Commit Graph

258 Commits

Author SHA1 Message Date
Eric Eastwood
29269d9d3f
Fix have_seen_event cache not being invalidated ()
Fix https://github.com/matrix-org/synapse/issues/13856
Fix https://github.com/matrix-org/synapse/issues/13865

> Discovered while trying to make Synapse fast enough for [this MSC2716 test for importing many batches](https://github.com/matrix-org/complement/pull/214#discussion_r741678240). As an example, disabling the `have_seen_event` cache saves 10 seconds for each `/messages` request in that MSC2716 Complement test because we're not making as many federation requests for `/state` (speeding up `have_seen_event` itself is related to https://github.com/matrix-org/synapse/issues/13625) 
> 
> But this will also make `/messages` faster in general so we can include it in the [faster `/messages` milestone](https://github.com/matrix-org/synapse/milestone/11).
> 
> *-- https://github.com/matrix-org/synapse/issues/13856*


### The problem

`_invalidate_caches_for_event` doesn't run in monolith mode which means we never even tried to clear the `have_seen_event` and other caches. And even in worker mode, it only runs on the workers, not the master (AFAICT).

Additionally there was bug with the key being wrong so `_invalidate_caches_for_event` never invalidates the `have_seen_event` cache even when it does run.

Because we were using the `@cachedList` wrong, it was putting items in the cache under keys like `((room_id, event_id),)` with a `set` in a `set` (ex. `(('!TnCIJPKzdQdUlIyXdQ:test', '$Iu0eqEBN7qcyF1S9B3oNB3I91v2o5YOgRNPwi_78s-k'),)`) and we we're trying to invalidate with just `(room_id, event_id)` which did nothing.
2022-09-27 15:55:43 -05:00
Mathieu Velten
6bd8763804
Add cache invalidation across workers to module API ()
Signed-off-by: Mathieu Velten <mathieuv@matrix.org>
2022-09-21 15:32:01 +02:00
reivilibre
cf65433de2
Fix a memory leak when running the unit tests. () 2022-09-14 15:29:05 +00:00
reivilibre
cf11919ddd
Fix cache metrics not being updated when not using the legacy exposition module. () 2022-09-08 15:30:48 +01:00
reivilibre
b455c2a5ec
Update Grafana dashboard to not use legacy metric names. () 2022-09-06 12:21:21 +01:00
reivilibre
be4250c7a8
Add experimental configuration option to allow disabling legacy Prometheus metric names. ()
Co-authored-by: David Robertson <davidr@element.io>
2022-08-24 11:35:54 +00:00
Erik Johnston
f7ddfe17a3
Speed up @cachedList ()
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.
2022-08-23 14:53:27 +00:00
Nick Mills-Barrett
5e7847dc92
Cache user IDs instead of profile objects ()
The profile objects are never used and increase cache size significantly.
2022-08-23 09:49:59 +00:00
Nick Mills-Barrett
41320a0554
Optimise async get event lookups ()
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.
2022-08-04 15:49:55 +01:00
Erik Johnston
0b87eb8e0c
Make DictionaryCache have better expiry properties () 2022-07-21 17:13:44 +01:00
Nick Mills-Barrett
cc21a431f3
Async get event cache prep ()
Some experimental prep work to enable external event caching based on  & . Doesn't actually move the cache at all, just lays the groundwork for async implemented caches.

Signed off by Nick @ Beeper (@Fizzadar)
2022-07-15 09:30:46 +00:00
David Robertson
6ba732fefe
Type tests.utils ()
* Cast to postgres types when handling postgres db

* Remove unused method

* Easy annotations

* Annotate create_room

* Use `ParamSpec` to annotate looping_call

* Annotate `default_config`

* Track `now` as a float

`time_ms` returns an int like the proper Synapse `Clock`

* Introduce a `Timer` dataclass

* Introduce a Looper type

* Suppress checking of a mock

* tests.utils is typed

* Changelog

* Whoops, import ParamSpec from typing_extensions

* ditch the psycopg2 casts
2022-07-05 15:13:47 +01:00
Patrick Cloke
759f9c09e1
Fix caching behavior for relations push rules. ()
By always returning all requested values from the function
wrapped by cachedList. Otherwise implicit None values get
added into the cache, which are unexpected.
2022-05-25 07:49:54 -04:00
Shay
cde8af9a49
Add config flags to allow for cache auto-tuning () 2022-05-13 12:32:39 -07:00
David Robertson
6463244375
Remove unused # type: ignores ()
Over time we've begun to use newer versions of mypy, typeshed, stub
packages---and of course we've improved our own annotations. This makes
some type ignore comments no longer necessary. I have removed them.

There was one exception: a module that imports `select.epoll`. The
ignore is redundant on Linux, but I've kept it ignored for those of us
who work on the source tree using not-Linux. ()

I'm more interested in the config line which enforces this. I want
unused ignores to be reported, because I think it's useful feedback when
annotating to know when you've fixed a problem you had to previously
ignore.

* Installing extras before typechecking

Lacking an easy way to install all extras generically, let's bite the bullet and
make install the hand-maintained `all` extra before typechecking.

Now that https://github.com/matrix-org/backend-meta/pull/6 is merged to
the release/v1 branch.
2022-04-27 14:03:44 +01:00
Patrick Cloke
11df4ec6c2
Add more type hints to the main state store. () 2022-03-31 13:38:09 -04:00
Sean Quah
2fcf4b3f6c
Add cancellation support to @cached and @cachedList decorators ()
These decorators mostly support cancellation already. Add cancellation
tests and fix use of finished logging contexts by delaying cancellation,
as suggested by @erikjohnston.

Signed-off-by: Sean Quah <seanq@element.io>
2022-03-14 19:04:29 +00:00
Patrick Cloke
bc9dff1d95
Remove unnecessary pass statements. () 2022-03-11 07:06:21 -05:00
Patrick Cloke
690cb4f3b3
Allow for ignoring some arguments when caching. ()
* `@cached` can now take an `uncached_args` which is an iterable of names to not use in the cache key.
* Requires `@cached`, @cachedList` and `@lru_cache` to use keyword arguments for clarity.
* Asserts that keyword-only arguments in cached functions are not accepted. (I tested this briefly and I don't believe this works properly.)
2022-03-09 18:07:41 +00:00
Richard van der Hoff
5458eb8551
Fix 'Unhandled error in Deferred' ()
* Fix 'Unhandled error in Deferred'

Fixes a CRITICAL "Unhandled error in Deferred" log message which happened when
a function wrapped with `@cachedList` failed

* Minor optimisation to cachedListDescriptor

we can avoid re-using `missing`, which saves looking up entries in
`deferreds_map`, and means we don't need to copy it.

* Improve type annotation on CachedListDescriptor
2022-03-01 09:51:38 +00:00
Erik Johnston
0dbbe33a65
Track cache invalidations ()
Currently we only track evictions due to size or time constraints.
2022-02-15 14:31:04 +00:00
David Robertson
f510fba4ba
Describe prune_unread_entries in docstrings ()
Should have been caught in .
2022-02-02 15:11:23 +00:00
Patrick Cloke
10a88ba91c
Use auto_attribs/native type hints for attrs classes. () 2022-01-13 13:49:28 +00:00
Shay
fcfe67578f
Update to the current version of Black and run it on Synapse codebase ()
* update black version

* run updated version of black on code

* newsfragment

* enumerate python versions
2021-12-23 20:22:15 -08:00
Richard van der Hoff
c3e38b88f2
Improve opentracing support for ResponseCache ()
This adds some opentracing annotations to ResponseCache, to make it easier to see what's going on; in particular, it adds a link back to the initial trace which is actually doing the work of generating the response.
2021-12-20 18:12:08 +00:00
Sean Quah
0147b3de20
Add missing type hints to synapse.logging.context () 2021-12-14 17:35:28 +00:00
Sean Quah
7ff22d6da4
Fix LruCache corruption bug with a size_callback that can return 0 ()
When all entries in an `LruCache` have a size of 0 according to the
provided `size_callback`, and `drop_from_cache` is called on a cache
node, the node would be unlinked from the LRU linked list but remain in
the cache dictionary. An assertion would be later be tripped due to the
inconsistency.

Avoid unintentionally calling `__len__` and use a strict `is None`
check instead when unwrapping the weak reference.
2021-11-30 16:28:02 +00:00
Sean Quah
5a0b652d36
Eliminate a few Anys in LruCache type hints () 2021-11-30 15:39:07 +00:00
Sean Quah
84fac0f814
Add type annotations to synapse.metrics () 2021-11-17 19:07:02 +00:00
Patrick Cloke
7468723697
Add most missing type hints to synapse.util () 2021-11-16 08:47:36 -05:00
David Robertson
f8d0f72b27
More types for synapse.util, part 1 ()
The following modules now pass `disallow_untyped_defs`:

* synapse.util.caches.cached_call 
* synapse.util.caches.lrucache
* synapse.util.caches.response_cache 
* synapse.util.caches.stream_change_cache
* synapse.util.caches.ttlcache pass
* synapse.util.daemonize
* synapse.util.patch_inline_callbacks pass `no-untyped-defs`
* synapse.util.versionstring

Additional typing in synapse.util.metrics. Didn't get this to pass `no-untyped-defs`, think I'll need to watch 
2021-10-06 11:20:49 +01:00
David Robertson
724aef9a87
Opt out of cache expiry for get_users_who_share_room_with_user ()
* Allow LruCaches to opt out of time-based expiry
* Don't expire `get_users_who_share_room` & friends
2021-09-22 14:21:58 +01:00
David Robertson
a2d7195e01
Track why we're evicting from caches ()
So we can see distinguish between "evicting because the cache is too big" and "evicting because the cache entries haven't been recently used".
2021-09-22 10:59:52 +01:00
Patrick Cloke
3eba047d38
Add type hints to state database module. () 2021-09-15 09:54:13 -04:00
reivilibre
524b8ead77
Add types to synapse.util. () 2021-09-10 17:03:18 +01:00
Richard van der Hoff
d9cb658c78
Fix up type hints for Twisted 21.7 ()
Mostly this involves decorating a few Deferred declarations with extra type hints. We wrap the types in quotes to avoid runtime errors when running against older versions of Twisted that don't have generics on Deferred.
2021-07-28 12:04:11 +00:00
Richard van der Hoff
9643dfde6a
improve typing annotations in CachedCall ()
tighten up some of the typing in CachedCall, which is going to be needed when
Twisted 21.7 brings better typing on Deferred.
2021-07-28 12:25:12 +01:00
Jonathan de Jong
95e47b2e78
[pyupgrade] synapse/ ()
This PR is tantamount to running 
```
pyupgrade --py36-plus --keep-percent-format `find synapse/ -type f -name "*.py"`
```

Part of 
2021-07-19 15:28:05 +01:00
Jonathan de Jong
bdfde6dca1
Use inline type hints in http/federation/, storage/ and util/ () 2021-07-15 12:46:54 -04:00
Erik Johnston
7a5873277e
Add support for evicting cache entries based on last access time. () 2021-07-05 16:32:12 +01:00
Richard van der Hoff
d7808a2dde
Extend ResponseCache to pass a context object into the callback ()
This is the first of two PRs which seek to address . This first PR lays the groundwork by extending ResponseCache; a second PR () will update the SyncHandler to actually use it, and fix the bug.

The idea here is that we allow the callback given to ResponseCache.wrap to decide whether its result should be cached or not. We do that by (optionally) passing a ResponseCacheContext into it, which it can modify.
2021-06-14 10:26:09 +01:00
Richard van der Hoff
224f2f949b
Combine LruCache.invalidate and invalidate_many ()
* Make `invalidate` and `invalidate_many` do the same thing

... so that we can do either over the invalidation replication stream, and also
because they always confused me a bit.

* Kill off `invalidate_many`

* changelog
2021-05-27 10:33:56 +01:00
Richard van der Hoff
c0df6bae06
Remove keylen from LruCache. ()
`keylen` seems to be a thing that is frequently incorrectly set, and we don't really need it.

The only time it was used was to figure out if we had removed a subtree in `del_multi`, which we can do better by changing `TreeCache.pop` to return a different type (`TreeCacheNode`).

Commits should be independently reviewable.
2021-05-24 14:02:01 +01:00
Richard van der Hoff
5090f26b63
Minor @cachedList enhancements ()
- use a tuple rather than a list for the iterable that is passed into the
  wrapped function, for performance

- test that we can pass an iterable and that keys are correctly deduped.
2021-05-14 11:12:36 +01:00
Erik Johnston
ef889c98a6
Optionally track memory usage of each LruCache ()
This will double count slightly in the presence of interned strings. It's off by default as it can consume a lot of resources.
2021-05-05 16:54:36 +01:00
Erik Johnston
391bfe9a7b
Reduce memory footprint of caches () 2021-04-28 11:59:28 +01:00
Andrew Morgan
fe604a022a
Remove various bits of compatibility code for Python <3.6 ()
I went through and removed a bunch of cruft that was lying around for compatibility with old Python versions. This PR also will now prevent Synapse from starting unless you're running Python 3.6+.
2021-04-27 13:13:07 +01:00
Richard van der Hoff
294c675033
Remove synapse.types.Collection ()
This is no longer required, since we have dropped support for Python 3.5.
2021-04-22 16:43:50 +01:00
Jonathan de Jong
4b965c862d
Remove redundant "coding: utf-8" lines ()
Part of 

Removes all redundant `# -*- coding: utf-8 -*-` lines from files, as python 3 automatically reads source code as utf-8 now.

`Signed-off-by: Jonathan de Jong <jonathan@automatia.nl>`
2021-04-14 15:34:27 +01:00
Jonathan de Jong
2ca4e349e9
Bugbear: Add Mutable Parameter fixes ()
Part of 

Adds in fixes for B006 and B008, both relating to mutable parameter lint errors.

Signed-off-by: Jonathan de Jong <jonathan@automatia.nl>
2021-04-08 22:38:54 +01:00