When backfilling, `_get_state_ids_after_missing_prev_event` calls [`get_metadata_for_events`](26bc26586b/synapse/handlers/federation_event.py (L1133)). For `#matrix:matrix.org`, it's called with 77k `state_events` which means 77 calls to the database and takes 28 seconds.
This is a re-do of 57d334a13d (#13365),
which was backed out in 12abd72497 (#13501).
The `room_id` field represented the parent space for each room
and was made redundant by changes in the API shape where the
`children_state` is now nested underneath each `room`.
The room ID of each child is in the `state_key` field and is still
available.
This avoids doing work that will never be used (since the
resulting unread counts will never be sent in a /sync
response).
The negative of doing this is that unread counts will be
incorrect when the feature is initially enabled.
* Add monthly active users documentation
* changelog
* Tidy up notes
* more tidyup
* Rewrite #1
* link back to mau docs
* fix links
* s/appservice|AS/application service
* further review
* a newline
* Remove bit about shadow banned users.
I think talking about them is confusing, and the current text doesn't imply they get any special treatment.
* Update docs/usage/administration/monthly_active_users.md
Co-authored-by: Patrick Cloke <clokep@users.noreply.github.com>
* Update docs/usage/administration/monthly_active_users.md
Co-authored-by: Patrick Cloke <clokep@users.noreply.github.com>
Co-authored-by: Brendan Abolivier <babolivier@matrix.org>
Co-authored-by: Patrick Cloke <clokep@users.noreply.github.com>
The method doesn't actually do any data fetching and the method that
does, `_get_joined_profile_from_event_id`, has its own cache.
Signed off by Nick @ Beeper (@Fizzadar).
We incorrectly didn't use the returned `Responder` if the client had
disconnected, which meant that the resource used by the Responder
wasn't correctly released.
In particular, this exhausted the thread pools so that *all* requests
timed out.
Media downloaded as part of a URL preview is normally deleted after two days.
However, while a background database migration is running, the process is
stopped. A long-running database migration can therefore cause the media
store to fill up with old preview files.
This logic was added in #2697 to make sure that we didn't try to run the expiry
without an index on `local_media_repository.created_ts`; the original logic that
needs that index was added in #2478 (in `get_url_cache_media_before`, as
amended by 93247a424a), and is still present.
Given that the background update was added before Synapse v1.0.0, just drop
this check and assume the index exists.
Optimize how we calculate `likely_domains` during backfill because I've seen this take 17s in production just to `get_current_state` which is used to `get_domains_from_state` (see case [*2. Loading tons of events* in the `/messages` investigation issue](https://github.com/matrix-org/synapse/issues/13356)).
There are 3 ways we currently calculate hosts that are in the room:
1. `get_current_state` -> `get_domains_from_state`
- Used in `backfill` to calculate `likely_domains` and `/timestamp_to_event` because it was cargo-culted from `backfill`
- This one is being eliminated in favor of `get_current_hosts_in_room` in this PR 🕳
1. `get_current_hosts_in_room`
- Used for other federation things like sending read receipts and typing indicators
1. `get_hosts_in_room_at_events`
- Used when pushing out events over federation to other servers in the `_process_event_queue_loop`
Fix https://github.com/matrix-org/synapse/issues/13626
Part of https://github.com/matrix-org/synapse/issues/13356
Mentioned in [internal doc](https://docs.google.com/document/d/1lvUoVfYUiy6UaHB6Rb4HicjaJAU40-APue9Q4vzuW3c/edit#bookmark=id.2tvwz3yhcafh)
### Query performance
#### Before
The query from `get_current_state` sucks just because we have to get all 80k events. And we see almost the exact same performance locally trying to get all of these events (16s vs 17s):
```
synapse=# SELECT type, state_key, event_id FROM current_state_events WHERE room_id = '!OGEhHVWSdvArJzumhm:matrix.org';
Time: 16035.612 ms (00:16.036)
synapse=# SELECT type, state_key, event_id FROM current_state_events WHERE room_id = '!OGEhHVWSdvArJzumhm:matrix.org';
Time: 4243.237 ms (00:04.243)
```
But what about `get_current_hosts_in_room`: When there is 8M rows in the `current_state_events` table, the previous query in `get_current_hosts_in_room` took 13s from complete freshness (when the events were first added). But takes 930ms after a Postgres restart or 390ms if running back to back to back.
```sh
$ psql synapse
synapse=# \timing on
synapse=# SELECT COUNT(DISTINCT substring(state_key FROM '@[^:]*:(.*)$'))
FROM current_state_events
WHERE
type = 'm.room.member'
AND membership = 'join'
AND room_id = '!OGEhHVWSdvArJzumhm:matrix.org';
count
-------
4130
(1 row)
Time: 13181.598 ms (00:13.182)
synapse=# SELECT COUNT(*) from current_state_events where room_id = '!OGEhHVWSdvArJzumhm:matrix.org';
count
-------
80814
synapse=# SELECT COUNT(*) from current_state_events;
count
---------
8162847
synapse=# SELECT pg_size_pretty( pg_total_relation_size('current_state_events') );
pg_size_pretty
----------------
4702 MB
```
#### After
I'm not sure how long it takes from complete freshness as I only really get that opportunity once (maybe restarting computer but that's cumbersome) and it's not really relevant to normal operating times. Maybe you get closer to the fresh times the more access variability there is so that Postgres caches aren't as exact. Update: The longest I've seen this run for is 6.4s and 4.5s after a computer restart.
After a Postgres restart, it takes 330ms and running back to back takes 260ms.
```sh
$ psql synapse
synapse=# \timing on
Timing is on.
synapse=# SELECT
substring(c.state_key FROM '@[^:]*:(.*)$') as host
FROM current_state_events c
/* Get the depth of the event from the events table */
INNER JOIN events AS e USING (event_id)
WHERE
c.type = 'm.room.member'
AND c.membership = 'join'
AND c.room_id = '!OGEhHVWSdvArJzumhm:matrix.org'
GROUP BY host
ORDER BY min(e.depth) ASC;
Time: 333.800 ms
```
#### Going further
To improve things further we could add a `limit` parameter to `get_current_hosts_in_room`. Realistically, we don't need 4k domains to choose from because there is no way we're going to query that many before we a) probably get an answer or b) we give up.
Another thing we can do is optimize the query to use a index skip scan:
- https://wiki.postgresql.org/wiki/Loose_indexscan
- Index Skip Scan, https://commitfest.postgresql.org/37/1741/
- https://www.timescale.com/blog/how-we-made-distinct-queries-up-to-8000x-faster-on-postgresql/
If things like the signing key file are missing, let's just try to generate
them on startup.
Again, this is useful for k8s-like deployments where we just want to generate
keys on the first run.
When loading current ids, sort by stream ID so that we don't want to overwrite the `current_position` of an instance to a lower stream ID than we're actually at ([discussion](https://github.com/matrix-org/synapse/pull/13585#discussion_r951795379)). Previously, it sorted alphabetically by instance name which can be `null` and throw errors but more importantly, accomplishes nothing.
Fixes the following startup error which is why I started looking into this area:
```
$ poetry run synapse_homeserver --config-path homeserver.yaml
****************************************************************
Error during initialisation:
'<' not supported between instances of 'NoneType' and 'str'
There may be more information in the logs.
****************************************************************
```
Somehow my database ended up looking like the following, notice the `instance_name` is `null` in the db, and we can't sort `NoneType` things. Another question is why do we see the `instance_name` as `null` sometimes instead of `master` in monolith mode?
```
$ psql synapse
synapse=# SELECT * FROM stream_positions;
stream_name | instance_name | stream_id
-----------------+---------------+-----------
account_data | master | 1242
events | master | 1787
to_device | master | 58
presence_stream | master | 485638
receipts | master | 341
backfill | master | -139106
(6 rows)
synapse=# SELECT instance_name, stream_id FROM receipts_linearized;
instance_name | stream_id
---------------+-----------
| 211
| 3
| 4
| 212
| 213
| 224
| 228
| 164
| 313
| 253
| 38
| 321
| 324
| 189
| 192
| 193
| 194
| 195
| 197
| 198
| 275
| 79
| 339
| 340
| 82
| 341
| 84
| 85
| 91
| 119
```
Use dedicated `get_local_users_in_room` to find local users when calculating `join_authorised_via_users_server` ("the authorising user for joining a restricted room") of a `/make_join` request.
Found while working on https://github.com/matrix-org/synapse/pull/13575#discussion_r953023755 but it's not related.
This speeds things up by ~2x.
The vast majority of the time is now spent in `LruCache` moving things around the linked lists.
We do this via two things:
1. Don't create a deferred per-key during bulk set operations in `DeferredCache`. Instead, only create them if a subsequent caller asks for the key.
2. Add a bulk lookup API to `DeferredCache` rather than use a loop.
Part of #13019
This changes all the permission-related methods to rely on the Requester instead of the UserID. This is a first step towards enabling scoped access tokens at some point, since I expect the Requester to have scope-related informations in it.
It also changes methods which figure out the user/device/appservice out of the access token to return a Requester instead of something else. This avoids having store-related objects in the methods signatures.
Use a state filter or accept partial state in a few places where we
request state, to avoid blocking.
To make lazy-loading `/sync`s work, we need to provide the memberships
of event senders, which are not guaranteed to be in the room state.
Instead we dig through auth events for memberships to present to
clients. The auth events of an event are guaranteed to contain a
passable membership event, otherwise the event would have been rejected.
Note that this only covers the common code paths encountered during
testing. There has been no exhaustive checking of all sync code paths.
Fixes#13146.
Signed-off-by: Sean Quah <seanq@matrix.org>
Broke by #13522
It looks like we have some rules in the DB with a priority class less
than 0 that don't override the base rules. Before these were just
dropped, but #13522 made that a hard error.
This improves load times for push rules:
| Version | Time per user | Time for 1k users |
| -------------------- | ------------- | ----------------- |
| Before | 138 µs | 138ms |
| Now (with custom) | 2.11 µs | 2.11ms |
| Now (without custom) | 49.7 ns | 0.05 ms |
This therefore has a large impact on send times for rooms
with large numbers of local users in the room.
This reverts commit f383b9b3ec. Other PRs
were seeing mypy failures that looked to be related to mypy-zope.
Confusingly, we didn't see this on #13521.
Revert this for now and investigate later.
* Clarifies comments.
* Fixes an erroneous comment (about return type) added in #13455
(ec24813220).
* Clarifies the name of a variable.
* Simplifies logic of pulling out the latest join for the requesting user.
Add some miscellaneous comments to document sync, especially around
`compute_state_delta`.
Signed-off-by: Sean Quah <seanq@matrix.org>
Co-authored-by: Richard van der Hoff <1389908+richvdh@users.noreply.github.com>
```py
@trace
@tag_args
async def get_oldest_event_ids_with_depth_in_room(...)
...
```
Before this PR, you would see a warning in the logs and the span was not exported:
```
2022-08-03 19:11:59,383 - synapse.logging.opentracing - 835 - ERROR - GET-0 - @trace may not have wrapped EventFederationWorkerStore.get_oldest_event_ids_with_depth_in_room correctly! The function is not async but returned a coroutine.
```
In state res v2, we apply two passes of iterative auth checks. The first
pass replays power events and events in their auth chains, but only
those belonging to the full conflicted set. The source code as written
suggests that we want only those belonging to the auth difference (which
is a smaller set of events).
At runtime we were doing the correct thing anyway, because the only
callsite of `_reverse_topological_power_sort` passes in the
`full_conflicted_set`. So this really is just a rename.
This adds support for the stable identifiers of MSC2285 while
continuing to support the unstable identifiers behind the configuration
flag. These will be removed in a future version.
Fix @tag_args being off-by-one (ahead)
Example:
```
argspec.args=[
'self',
'room_id'
]
args=(
<synapse.storage.databases.main.DataStore object at 0x10d0b8d00>,
'!HBehERstyQBxyJDLfR:my.synapse.server'
)
```
---
The previous logic was also flawed and we can end up in a situation like this:
```
argspec.args=['self', 'dest', 'room_id', 'limit', 'extremities']
args=(<synapse.federation.federation_client.FederationClient object at 0x7f1651c18160>, 'hs1', '!jAEHKIubyIfuLOdfpY:hs1')
```
From this source:
```py
async def backfill(
self, dest: str, room_id: str, limit: int, extremities: Collection[str]
) -> Optional[List[EventBase]]:
```
And this usage:
```py
events = await self._federation_client.backfill(
dest, room_id, limit=limit, extremities=extremities
)
```
which would previously cause this error:
```
synapse_main | 2022-08-04 06:13:12,051 - synapse.handlers.federation - 424 - ERROR - GET-5 - Failed to backfill from hs1 because tuple index out of range
synapse_main | Traceback (most recent call last):
synapse_main | File "/usr/local/lib/python3.9/site-packages/synapse/handlers/federation.py", line 392, in try_backfill
synapse_main | await self._federation_event_handler.backfill(
synapse_main | File "/usr/local/lib/python3.9/site-packages/synapse/logging/tracing.py", line 828, in _wrapper
synapse_main | return await func(*args, **kwargs)
synapse_main | File "/usr/local/lib/python3.9/site-packages/synapse/handlers/federation_event.py", line 593, in backfill
synapse_main | events = await self._federation_client.backfill(
synapse_main | File "/usr/local/lib/python3.9/site-packages/synapse/logging/tracing.py", line 828, in _wrapper
synapse_main | return await func(*args, **kwargs)
synapse_main | File "/usr/local/lib/python3.9/site-packages/synapse/logging/tracing.py", line 827, in _wrapper
synapse_main | with wrapping_logic(func, *args, **kwargs):
synapse_main | File "/usr/local/lib/python3.9/contextlib.py", line 119, in __enter__
synapse_main | return next(self.gen)
synapse_main | File "/usr/local/lib/python3.9/site-packages/synapse/logging/tracing.py", line 922, in _wrapping_logic
synapse_main | set_attribute("ARG_" + arg, str(args[i + 1])) # type: ignore[index]
synapse_main | IndexError: tuple index out of range
```
* Adds docstrings and inline comments.
* Formats SQL queries using triple quoted strings.
* Minor formatting changes.
* Avoid fetching `event_push_summary_stream_ordering` multiple times
in the same transactions.
Still maintains local in memory lookup optimisation, but does any external
lookup as part of the deferred that prevents duplicate lookups for the same
event at once. This makes the assumption that fetching from an external
cache is a non-zero load operation.
Part of my continuing quest to make the docker images build quicker: copy nginx and redis in from base docker images, rather than apt installing each time.
Signed-off-by: Andrew Doh <andrewddo@gmail.com>
Co-authored-by: Patrick Cloke <clokep@users.noreply.github.com>
Co-authored-by: Andrew Morgan <andrewm@element.io>
Co-authored-by: Brendan Abolivier <babolivier@matrix.org>
* Improved section regarding server admin
Added steps describing how to elevate an existing user to administrator by manipulating a `postgres` database.
Signed-off-by: jejo86 28619134+jejo86@users.noreply.github.com
* Improved section regarding server admin
* Reference database settings
Add instructions to check database settings to find out the database name, instead of listing all available PostgreSQL databases.
* Add suggestions from PR conversation
Replace config filename `homeserver.yaml`. with "config file".
Remove instructions to switch to `postgres` user.
Add instructions how to connect to SQLite database.
* Update changelog.d/13230.doc
Co-authored-by: reivilibre <olivier@librepush.net>
Previously, `_resolve_state_at_missing_prevs` returned the resolved
state before an event and a partial state flag. These were unwieldy to
carry around would only ever be used to build an event context. Build
the event context directly instead.
Signed-off-by: Sean Quah <seanq@matrix.org>
The `room_id` field represented the parent space for each room
and was made redundant by changes in the API shape where the
`children_state` is now nested underneath each `room`.
The room ID of each child is in the `state_key` field and is still
available.
Avoid blocking on full state in `_resolve_state_at_missing_prevs` and
return a new flag indicating whether the resolved state is partial.
Thread that flag around so that it makes it into the event context.
Co-authored-by: Richard van der Hoff <1389908+richvdh@users.noreply.github.com>
When registering a new account via SSO on iOS, the text field becomes pretty annoying as it autocapitalises and autocorrects your input. This PR fixes that (although I have only tested the raw HTML file on the simulator, I'm not sure how to get the complete setup available for testing in the flow).
Previously, TLS could only be used with STARTTLS.
Add a new option `force_tls`, where TLS is used from the start.
Implicit TLS is recommended over STARTLS,
see https://datatracker.ietf.org/doc/html/rfc8314Fixes#8046.
Signed-off-by: Jan Schär <jan@jschaer.ch>
See #10826 and #10786 for context as to why we had to disable pruning on
those caches.
Now that `get_users_who_share_room_with_user` is called frequently only
for presence, we just need to make calls to it less frequent and then we
can remove the various levels of caching that is going on.
When a room has the partial state flag, we may not have an accurate
`m.room.member` event for event senders in the room's current state, and
so cannot perform soft fail checks correctly. Skip the soft fail check
entirely in this case.
As an alternative, we could block until we have full state, but that
would prevent us from receiving incoming events over federation, which
is undesirable.
Signed-off-by: Sean Quah <seanq@matrix.org>
Add another bash script to the contrib directory. It creates multiple stream writers and also prints out the example configuration for homeserver.yaml.
Signed-off-by: Ville Petteri Huh.
Fix race conditions in the async cache invalidation logic, by separating
the async & local invalidation calls and ensuring any async call i
executed first.
Signed off by Nick @ Beeper (@Fizzadar).
More prep work for asyncronous caching, also makes all process_replication_rows methods consistent (presence handler already is so).
Signed off by Nick @ Beeper (@Fizzadar)
Fix https://github.com/matrix-org/synapse/issues/13016
## New error code and status
### Before
Previously, we returned a `404` for `/thumbnail` which isn't even in the spec.
```json
{
"errcode": "M_NOT_FOUND",
"error": "Not found [b'hs1', b'tefQeZhmVxoiBfuFQUKRzJxc']"
}
```
### After
What does the spec say?
> 400: The request does not make sense to the server, or the server cannot thumbnail the content. For example, the client requested non-integer dimensions or asked for negatively-sized images.
>
> *-- https://spec.matrix.org/v1.1/client-server-api/#get_matrixmediav3thumbnailservernamemediaid*
Now with this PR, we respond with a `400` when we don't have thumbnails to serve and we explain why we might not have any thumbnails.
```json
{
"errcode": "M_UNKNOWN",
"error": "Cannot find any thumbnails for the requested media ([b'example.com', b'12345']). This might mean the media is not a supported_media_format=(image/jpeg, image/jpg, image/webp, image/gif, image/png) or that thumbnailing failed for some other reason. (Dynamic thumbnails are disabled on this server.)",
}
```
> Cannot find any thumbnails for the requested media ([b'example.com', b'12345']). This might mean the media is not a supported_media_format=(image/jpeg, image/jpg, image/webp, image/gif, image/png) or that thumbnailing failed for some other reason. (Dynamic thumbnails are disabled on this server.)
---
We still respond with a 404 in many other places. But we can iterate on those later and maybe keep some in some specific places after spec updates/clarification: https://github.com/matrix-org/matrix-spec/issues/1122
We can also iterate on the bugs where Synapse doesn't thumbnail when it should in other issues/PRs.
`frozendict` 2.3.2 includes a fix for a memory leak in
`frozendict.__hash__`. This likely has no impact outside of the
deprecated `/initialSync` endpoint, which uses `StreamToken`s,
containing `RoomStreamToken`s, containing `frozendict`s, as cache keys.
Signed-off-by: Sean Quah <seanq@matrix.org>
These columns were added back in Synapse 1.52, and have been populated for new
events since then. It's now (beyond) time to back-populate them for existing
events.
There are two fixes here:
1. A long-standing bug where we incorrectly calculated `delta_ids`; and
2. A bug introduced in #13267 where we got current state incorrect.
When building the docker images for complement testing, copy a preinstalled
complement over from a base image, rather than apt installing it. This avoids
network traffic and is much faster.
Some experimental prep work to enable external event caching based on #9379 & #12955. Doesn't actually move the cache at all, just lays the groundwork for async implemented caches.
Signed off by Nick @ Beeper (@Fizzadar)
* Replace `get_new_events_for_appservice` with `get_all_new_events_stream`
The functions were near identical and this brings the AS worker closer
to the way federation senders work which can allow for multiple workers
to handle AS traffic.
* Pull received TS alongside events when processing the stream
This avoids an extra query -per event- when both federation sender
and appservice pusher process events.
There is a corner in `_check_event_auth` (long known as "the weird corner") where, if we get an event with auth_events which don't match those we were expecting, we attempt to resolve the diffence between our state and the remote's with a state resolution.
This isn't specced, and there's general agreement we shouldn't be doing it.
However, it turns out that the faster-joins code was relying on it, so we need to introduce something similar (but rather simpler) for that.
* Drop support for v1 unbind
Signed-off-by: Jacek Kusnierz <jacek.kusnierz@tum.de>
* Add changelog
Signed-off-by: Jacek Kusnierz <jacek.kusnierz@tum.de>
* Update changelog.d/13240.misc
* Admin API request explanation improved
Pointed out, that the Admin API is not accessible by default from any remote computer, but only from the PC `matrix-synapse` is running on.
Added a full, working example, making sure to include the cURL flag `-X`, which needs to be prepended to `GET`, `POST`, `PUT` etc. and listing the full query string including protocol, IP address and port.
* Admin API request explanation improved
* Apply suggestions from code review
Update changelog. Reword prose.
Co-authored-by: David Robertson <david.m.robertson1@gmail.com>
* Drop support for delegating email validation
Delegating email validation to an IS is insecure (since it allows the owner of
the IS to do a password reset on your HS), and has long been deprecated. It
will now cause a config error at startup.
* Update unit test which checks for email verification
Give it an `email` config instead of a threepid delegate
* Remove unused method `requestEmailToken`
* Simplify config handling for email verification
Rather than an enum and a boolean, all we need here is a single bool, which
says whether we are or are not doing email verification.
* update docs
* changelog
* upgrade.md: fix typo
* update version number
this will be in 1.64, not 1.63
* update version number
this one too
This gets rid of another usage of get_appservice_by_req, with all the benefits, including correctly tracking the appservice IP and setting the tracing attributes correctly.
Signed-off-by: Quentin Gliech <quenting@element.io>
