This is because if a worker reaches ~100% CPU then everything starts
lagging and we hit the log line a lot. When at error we invoke sentry
and that has a lot of overhead, which then puts even more pressure on
the worker.
Refactoring to use both the user ID & the device ID when tracking
the currently syncing users in the presence handler.
This is done both locally and over replication. Note that the device
ID is discarded but will be used in a future change.
Refactoring to pass the device ID (in addition to the user ID) through
the presence handler (specifically the `user_syncing`, `set_state`,
and `bump_presence_active_time` methods and their replication
versions).
Simplify some of the presence code by reducing duplicated code between
worker & non-worker modes.
The main change is to push some of the logic from `user_syncing` into
`set_state`. This is done by passing whether the user is setting the presence
via a `/sync` with a new `is_sync` flag to `set_state`. If this is `true` some
additional logic is performed:
* Don't override `busy` presence.
* Update the `last_user_sync_ts`.
* Never update the status message.
All the information needed is already in the `instance_map`, so
use that instead of passing the hostname / IP & port manually
for each replication request.
This consolidates logic for future improvements of using e.g.
UNIX sockets for workers.
* Add SSL options to redis config
* fix lint issues
* Add documentation and changelog file
* add missing . at the end of the changelog
* Move client context factory to new file
* Rename ssl to tls and fix typo
* fix lint issues
* Added when redis attributes were added
* Add master to the instance_map as part of Complement, have ReplicationEndpoint look at instance_map for master.
* Fix typo in drive by.
* Remove unnecessary worker_replication_* bits from unit tests and add master to instance_map(hopefully in the right place)
* Several updates:
1. Switch from master to main for naming the main process in the instance_map. Add useful constants for easier adjustment of names in the future.
2. Add backwards compatibility for worker_replication_* to allow time to transition to new style. Make sure to prioritize declaring main directly on the instance_map.
3. Clean up old comments/commented out code.
4. Adjust unit tests to match with new code.
5. Adjust Complement setup infrastructure to only add main to the instance_map if workers are used and remove now unused options from the worker.yaml template.
* Initial Docs upload
* Changelog
* Missed some commented out code that can go now
* Remove TODO comment that no longer holds true.
* Fix links in docs
* More docs
* Remove debug logging
* Apply suggestions from code review
Co-authored-by: reivilibre <olivier@librepush.net>
* Apply suggestions from code review
Co-authored-by: reivilibre <olivier@librepush.net>
* Update version to latest, include completeish before/after examples in upgrade notes.
* Fix up and docs too
---------
Co-authored-by: reivilibre <olivier@librepush.net>
Separate out a HTTP client for replication in preparation for
also supporting using UNIX sockets. The major difference from
the base class is that this does not use treq to handle HTTP
requests.
* Have replication clients remove _INT_STREAM_POS
Suppose worker A makes an internal http request from worker B. B may
make changes that A later learns about over replication. We want A's
request to block until it has seen those changes—mainly to ensure A's
caches are invalidated promptly. This helps provide read-after-write
consistency, eliminating entire categories of races and test flakes.
To implement this, B includes a top-level field `_INT_STREAM_POS` in its
response JSON. Roughly speaking, the field's value tells A what to wait
for. But we weren't removing that internal field before A's request
completed!
Introduced in https://github.com/matrix-org/synapse/pull/14820.
Fixes#15308.
* Changelog
With Redis commands do not need to be re-issued by the main
process (they fan-out to all processes at once) and thus it is no
longer necessary to worry about them reflecting recursively forever.
* Allow `AbstractSet` in `StrCollection`
Or else frozensets are excluded. This will be useful in an upcoming
commit where I plan to change a function that accepts `List[str]` to
accept `StrCollection` instead.
* `rooms_to_exclude` -> `rooms_to_exclude_globally`
I am about to make use of this exclusion mechanism to exclude rooms for
a specific user and a specific sync. This rename helps to clarify the
distinction between the global config and the rooms to exclude for a
specific sync.
* Better function names for internal sync methods
* Track a list of excluded rooms on SyncResultBuilder
I plan to feed a list of partially stated rooms for this sync to ignore
* Exclude partial state rooms during eager sync
using the mechanism established in the previous commit
* Track un-partial-state stream in sync tokens
So that we can work out which rooms have become fully-stated during a
given sync period.
* Fix mutation of `@cached` return value
This was fouling up a complement test added alongside this PR.
Excluding a room would mean the set of forgotten rooms in the cache
would be extended. This means that room could be erroneously considered
forgotten in the future.
Introduced in #12310, Synapse 1.57.0. I don't think this had any
user-visible side effects (until now).
* SyncResultBuilder: track rooms to force as newly joined
Similar plan as before. We've omitted rooms from certain sync responses;
now we establish the mechanism to reintroduce them into future syncs.
* Read new field, to present rooms as newly joined
* Force un-partial-stated rooms to be newly-joined
for eager incremental syncs only, provided they're still fully stated
* Notify user stream listeners to wake up long polling syncs
* Changelog
* Typo fix
Co-authored-by: Sean Quah <8349537+squahtx@users.noreply.github.com>
* Unnecessary list cast
Co-authored-by: Sean Quah <8349537+squahtx@users.noreply.github.com>
* Rephrase comment
Co-authored-by: Sean Quah <8349537+squahtx@users.noreply.github.com>
* Another comment
Co-authored-by: Sean Quah <8349537+squahtx@users.noreply.github.com>
* Fixup merge(?)
* Poke notifier when receiving un-partial-stated msg over replication
* Fixup merge whoops
Thanks MV :)
Co-authored-by: Mathieu Velen <mathieuv@matrix.org>
Co-authored-by: Mathieu Velten <mathieuv@matrix.org>
Co-authored-by: Sean Quah <8349537+squahtx@users.noreply.github.com>
* Faster joins: Update room stats and user directory on workers when done
When finishing a partial state join to a room, we update the current
state of the room without persisting additional events. Workers receive
notice of the current state update over replication, but neglect to wake
the room stats and user directory updaters, which then get incidentally
triggered the next time an event is persisted or an unrelated event
persister sends out a stream position update.
We wake the room stats and user directory updaters at the appropriate
time in this commit.
Part of #12814 and #12815.
Signed-off-by: Sean Quah <seanq@matrix.org>
* fixup comment
Signed-off-by: Sean Quah <seanq@matrix.org>
* Enable Complement tests for Faster Remote Room Joins on worker-mode
* (dangerous) Add an override to allow Complement to use FRRJ under workers
* Newsfile
Signed-off-by: Olivier Wilkinson (reivilibre) <oliverw@matrix.org>
* Fix race where we didn't send out replication notification
* MORE HACKS
* Fix get_un_partial_stated_rooms_token to take instance_name
* Fix bad merge
* Remove warning
* Correctly advance un_partial_stated_room_stream
* Fix merge
* Add another notify_replication
* Fixups
* Create a separate ReplicationNotifier
* Fix test
* Fix portdb
* Create a separate ReplicationNotifier
* Fix test
* Fix portdb
* Fix presence test
* Newsfile
* Apply suggestions from code review
* Update changelog.d/14752.misc
Co-authored-by: Erik Johnston <erik@matrix.org>
* lint
Signed-off-by: Olivier Wilkinson (reivilibre) <oliverw@matrix.org>
Co-authored-by: Erik Johnston <erik@matrix.org>
Now that we wait for stream positions whenever we do a HTTP replication
hit, we need to be less brutal in the case where we do timeout (as we
have bugs around this).
We were incorrectly checking if the *local* token had been advanced, rather than the token for the remote instance.
In practice, I don't think this has caused any bugs due to where we use `wait_for_stream_position`, as critically we don't use it on instances that also write to the given streams (and so the local token will lag behind all remote tokens).
This creates a new store method, `process_replication_position` that
is called after `process_replication_rows`. By moving stream ID advances
here this guarantees any relevant cache invalidations will have been
applied before the stream is advanced.
This avoids race conditions where Python switches between threads mid
way through processing the `process_replication_rows` method where stream
IDs may be advanced before caches are invalidated due to class resolution
ordering.
See this comment/issue for further discussion:
https://github.com/matrix-org/synapse/issues/14158#issuecomment-1344048703
Patrick Cloke