This should stop state events disappearing down a gap where we'd try to
separate out the sections *before* applying history visibility instead
of after.
This may be a better approach than #2843 but I hope @tak-hntlabs will
shout if it isn't.
If we're going backwards, we were selecting potentially thousands of
events, which in turn were fed to history visibility checks, resulting
in bad sync performance.
The problem was that we weren't getting enough recent events, as most of
them were removed by the history visibility filter. Now we're getting
all events between the given input range and re-slice the returned
values after applying history visibility.
Makes the tests
```
Can get rooms/{roomId}/members at a given point
Can filter rooms/{roomId}/members
```
pass, by moving `/members` and `/joined_members` to the SyncAPI.
This is going to make `Can get rooms/{roomId}/messages for a departed
room (SPEC-216)` pass, since we now only grep events from before the
user left the room.
This makes the following changes:
- get state deltas without the user supplied filter, so we can actually
"calculate" state transitions
- closes `stmt` when using SQLite
- Adds presence for users who newly joined a room, even if the syncing
user already knows about the presence status (should fix
https://github.com/matrix-org/complement/pull/516)
Improves the control flow of `GetStateDeltas` for clarity and possibly
also fixes a bug where duplicate state delta entries could be inserted
with different memberships instead of being correctly overridden by
`join`.
Use the stream positions of the notifier, which might have advanced
since setting it at the beginning of the loop. This possibly helps in
reducing roundtrips to the SyncAPI, just because we didn't fetch the
latest data.
Also fixes a minor oversight in the receipts stream.
First attempt at removing empty fields from `/sync` responses. Needs
https://github.com/matrix-org/sytest/pull/1298 to keep Sytest happy.
Co-authored-by: Neil Alexander <neilalexander@users.noreply.github.com>
This fixes a temporary workaround with the `selectEventsWithEventIDsSQL`
queries where fields need to be artificially added to the queries so the
row results match the format of the `syncapi_output_room_events` table.
I made similar functions that accept row results from the
`syncapi_current_room_state` table and convert them into StreamEvents
without the fields that are specific to output room events.
There is also a unit test in the first commit to ensure the resulting
behavior doesn't change from the modified queries and functions.
Fixes#601.
### Pull Request Checklist
<!-- Please read docs/CONTRIBUTING.md before submitting your pull
request -->
* [x] I have added tests for PR _or_ I have justified why this PR
doesn't need tests.
* [x] Pull request includes a [sign
off](https://github.com/matrix-org/dendrite/blob/main/docs/CONTRIBUTING.md#sign-off)
Signed-off-by: `Ashley Nelson <fant@shley.email>`
Co-authored-by: Neil Alexander <neilalexander@users.noreply.github.com>
This now uses a transaction per stream, so that errors in one stream
don't propagate to another, and we therefore no longer need to do hacks
to reopen a new transaction after aborting a failed one.
This should transactional snapshot isolation for `/sync` etc requests.
For now we don't use repeatable read due to some odd test failures with
invites.
…ce {}, a slice of interface` in new notifications select
The sqlite3 version was just not working, original pr here:
https://github.com/matrix-org/dendrite/pull/2688
signed off by: austin ellis <austin@hntlabs.com>
This doesn't fix the notification counts, they still only work about 1
out of every 5 times in my tests. I will stick with my other fix locally
for reliable notification delivery:
https://github.com/matrix-org/dendrite/pull/2701
Based on #2480
This actually indexes events based on their event type. They are removed
from the index if we receive a `m.room.redaction` event on the
`OutputRoomEvent` stream.
An admin endpoint is added to reindex all existing events.
Co-authored-by: Neil Alexander <neilalexander@users.noreply.github.com>
This PR changes the handling of notifications
- removes the `StreamEvent` and `ReadUpdate` stream
- listens on the `OutputRoomEvent` stream in the UserAPI to inform the
SyncAPI about unread notifications
- listens on the `OutputReceiptEvent` stream in the UserAPI to set
receipts/update notifications
- sets the `read_markers` directly from within the internal UserAPI
Co-authored-by: Neil Alexander <neilalexander@users.noreply.github.com>
Recently I have observed that dendrite spends a lot of time (~390s) in
`selectRoomIDsWithAnyMembershipSQL` query
```
dendrite_syncapi=# select total_exec_time, left(query,100) from pg_stat_statements order by total_exec_time desc limit 5 ;
total_exec_time | left
--------------------+------------------------------------------------------------------------------------------------------
747826.5800519128 | SELECT event_id, id, headered_event_json, session_id, exclude_from_sync, transaction_id, history_vis
389130.5490339942 | SELECT DISTINCT room_id, membership FROM syncapi_current_room_state WHERE type = $2 AND state_key =
376104.17514700035 | SELECT psd.datname, xact_commit, xact_rollback, blks_read, blks_hit, tup_returned, tup_fetched, tup_
363644.164092031 | SELECT event_type_nid, event_state_key_nid, event_nid FROM roomserver_events WHERE event_nid = ANY($
58570.48104699995 | SELECT event_id, headered_event_json FROM syncapi_current_room_state WHERE room_id = $1 AND ( $2::te
(5 rows)
```
Explain analyze showed correct usage of `syncapi_room_state_unique`
index:
```
dendrite_syncapi=#
explain analyze SELECT distinct room_id, membership FROM syncapi_current_room_state WHERE type = 'm.room.member' AND state_key = '@qjfl:dendrite.stg.globekeeper.com';
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Unique (cost=2749.38..2749.56 rows=24 width=52) (actual time=2.933..2.956 rows=65 loops=1)
-> Sort (cost=2749.38..2749.44 rows=24 width=52) (actual time=2.932..2.937 rows=65 loops=1)
Sort Key: room_id, membership
Sort Method: quicksort Memory: 34kB
-> Index Scan using syncapi_room_state_unique on syncapi_current_room_state (cost=0.41..2748.83 rows=24 width=52) (actual time=0.030..2.890 rows=65 loops=1)
Index Cond: ((type = 'm.room.member'::text) AND (state_key = '@qjfl:dendrite.stg.globekeeper.com'::text))
Planning Time: 0.140 ms
Execution Time: 2.990 ms
(8 rows)
```
Multi-column indexes in Postgres shall perform well for leftmost
columns, but I gave it a try and created
`syncapi_current_room_state_type_state_key_idx` index. I could observe
significant performance improvement. Execution time dropped from 2.9 ms
to 0.24 ms:
```
explain analyze SELECT distinct room_id, membership FROM syncapi_current_room_state WHERE type = 'm.room.member' AND state_key = '@qjfl:dendrite.stg.globekeeper.com';
QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
Unique (cost=96.46..96.64 rows=24 width=52) (actual time=0.199..0.218 rows=65 loops=1)
-> Sort (cost=96.46..96.52 rows=24 width=52) (actual time=0.199..0.202 rows=65 loops=1)
Sort Key: room_id, membership
Sort Method: quicksort Memory: 34kB
-> Bitmap Heap Scan on syncapi_current_room_state (cost=4.53..95.91 rows=24 width=52) (actual time=0.048..0.139 rows=65 loops=1)
Recheck Cond: ((type = 'm.room.member'::text) AND (state_key = '@qjfl:dendrite.stg.globekeeper.com'::text))
Heap Blocks: exact=59
-> Bitmap Index Scan on syncapi_current_room_state_type_state_key_idx (cost=0.00..4.53 rows=24 width=0) (actual time=0.037..0.037 rows=65 loops=1)
Index Cond: ((type = 'm.room.member'::text) AND (state_key = '@qjfl:dendrite.stg.globekeeper.com'::text))
Planning Time: 0.236 ms
Execution Time: 0.242 ms
(11 rows)
```
Next improvement is skipping DISTINCT and rely on map assignment in
`SelectRoomIDsWithAnyMembership`. Execution time drops by almost half:
```
explain analyze SELECT room_id, membership FROM syncapi_current_room_state WHERE type = 'm.room.member' AND state_key = '@qjfl:dendrite.stg.globekeeper.com';
QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------
Bitmap Heap Scan on syncapi_current_room_state (cost=4.53..95.91 rows=24 width=52) (actual time=0.032..0.113 rows=65 loops=1)
Recheck Cond: ((type = 'm.room.member'::text) AND (state_key = '@qjfl:dendrite.stg.globekeeper.com'::text))
Heap Blocks: exact=59
-> Bitmap Index Scan on syncapi_current_room_state_type_state_key_idx (cost=0.00..4.53 rows=24 width=0) (actual time=0.021..0.021 rows=65 loops=1)
Index Cond: ((type = 'm.room.member'::text) AND (state_key = '@qjfl:dendrite.stg.globekeeper.com'::text))
Planning Time: 0.087 ms
Execution Time: 0.136 ms
(7 rows)
```
In our env we spend only 1s on inserting to table, so the write penalty
of creating an index should be small.
```
dendrite_syncapi=# select total_exec_time, left(query,100) from pg_stat_statements where query like '%INSERT%syncapi_current_room_state%' order by total_exec_time desc;
total_exec_time | left
--------------------+------------------------------------------------------------------------------------------------------
1139.9057619999971 | INSERT INTO syncapi_current_room_state (room_id, event_id, type, sender, contains_url, state_key, he
(1 row)
```
This PR does not require test modifications.
### Pull Request Checklist
<!-- Please read docs/CONTRIBUTING.md before submitting your pull
request -->
* [x] I have added added tests for PR _or_ I have justified why this PR
doesn't need tests.
* [x] Pull request includes a [sign
off](https://github.com/matrix-org/dendrite/blob/main/docs/CONTRIBUTING.md#sign-off)
Signed-off-by: `Piotr Kozimor <p1996k@gmail.com>`
Previously `LoadMembershipAtEvent` would fail if the state before one of
the events was not known, i.e. because it was an outlier. This modifies
it so that it gracefully handles not knowing the state and returns no
memberships instead, so that history visibility doesn't freak out and
kill `/sync` requests dead.
Some tweaks for the send-to-device consumers/producers:
- use `json.RawMessage` without marshalling it first
- try further devices (if available) if we failed to `PublishMsg` in the
producers
- some logging changes (to better debug E2EE issues)
Introduced index improves select query performance. Example execution time of `selectSendToDeviceMessagesSQL` query dropped from 80 ms to 15 ms. No sytest modifications are required.
### Pull Request Checklist
* [x] I have added added tests for PR _or_ I have justified why this PR doesn't need tests.
* [x] Pull request includes a [sign off](https://github.com/matrix-org/dendrite/blob/main/docs/CONTRIBUTING.md#sign-off)
Signed-off-by: `Piotr Kozimor <p1996k@gmail.com>`
This should fix an issue where we return less than the expected membership events, when doing an initial sync.
When doing an initial sync, the state limit is set to `math.MaxInt32`, while the default filter is set to 20.
- Reverts 9dc57122d9 as it was causing issues https://github.com/matrix-org/dendrite/issues/2660
- Updates the GMSL `DefaultStateFilter` to use a limit of 20 events
- Uses the timeline events to determine the new position instead of the state events
This should hopefully deflake Backfill works correctly with history visibility set to joined as we were using the default shared visibility, even if the events are set to joined (or something else)
