self.store.get_rooms_in_common_for_users(user_id, other_user_id)
# If they dont share a room anymore, remove the mapping
if not rooms:
yield self.store.remove_user_who_share_room(
user_id, other_user_id,
)
continue
found_public_share = None
for j_room_id in rooms:
is_public = yield self.store.is_room_world_readable_or_publicly_joinable(
j_room_id
)
if is_public:
found_public_share = j_room_id
else:
found_public_share = None
yield self.store.update_users_who_share_room(
room_id, not is_public, [(user_id, other_user_id)],
)
break
if found_public_share:
yield self.store.update_users_who_share_room(
room_id, not is_public, [(user_id, other_user_id)],
)
@defer.inlineCallbacks
def _handle_profile_change(self, user_id, room_id, prev_event_id, event_id):
"""Check member event changes for any profile changes and update the
database if there are.
"""
if not prev_event_id or not event_id:
return
prev_event = yield self.store.get_event(prev_event_id, allow_none=True)
event = yield self.store.get_event(event_id, allow_none=True)
if not prev_event or not event:
return
if event.membership != Membership.JOIN:
return
prev_name = prev_event.content.get("displayname")
new_name = event.content.get("displayname")
prev_avatar = prev_event.content.get("avatar_url")
new_avatar = event.content.get("avatar_url")
if prev_name != new_name or prev_avatar != new_avatar:
yield self.store.update_profile_in_user_dir(
user_id, new_name, new_avatar, room_id,
)
@defer.inlineCallbacks
def _get_key_change(self, prev_event_id, event_id, key_name, public_value):
"""Given two events check if the `key_name` field in content changed
from not matching `public_value` to doing so.
For example, check if `history_visibility` (`key_name`) changed from
`shared` to `world_readable` (`public_value`).
Returns:
None if the field in the events either both match `public_value`
or if neither do, i.e. there has been no change.
True if it didnt match `public_value` but now does
False if it did match `public_value` but now doesn't
"""
prev_event = None
event = None
if prev_event_id:
prev_event = yield self.store.get_event(prev_event_id, allow_none=True)
if event_id:
event = yield self.store.get_event(event_id, allow_none=True)
if not event and not prev_event:
logger.debug("Neither event exists: %r %r", prev_event_id, event_id)
defer.returnValue(None)
prev_value = None
value = None
if prev_event:
prev_value = prev_event.content.get(key_name)
if event:
value = event.content.get(key_name)
logger.debug("prev_value: %r -> value: %r", prev_value, value)
if value == public_value and prev_value != public_value:
defer.returnValue(True)
elif value != public_value and prev_value == public_value:
defer.returnValue(False)
else:
defer.returnValue(None)
+55
View File
@@ -0,0 +1,55 @@
# -*- coding: utf-8 -*-
# Copyright 2017 New Vector Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from synapse.http.server import wrap_request_handler
from twisted.web.resource import Resource
from twisted.web.server import NOT_DONE_YET
class AdditionalResource(Resource):
"""Resource wrapper for additional_resources
If the user has configured additional_resources, we need to wrap the
handler class with a Resource so that we can map it into the resource tree.
This class is also where we wrap the request handler with logging, metrics,
and exception handling.
"""
def __init__(self, hs, handler):
"""Initialise AdditionalResource
The ``handler`` should return a deferred which completes when it has
done handling the request. It should write a response with
``request.write()``, and call ``request.finish()``.
Args:
hs (synapse.server.HomeServer): homeserver
handler ((twisted.web.server.Request) -> twisted.internet.defer.Deferred):
function to be called to handle the request.
"""
Resource.__init__(self)
self._handler = handler
# these are required by the request_handler wrapper
self.version_string = hs.version_string
self.clock = hs.get_clock()
def render(self, request):
self._async_render(request)
return NOT_DONE_YET
@wrap_request_handler
def _async_render(self, request):
return self._handler(request)
+89 -38
View File
@@ -18,7 +18,7 @@ from OpenSSL.SSL import VERIFY_NONE
from synapse.api.errors import (
CodeMessageException, MatrixCodeMessageException, SynapseError, Codes,
)
from synapse.util.logcontext import preserve_context_over_fn
from synapse.util.logcontext import make_deferred_yieldable
from synapse.util import logcontext
import synapse.metrics
from synapse.http.endpoint import SpiderEndpoint
@@ -114,43 +114,73 @@ class SimpleHttpClient(object):
raise e
@defer.inlineCallbacks
def post_urlencoded_get_json(self, uri, args={}):
def post_urlencoded_get_json(self, uri, args={}, headers=None):
"""
Args:
uri (str):
args (dict[str, str|List[str]]): query params
headers (dict[str, List[str]]|None): If not None, a map from
header name to a list of values for that header
Returns:
Deferred[object]: parsed json
"""
# TODO: Do we ever want to log message contents?
logger.debug("post_urlencoded_get_json args: %s", args)
query_bytes = urllib.urlencode(encode_urlencode_args(args), True)
actual_headers = {
b"Content-Type": [b"application/x-www-form-urlencoded"],
b"User-Agent": [self.user_agent],
}
if headers:
actual_headers.update(headers)
response = yield self.request(
"POST",
uri.encode("ascii"),
headers=Headers({
b"Content-Type": [b"application/x-www-form-urlencoded"],
b"User-Agent": [self.user_agent],
}),
headers=Headers(actual_headers),
bodyProducer=FileBodyProducer(StringIO(query_bytes))
)
body = yield preserve_context_over_fn(readBody, response)
body = yield make_deferred_yieldable(readBody(response))
defer.returnValue(json.loads(body))
@defer.inlineCallbacks
def post_json_get_json(self, uri, post_json):
def post_json_get_json(self, uri, post_json, headers=None):
"""
Args:
uri (str):
post_json (object):
headers (dict[str, List[str]]|None): If not None, a map from
header name to a list of values for that header
Returns:
Deferred[object]: parsed json
"""
json_str = encode_canonical_json(post_json)
logger.debug("HTTP POST %s -> %s", json_str, uri)
actual_headers = {
b"Content-Type": [b"application/json"],
b"User-Agent": [self.user_agent],
}
if headers:
actual_headers.update(headers)
response = yield self.request(
"POST",
uri.encode("ascii"),
headers=Headers({
b"Content-Type": [b"application/json"],
b"User-Agent": [self.user_agent],
}),
headers=Headers(actual_headers),
bodyProducer=FileBodyProducer(StringIO(json_str))
)
body = yield preserve_context_over_fn(readBody, response)
body = yield make_deferred_yieldable(readBody(response))
if 200 <= response.code < 300:
defer.returnValue(json.loads(body))
@@ -160,7 +190,7 @@ class SimpleHttpClient(object):
defer.returnValue(json.loads(body))
@defer.inlineCallbacks
def get_json(self, uri, args={}):
def get_json(self, uri, args={}, headers=None):
""" Gets some json from the given URI.
Args:
@@ -169,6 +199,8 @@ class SimpleHttpClient(object):
None.
**Note**: The value of each key is assumed to be an iterable
and *not* a string.
headers (dict[str, List[str]]|None): If not None, a map from
header name to a list of values for that header
Returns:
Deferred: Succeeds when we get *any* 2xx HTTP response, with the
HTTP body as JSON.
@@ -177,13 +209,13 @@ class SimpleHttpClient(object):
error message.
"""
try:
body = yield self.get_raw(uri, args)
body = yield self.get_raw(uri, args, headers=headers)
defer.returnValue(json.loads(body))
except CodeMessageException as e:
raise self._exceptionFromFailedRequest(e.code, e.msg)
@defer.inlineCallbacks
def put_json(self, uri, json_body, args={}):
def put_json(self, uri, json_body, args={}, headers=None):
""" Puts some json to the given URI.
Args:
@@ -193,6 +225,8 @@ class SimpleHttpClient(object):
None.
**Note**: The value of each key is assumed to be an iterable
and *not* a string.
headers (dict[str, List[str]]|None): If not None, a map from
header name to a list of values for that header
Returns:
Deferred: Succeeds when we get *any* 2xx HTTP response, with the
HTTP body as JSON.
@@ -205,17 +239,21 @@ class SimpleHttpClient(object):
json_str = encode_canonical_json(json_body)
actual_headers = {
b"Content-Type": [b"application/json"],
b"User-Agent": [self.user_agent],
}
if headers:
actual_headers.update(headers)
response = yield self.request(
"PUT",
uri.encode("ascii"),
headers=Headers({
b"User-Agent": [self.user_agent],
"Content-Type": ["application/json"]
}),
headers=Headers(actual_headers),
bodyProducer=FileBodyProducer(StringIO(json_str))
)
body = yield preserve_context_over_fn(readBody, response)
body = yield make_deferred_yieldable(readBody(response))
if 200 <= response.code < 300:
defer.returnValue(json.loads(body))
@@ -226,7 +264,7 @@ class SimpleHttpClient(object):
raise CodeMessageException(response.code, body)
@defer.inlineCallbacks
def get_raw(self, uri, args={}):
def get_raw(self, uri, args={}, headers=None):
""" Gets raw text from the given URI.
Args:
@@ -235,6 +273,8 @@ class SimpleHttpClient(object):
None.
**Note**: The value of each key is assumed to be an iterable
and *not* a string.
headers (dict[str, List[str]]|None): If not None, a map from
header name to a list of values for that header
Returns:
Deferred: Succeeds when we get *any* 2xx HTTP response, with the
HTTP body at text.
@@ -246,15 +286,19 @@ class SimpleHttpClient(object):
query_bytes = urllib.urlencode(args, True)
uri = "%s?%s" % (uri, query_bytes)
actual_headers = {
b"User-Agent": [self.user_agent],
}
if headers:
actual_headers.update(headers)
response = yield self.request(
"GET",
uri.encode("ascii"),
headers=Headers({
b"User-Agent": [self.user_agent],
})
headers=Headers(actual_headers),
)
body = yield preserve_context_over_fn(readBody, response)
body = yield make_deferred_yieldable(readBody(response))
if 200 <= response.code < 300:
defer.returnValue(body)
@@ -274,27 +318,33 @@ class SimpleHttpClient(object):
# The two should be factored out.
@defer.inlineCallbacks
def get_file(self, url, output_stream, max_size=None):
def get_file(self, url, output_stream, max_size=None, headers=None):
"""GETs a file from a given URL
Args:
url (str): The URL to GET
output_stream (file): File to write the response body to.
headers (dict[str, List[str]]|None): If not None, a map from
header name to a list of values for that header
Returns:
A (int,dict,string,int) tuple of the file length, dict of the response
headers, absolute URI of the response and HTTP response code.
"""
actual_headers = {
b"User-Agent": [self.user_agent],
}
if headers:
actual_headers.update(headers)
response = yield self.request(
"GET",
url.encode("ascii"),
headers=Headers({
b"User-Agent": [self.user_agent],
})
headers=Headers(actual_headers),
)
headers = dict(response.headers.getAllRawHeaders())
resp_headers = dict(response.headers.getAllRawHeaders())
if 'Content-Length' in headers and headers['Content-Length'] > max_size:
if 'Content-Length' in resp_headers and resp_headers['Content-Length'] > max_size:
logger.warn("Requested URL is too large > %r bytes" % (self.max_size,))
raise SynapseError(
502,
@@ -315,10 +365,9 @@ class SimpleHttpClient(object):
# straight back in again
try:
length = yield preserve_context_over_fn(
_readBodyToFile,
response, output_stream, max_size
)
length = yield make_deferred_yieldable(_readBodyToFile(
response, output_stream, max_size,
))
except Exception as e:
logger.exception("Failed to download body")
raise SynapseError(
@@ -327,7 +376,9 @@ class SimpleHttpClient(object):
Codes.UNKNOWN,
)
defer.returnValue((length, headers, response.request.absoluteURI, response.code))
defer.returnValue(
(length, resp_headers, response.request.absoluteURI, response.code),
)
# XXX: FIXME: This is horribly copy-pasted from matrixfederationclient.
@@ -395,7 +446,7 @@ class CaptchaServerHttpClient(SimpleHttpClient):
)
try:
body = yield preserve_context_over_fn(readBody, response)
body = yield make_deferred_yieldable(readBody(response))
defer.returnValue(body)
except PartialDownloadError as e:
# twisted dislikes google's response, no content length.
+110 -20
View File
@@ -12,6 +12,7 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import socket
from twisted.internet.endpoints import HostnameEndpoint, wrapClientTLS
from twisted.internet import defer, reactor
@@ -30,7 +31,10 @@ logger = logging.getLogger(__name__)
SERVER_CACHE = {}
# our record of an individual server which can be tried to reach a destination.
#
# "host" is actually a dotted-quad or ipv6 address string. Except when there's
# no SRV record, in which case it is the original hostname.
_Server = collections.namedtuple(
"_Server", "priority weight host port expires"
)
@@ -219,9 +223,10 @@ class SRVClientEndpoint(object):
return self.default_server
else:
raise ConnectError(
"Not server available for %s" % self.service_name
"No server available for %s" % self.service_name
)
# look for all servers with the same priority
min_priority = self.servers[0].priority
weight_indexes = list(
(index, server.weight + 1)
@@ -231,11 +236,22 @@ class SRVClientEndpoint(object):
total_weight = sum(weight for index, weight in weight_indexes)
target_weight = random.randint(0, total_weight)
for index, weight in weight_indexes:
target_weight -= weight
if target_weight <= 0:
server = self.servers[index]
# XXX: this looks totally dubious:
#
# (a) we never reuse a server until we have been through
# all of the servers at the same priority, so if the
# weights are A: 100, B:1, we always do ABABAB instead of
# AAAA...AAAB (approximately).
#
# (b) After using all the servers at the lowest priority,
# we move onto the next priority. We should only use the
# second priority if servers at the top priority are
# unreachable.
#
del self.servers[index]
self.used_servers.append(server)
return server
@@ -280,26 +296,21 @@ def resolve_service(service_name, dns_client=client, cache=SERVER_CACHE, clock=t
continue
payload = answer.payload
host = str(payload.target)
srv_ttl = answer.ttl
try:
answers, _, _ = yield dns_client.lookupAddress(host)
except DNSNameError:
continue
hosts = yield _get_hosts_for_srv_record(
dns_client, str(payload.target)
)
for answer in answers:
if answer.type == dns.A and answer.payload:
ip = answer.payload.dottedQuad()
host_ttl = min(srv_ttl, answer.ttl)
for (ip, ttl) in hosts:
host_ttl = min(answer.ttl, ttl)
servers.append(_Server(
host=ip,
port=int(payload.port),
priority=int(payload.priority),
weight=int(payload.weight),
expires=int(clock.time()) + host_ttl,
))
servers.append(_Server(
host=ip,
port=int(payload.port),
priority=int(payload.priority),
weight=int(payload.weight),
expires=int(clock.time()) + host_ttl,
))
servers.sort()
cache[service_name] = list(servers)
@@ -317,3 +328,82 @@ def resolve_service(service_name, dns_client=client, cache=SERVER_CACHE, clock=t
raise e
defer.returnValue(servers)
@defer.inlineCallbacks
def _get_hosts_for_srv_record(dns_client, host):
"""Look up each of the hosts in a SRV record
Args:
dns_client (twisted.names.dns.IResolver):
host (basestring): host to look up
Returns:
Deferred[list[(str, int)]]: a list of (host, ttl) pairs
"""
ip4_servers = []
ip6_servers = []
def cb(res):
# lookupAddress and lookupIP6Address return a three-tuple
# giving the answer, authority, and additional sections of the
# response.
#
# we only care about the answers.
return res[0]
def eb(res, record_type):
if res.check(DNSNameError):
return []
logger.warn("Error looking up %s for %s: %s",
record_type, host, res, res.value)
return res
# no logcontexts here, so we can safely fire these off and gatherResults
d1 = dns_client.lookupAddress(host).addCallbacks(
cb, eb, errbackArgs=("A", ))
d2 = dns_client.lookupIPV6Address(host).addCallbacks(
cb, eb, errbackArgs=("AAAA", ))
results = yield defer.DeferredList(
[d1, d2], consumeErrors=True)
# if all of the lookups failed, raise an exception rather than blowing out
# the cache with an empty result.
if results and all(s == defer.FAILURE for (s, _) in results):
defer.returnValue(results[0][1])
for (success, result) in results:
if success == defer.FAILURE:
continue
for answer in result:
if not answer.payload:
continue
try:
if answer.type == dns.A:
ip = answer.payload.dottedQuad()
ip4_servers.append((ip, answer.ttl))
elif answer.type == dns.AAAA:
ip = socket.inet_ntop(
socket.AF_INET6, answer.payload.address,
)
ip6_servers.append((ip, answer.ttl))
else:
# the most likely candidate here is a CNAME record.
# rfc2782 says srvs may not point to aliases.
logger.warn(
"Ignoring unexpected DNS record type %s for %s",
answer.type, host,
)
continue
except Exception as e:
logger.warn("Ignoring invalid DNS response for %s: %s",
host, e)
continue
# keep the ipv4 results before the ipv6 results, mostly to match historical
# behaviour.
defer.returnValue(ip4_servers + ip6_servers)

Some files were not shown because too many files have changed in this diff Show More