Commit Graph

627 Commits

Author SHA1 Message Date
Erik Johnston
b660327056
Resync remote device list when detected as stale. (#6786) 2020-01-30 17:06:38 +00:00
Erik Johnston
e17a110661
Detect unknown remote devices and mark cache as stale (#6776)
We just mark the fact that the cache may be stale in the database for
now.
2020-01-28 14:43:21 +00:00
Richard van der Hoff
a8ce7aeb43
Pass room version object into event_auth.check and check_redaction (#6788)
These are easier to work with than the strings and we normally have one around.

This fixes `FederationHander._persist_auth_tree` which was passing a
RoomVersion object into event_auth.check instead of a string.
2020-01-28 14:18:29 +00:00
Erik Johnston
8df862e45d
Add rooms.room_version column (#6729)
This is so that we don't have to rely on pulling it out from `current_state_events` table.
2020-01-27 14:30:57 +00:00
Erik Johnston
d386f2f339
Add StateMap type alias (#6715) 2020-01-16 13:31:22 +00:00
Brendan Abolivier
cd428a93e2
Fix conditions failing if min_depth = 0
This could result in Synapse not fetching prev_events for new events in the room if it has missed some events.
2020-01-07 12:08:58 +00:00
Richard van der Hoff
56ad230efa Merge branch 'master' into develop 2019-12-20 11:11:34 +00:00
Erik Johnston
fa780e9721
Change EventContext to use the Storage class (#6564) 2019-12-20 10:32:02 +00:00
Richard van der Hoff
03d3792f3c
Fix exceptions when attempting to backfill (#6576)
Fixes #6575
2019-12-20 09:55:45 +00:00
Richard van der Hoff
6e8f8e14f2 Merge release-v1.7.1 into develop 2019-12-18 09:51:51 +00:00
Richard van der Hoff
bbb75ff6ee Exclude rejected state events when calculating state at backwards extrems (#6527)
This fixes a weird bug where, if you were determined enough, you could end up with a rejected event forming part of the state at a backwards-extremity. Authing that backwards extrem would then lead to us trying to pull the rejected event from the db (with allow_rejected=False), which would fail with a 404.
2019-12-16 14:07:29 +00:00
Richard van der Hoff
ff773ff724 Persist auth/state events at backwards extremities when we fetch them (#6526)
The main point here is to make sure that the state returned by _get_state_in_room has been authed before we try to use it as state in the room.
2019-12-16 14:07:11 +00:00
Richard van der Hoff
83895316d4 sanity-checking for events used in state res (#6531)
When we perform state resolution, check that all of the events involved are in
the right room.
2019-12-16 13:53:53 +00:00
Richard van der Hoff
35bbe4ca79 Check the room_id of events when fetching room state/auth (#6524)
When we request the state/auth_events to populate a backwards extremity (on
backfill or in the case of missing events in a transaction push), we should
check that the returned events are in the right room rather than blindly using
them in the room state or auth chain.

Given that _get_events_from_store_or_dest takes a room_id, it seems clear that
it should be sanity-checking the room_id of the requested events, so let's do
it there.
2019-12-16 13:47:07 +00:00
Richard van der Hoff
20d5ba16e6 Add include_event_in_state to _get_state_for_room (#6521)
Make it return the state *after* the requested event, rather than the one
before it. This is a bit easier and requires fewer calls to
get_events_from_store_or_dest.
2019-12-16 13:26:12 +00:00
Richard van der Hoff
be294d6fde Move get_state methods into FederationHandler (#6503)
This is a non-functional refactor as a precursor to some other work.
2019-12-16 13:20:21 +00:00
Richard van der Hoff
6920d88892
Exclude rejected state events when calculating state at backwards extrems (#6527)
This fixes a weird bug where, if you were determined enough, you could end up with a rejected event forming part of the state at a backwards-extremity. Authing that backwards extrem would then lead to us trying to pull the rejected event from the db (with allow_rejected=False), which would fail with a 404.
2019-12-16 13:14:37 +00:00
Richard van der Hoff
bc7de87650
Persist auth/state events at backwards extremities when we fetch them (#6526)
The main point here is to make sure that the state returned by _get_state_in_room has been authed before we try to use it as state in the room.
2019-12-16 12:26:28 +00:00
Richard van der Hoff
1da15f05f5
sanity-checking for events used in state res (#6531)
When we perform state resolution, check that all of the events involved are in
the right room.
2019-12-13 12:55:32 +00:00
Richard van der Hoff
25f1244329
Check the room_id of events when fetching room state/auth (#6524)
When we request the state/auth_events to populate a backwards extremity (on
backfill or in the case of missing events in a transaction push), we should
check that the returned events are in the right room rather than blindly using
them in the room state or auth chain.

Given that _get_events_from_store_or_dest takes a room_id, it seems clear that
it should be sanity-checking the room_id of the requested events, so let's do
it there.
2019-12-12 12:57:45 +00:00
Richard van der Hoff
2045356517
Add include_event_in_state to _get_state_for_room (#6521)
Make it return the state *after* the requested event, rather than the one
before it. This is a bit easier and requires fewer calls to
get_events_from_store_or_dest.
2019-12-11 16:37:51 +00:00
Richard van der Hoff
6637d90d77 convert to async: FederationHandler._process_received_pdu
also fix user_joined_room to consistently return deferreds
2019-12-11 14:39:26 +00:00
Richard van der Hoff
4db394a4b3 convert to async: FederationHandler._get_state_for_room
... and _get_events_from_store_or_dest
2019-12-11 14:39:26 +00:00
Richard van der Hoff
e77237b935 convert to async: FederationHandler.on_receive_pdu
and associated functions:
 * on_receive_pdu
 * handle_queued_pdus
 * get_missing_events_for_pdu
2019-12-11 14:39:26 +00:00
Richard van der Hoff
7712e751b8 Convert federation backfill to async
PaginationHandler.get_messages is only called by RoomMessageListRestServlet,
which is async.

Chase the code path down from there:
 - FederationHandler.maybe_backfill (and nested try_backfill)
 - FederationHandler.backfill
2019-12-11 14:39:25 +00:00
Richard van der Hoff
7c429f92d6
Clean up some logging (#6515)
This just makes some of the logging easier to follow when things start going
wrong.
2019-12-11 14:32:25 +00:00
Andrew Morgan
fc316a4894
Prevent redacted events from appearing in message search (#6377) 2019-12-11 13:39:47 +00:00
Andrew Morgan
ea0f0ad414
Prevent message search in upgraded rooms we're not in (#6385) 2019-12-11 13:07:25 +00:00
Richard van der Hoff
f8bc2ae883
Move get_state methods into FederationHandler (#6503)
This is a non-functional refactor as a precursor to some other work.
2019-12-10 17:42:46 +00:00
Richard van der Hoff
63d6ad1064
Stronger typing in the federation handler (#6480)
replace the event_info dict with an attrs thing
2019-12-05 15:02:35 +00:00
Richard van der Hoff
e1f4c83f41
Sanity-check the rooms of auth events before pulling them in. (#6472) 2019-12-05 14:14:45 +00:00
Richard van der Hoff
e203874caa
get rid of (most of) have_events from _update_auth_events_and_context_for_auth (#6468)
have_events was a map from event_id to rejection reason (or None) for events
which are in our local database. It was used as filter on the list of
event_ids being passed into get_events_as_list. However, since
get_events_as_list will ignore any event_ids that are unknown or rejected, we
can equivalently just leave it to get_events_as_list to do the filtering.

That means that we don't have to keep `have_events` up-to-date, and can use
`have_seen_events` instead of `get_seen_events_with_rejection` in the one place
we do need it.
2019-12-04 17:27:32 +00:00
Brendan Abolivier
54dd5dc12b
Add ephemeral messages support (MSC2228) (#6409)
Implement part [MSC2228](https://github.com/matrix-org/matrix-doc/pull/2228). The parts that differ are:

* the feature is hidden behind a configuration flag (`enable_ephemeral_messages`)
* self-destruction doesn't happen for state events
* only implement support for the `m.self_destruct_after` field (not the `m.self_destruct` one)
* doesn't send synthetic redactions to clients because for this specific case we consider the clients to be able to destroy an event themselves, instead we just censor it (by pruning its JSON) in the database
2019-12-03 19:19:45 +00:00
Erik Johnston
2173785f0d Propagate reason in remotely rejected invites 2019-11-28 11:31:56 +00:00
Brendan Abolivier
d31f69afa0
Merge pull request #6358 from matrix-org/babolivier/message_retention
Implement message retention policies (MSC1763)
2019-11-27 15:04:38 +00:00
Brendan Abolivier
9e937c28ee Merge branch 'develop' into babolivier/message_retention 2019-11-26 17:53:57 +00:00
Richard van der Hoff
4d394d6415
remove confusing fixme 2019-11-26 12:32:37 +00:00
Brendan Abolivier
a6fc6754f8
Fix 3PID invite exchange 2019-11-19 14:07:39 +00:00
Richard van der Hoff
870c00e278 Merge remote-tracking branch 'origin/develop' into rav/event_auth/4 2019-11-18 12:05:36 +00:00
Andrew Morgan
bc29a19731 Replace instance variations of homeserver with correct case/spacing 2019-11-12 13:08:12 +00:00
Richard van der Hoff
f41027f746 Use get_events_as_list rather than lots of calls to get_event
It's more efficient and clearer.
2019-11-08 12:21:28 +00:00
Richard van der Hoff
f8407975e7 Update some docstrings and comments 2019-11-08 12:18:20 +00:00
Richard van der Hoff
772d414975 Simplify _update_auth_events_and_context_for_auth
move event_key calculation into _update_context_for_auth_events, since it's
only used there.
2019-11-08 11:40:11 +00:00
Richard van der Hoff
807ec3bd99
Fix bug which caused rejected events to be stored with the wrong room state (#6320)
Fixes a bug where rejected events were persisted with the wrong state group.

Also fixes an occasional internal-server-error when receiving events over
federation which are rejected and (possibly because they are
backwards-extremities) have no prev_group.

Fixes #6289.
2019-11-06 10:01:39 +00:00
Richard van der Hoff
0e3ab8afdc
Add some checks that we aren't using state from rejected events (#6330)
* Raise an exception if accessing state for rejected events

Add some sanity checks on accessing state_group etc for
rejected events.

* Skip calculating push actions for rejected events

It didn't actually cause any bugs, because rejected events get filtered out at
various later points, but there's not point in trying to calculate the push
actions for a rejected event.
2019-11-05 22:13:37 +00:00
Brendan Abolivier
09957ce0e4
Implement per-room message retention policies 2019-11-04 17:09:22 +00:00
Richard van der Hoff
c6516adbe0
Factor out an _AsyncEventContextImpl (#6298)
The intention here is to make it clearer which fields we can expect to be
populated when: notably, that the _event_type etc aren't used for the
synchronous impl of EventContext.
2019-11-01 16:19:09 +00:00
Andrew Morgan
ace947e8da
Depublish a room from the public rooms list when it is upgraded (#6232) 2019-11-01 10:28:09 +00:00
Amber Brown
020add5099
Update black to 19.10b0 (#6304)
* update version of black and also fix the mypy config being overridden
2019-11-01 02:43:24 +11:00
Erik Johnston
dfe0cd71b6
Merge pull request #6294 from matrix-org/erikj/add_state_storage
Add StateGroupStorage interface
2019-10-31 16:17:53 +01:00
Andrew Morgan
54fef094b3
Remove usage of deprecated logger.warn method from codebase (#6271)
Replace every instance of `logger.warn` with `logger.warning` as the former is deprecated.
2019-10-31 10:23:24 +00:00
Erik Johnston
69f0054ce6 Port to use state storage 2019-10-30 14:46:54 +00:00
Erik Johnston
ec6de1cc7d Merge branch 'develop' of github.com:matrix-org/synapse into erikj/split_out_persistence_store 2019-10-30 13:37:04 +00:00
Andrew Morgan
a2276d4d3c
Fix log line that was printing undefined value (#6278) 2019-10-30 11:28:48 +00:00
Richard van der Hoff
172f264ed3
Improve signature checking on some federation APIs (#6262)
Make sure that we check that events sent over /send_join, /send_leave, and
/invite, are correctly signed and come from the expected servers.
2019-10-28 12:43:23 +00:00
Erik Johnston
3ca4c7c516 Use new EventPersistenceStore 2019-10-23 16:15:03 +01:00
Richard van der Hoff
93eaeec75a
Remove Auth.check method (#6217)
This method was somewhat redundant, and confusing.
2019-10-18 19:43:36 +02:00
Richard van der Hoff
1ba359a11f rip out some unreachable code
The only possible rejection reason is AUTH_ERROR, so all of this is unreachable.
2019-10-17 21:45:59 +01:00
Brendan Abolivier
21d51ab598
Typo 2019-10-04 11:21:24 +01:00
Brendan Abolivier
4676732ca0
Lint 2019-10-04 11:18:28 +01:00
Brendan Abolivier
81d51ce48b
Incorporate review 2019-10-04 11:16:19 +01:00
Brendan Abolivier
6527fa18c1 Add test case 2019-10-03 11:24:36 +01:00
Brendan Abolivier
972c9f65d7
Lint 2019-10-02 12:17:46 +01:00
Brendan Abolivier
5705ecaec6
Don't 500 code when trying to exchange a revoked 3PID invite
While this is not documented in the spec (but should be), Riot (and other clients) revoke 3PID invites by sending a m.room.third_party_invite event with an empty ({}) content to the room's state.
When the invited 3PID gets associated with a MXID, the identity server (which doesn't know about revocations) sends down to the MXID's homeserver all of the undelivered invites it has for this 3PID. The homeserver then tries to talk to the inviting homeserver in order to exchange these invite for m.room.member events.
When one of the invite is revoked, the inviting homeserver responds with a 500 error because it tries to extract a 'display_name' property from the content, which is empty. This might cause the invited server to consider that the server is down and not try to exchange other, valid invites (or at least delay it).

This fix handles the case of revoked invites by avoiding trying to fetch a 'display_name' from the original invite's content, and letting the m.room.member event fail the auth rules (because, since the original invite's content is empty, it doesn't have public keys), which results in sending a 403 with the correct error message to the invited server.
2019-10-02 11:19:43 +01:00
Andrew Morgan
cd17a2085e
Remove origin parameter from add_display_name_to_third_party_invite and add params to docstring (#6010)
Another small fixup noticed during work on a larger PR. The `origin` field of `add_display_name_to_third_party_invite` is not used and likely was just carried over from the `on_PUT` method of `FederationThirdPartyInviteExchangeServlet` which, like all other servlets, provides an `origin` argument.

Since it's not used anywhere in the handler function though, we should remove it from the function arguments.
2019-09-11 10:37:17 +01:00
Andrew Morgan
4548d1f87e
Remove unnecessary parentheses around return statements (#5931)
Python will return a tuple whether there are parentheses around the returned values or not.

I'm just sick of my editor complaining about this all over the place :)
2019-08-30 16:28:26 +01:00
Amber Brown
7dc398586c
Implement a structured logging output system. (#5680) 2019-08-28 21:18:53 +10:00
Erik Johnston
0f632f3a57
Merge pull request #5790 from matrix-org/erikj/groups_request_errors
Handle RequestSendFailed exception correctly in more places.
2019-08-01 13:14:08 +01:00
Erik Johnston
b4d5ff0af7 Don't log as exception when failing durig backfill 2019-07-30 13:19:22 +01:00
Amber Brown
865077f1d1
Room Complexity Client Implementation (#5783) 2019-07-30 02:47:27 +10:00
Richard van der Hoff
1a93daf353
Merge pull request #5744 from matrix-org/erikj/log_leave_origin_mismatch
Log when we receive a /make_* request from a different origin
2019-07-26 12:38:37 +01:00
Richard van der Hoff
d1020653fc Log when we receive a /make_* request from a different origin 2019-07-26 10:08:22 +01:00
Amber Brown
4806651744
Replace returnValue with return (#5736) 2019-07-23 23:00:55 +10:00
Amber Brown
463b072b12
Move logging utilities out of the side drawer of util/ and into logging/ (#5606) 2019-07-04 00:07:04 +10:00
Amber Brown
32e7c9e7f2
Run Black. (#5482) 2019-06-20 19:32:02 +10:00
Erik Johnston
7dcf984075
Merge pull request #5042 from matrix-org/erikj/fix_get_missing_events_error
Handle the case of `get_missing_events` failing
2019-06-19 13:20:09 +01:00
David Baker
f2d2ae03da Add some logging to 3pid invite sig verification (#5015)
I had to add quite a lot of logging to diagnose a problem with 3pid
invites - we only logged the one failure which isn't all that
informative.

NB. I'm not convinced the logic of this loop is right: I think it
should just accept a single valid signature from a trusted source
rather than fail if *any* signature is invalid. Also it should
probably not skip the rest of middle loop if a check fails? However,
I'm deliberately not changing the logic here.
2019-06-18 22:51:24 +01:00
Erik Johnston
2b20d0fb59 Fix logline 2019-06-18 16:12:53 +01:00
Erik Johnston
19b80fe68a Merge branch 'develop' of github.com:matrix-org/synapse into erikj/fix_get_missing_events_error 2019-06-18 16:11:43 +01:00
Erik Johnston
a9dab970b8
Merge pull request #5464 from matrix-org/erikj/3pid_remote_invite_state
Fix 3PID invite room state over federation.
2019-06-17 10:18:28 +01:00
Brendan Abolivier
f874b16b2e Add plugin APIs for implementations of custom event rules. 2019-06-14 18:16:03 +01:00
Erik Johnston
304a1376c2 Fix 3PID invite room state over federation.
Fixes that when a user exchanges a 3PID invite for a proper invite over
federation it does not include the `invite_room_state` key.

This was due to synapse incorrectly sending out two invite requests.
2019-06-14 16:19:11 +01:00
Erik Johnston
bc3d6b918b Add logging when request fails and clarify we ignore errors. 2019-06-05 11:37:51 +01:00
Erik Johnston
220a733d73 Fix handling of failures when calling /event_auth.
When processing an incoming event over federation, we may try and
resolve any unexpected differences in auth events. This is a
non-essential process and so should not stop the processing of the event
if it fails (e.g. due to the remote disappearing or not implementing the
necessary endpoints).

Fixes #3330
2019-06-03 09:56:45 +01:00
Richard van der Hoff
85d1e03b9d
Simplifications and comments in do_auth (#5227)
I was staring at this function trying to figure out wtf it was actually
doing. This is (hopefully) a non-functional refactor which makes it a bit
clearer.
2019-05-23 11:17:42 +01:00
Richard van der Hoff
c4aef549ad
Exclude soft-failed events from fwd-extremity candidates. (#5146)
When considering the candidates to be forward-extremities, we must exclude soft
failures.

Hopefully fixes #5090.
2019-05-21 16:10:54 +01:00
Erik Johnston
c132c8e505 Handle the case of get_missing_events failing
Currently if a call to `/get_missing_events` fails we log an exception
and stop processing the top level event we received over federation.
Instead let's try and handle it sensibly given it is a somewhat expected
failure mode.
2019-04-10 10:39:54 +01:00
Richard van der Hoff
54a87a7b08
Collect room-version variations into one place (#4969)
Collect all the things that make room-versions different to one another into
one place, so that it's easier to define new room versions.
2019-04-01 10:24:38 +01:00
Erik Johnston
0ff8163eae Factor out soft fail checks 2019-03-08 11:26:33 +00:00
Erik Johnston
a9de04be72 Implement soft fail 2019-03-06 16:22:16 +00:00
Erik Johnston
aa06d26ae0 clarify comments 2019-03-05 09:16:35 +00:00
Erik Johnston
d1523aed6b Only check history visibility when filtering
When filtering events to send to server we check more than just history
visibility. However when deciding whether to backfill or not we only
care about the history visibility.
2019-03-04 14:43:42 +00:00
Erik Johnston
8b63fe4c26 s/get_forward_events/get_successor_events/ 2019-03-04 11:56:03 +00:00
Erik Johnston
fbc047f2a5 Merge branch 'develop' of github.com:matrix-org/synapse into erikj/stop_fed_not_in_room 2019-03-04 11:54:58 +00:00
Erik Johnston
b183fef9ac Update comments 2019-02-27 13:06:10 +00:00
Erik Johnston
9c598dddcb Fix typo 2019-02-25 16:32:02 +00:00
Erik Johnston
d730c2c22b More comments 2019-02-25 14:45:02 +00:00
Erik Johnston
890cb048fd Assert rather than clobber the values 2019-02-25 14:42:11 +00:00
Erik Johnston
9342cc6ab1 Add comments and paranoia 2019-02-25 10:02:12 +00:00
Erik Johnston
7b288826b7 Fix backfill storing incorrect state for events 2019-02-22 11:33:51 +00:00