Commit graph

11229 commits

Author SHA1 Message Date
Erik Johnston c7320a5564
Merge pull request #3544 from matrix-org/erikj/fixup_stream_cache
Fix perf regression in PR #3530
2018-07-17 11:16:59 +01:00
Erik Johnston b2aa05a8d6 Use efficient .intersection 2018-07-17 11:07:04 +01:00
Erik Johnston 850238b4ef Add unit test 2018-07-17 10:59:02 +01:00
Erik Johnston 9952d18e4d Newsfile 2018-07-17 10:31:51 +01:00
Erik Johnston 547b1355d3 Fix perf regression in PR #3530
The get_entities_changed function was changed to return all changed
entities since the given stream position, rather than only those changed
from a given list of entities. This resulted in the function incorrectly
returning large numbers of entities that, for example, caused large
increases in database usage.
2018-07-17 10:27:51 +01:00
Amber Brown 3fe0938b76
Merge pull request #3530 from matrix-org/erikj/stream_cache
Don't return unknown entities in get_entities_changed
2018-07-17 13:44:46 +10:00
Amber Brown fe10dd9fb2
Merge pull request #3540 from krombel/enforce_isort
check isort by travis
2018-07-17 13:41:59 +10:00
Krombel 9677b1d1c0 rename 'isort' to 'check_isort' as requested 2018-07-16 16:03:41 +02:00
Krombel 78a9ddcf9a rerun isort with latest version 2018-07-16 14:23:25 +02:00
Krombel 4a27000548 check isort by travis 2018-07-16 13:57:33 +02:00
Amber Brown 8a4f05fefb
Fix develop because I broke it :( (#3535) 2018-07-14 09:51:00 +10:00
Amber Brown 8532953c04
Merge pull request #3534 from krombel/use_parse_and_asserts_from_servlet
Use parse and asserts from http.servlet
2018-07-14 09:09:19 +10:00
Amber Brown a2374b2c7f
fix sytests 2018-07-14 07:52:58 +10:00
Amber Brown 33b60c01b5
Make auth & transactions more testable (#3499) 2018-07-14 07:34:49 +10:00
Krombel 516f960ad8 add changelog 2018-07-13 22:19:19 +02:00
Krombel 3366b9c534 rename assert_params_in_request to assert_params_in_dict
the method "assert_params_in_request" does handle dicts and not
requests. A request body has to be parsed to json before this method
can be used
2018-07-13 21:53:01 +02:00
Krombel 32fd6910d0 Use parse_{int,str} and assert from http.servlet
parse_integer and parse_string can take a request and raise errors
in case we have wrong or missing params.
This PR tries to use them more to deduplicate some code and make it
better readable
2018-07-13 21:40:14 +02:00
Erik Johnston bc832f822f Fixup unit test 2018-07-13 17:03:04 +01:00
Richard van der Hoff 2aba1f549c
Merge pull request #3533 from matrix-org/rav/fix_federation_ratelimite_queue
Make FederationRateLimiter queue requests properly
2018-07-13 16:59:18 +01:00
Richard van der Hoff 08546be40f better changelog 2018-07-13 16:47:13 +01:00
Richard van der Hoff b03a0e14db changelog 2018-07-13 16:28:28 +01:00
Richard van der Hoff 3b391d9c45 Fix unit tests 2018-07-13 16:28:04 +01:00
Richard van der Hoff 33b40d0a25 Make FederationRateLimiter queue requests properly
popitem removes the *most recent* item by default [1]. We want the oldest.

Fixes #3524

[1]: https://docs.python.org/2/library/collections.html#collections.OrderedDict.popitem
2018-07-13 16:19:40 +01:00
Erik Johnston 5f263b607e Newsfile 2018-07-13 15:47:25 +01:00
Erik Johnston 77b692e65d Don't return unknown entities in get_entities_changed
The stream cache keeps track of all entities that have changed since
a particular stream position, so get_entities_changed does not need to
return unknown entites when given a larger stream position.

This makes it consistent with the behaviour of has_entity_changed.
2018-07-13 15:26:10 +01:00
Matthew Hodgson ba22b6a456 typo 2018-07-13 12:03:39 +01:00
Richard van der Hoff 6dff49b8a9
Merge pull request #3521 from matrix-org/rav/optimise_stream_change_cache
Reduce set building in get_entities_changed
2018-07-12 12:08:49 +01:00
Richard van der Hoff 38c5fa7ee4 changelog 2018-07-12 11:45:28 +01:00
Richard van der Hoff fa5c2bc082 Reduce set building in get_entities_changed
This line shows up as about 5% of cpu time on a synchrotron:

    not_known_entities = set(entities) - set(self._entity_to_key)

Presumably the problem here is that _entity_to_key can be largeish, and
building a set for its keys every time this function is called is slow.

Here we rewrite the logic to avoid building so many sets.
2018-07-12 11:37:44 +01:00
Richard van der Hoff 4c4dd6299d
Merge pull request #3316 from matrix-org/rav/enforce_report_api
Enforce the specified API for report_event
2018-07-12 10:50:05 +01:00
Richard van der Hoff c969754a91 changelog 2018-07-12 10:00:57 +01:00
Richard van der Hoff 482d17b58b Merge branch 'develop' into rav/enforce_report_api 2018-07-12 09:56:28 +01:00
Erik Johnston 0456e05977
Merge pull request #3505 from matrix-org/erikj/receipts_cahce
Use stream cache in get_linearized_receipts_for_room
2018-07-12 09:46:29 +01:00
Erik Johnston aff1dfdf3d Update return value docstring 2018-07-12 09:45:37 +01:00
Amber Brown 129ffd7b88
Merge pull request #3498 from OlegGirko/fix_attrs_syntax
* Use more portable syntax using attrs package.

Newer syntax

    attr.ib(factory=dict)

is just a syntactic sugar for

    attr.ib(default=attr.Factory(dict))

It was introduced in newest version of attrs package (18.1.0)
and doesn't work with older versions.

We should either require minimum version of attrs to be 18.1.0,
or use older (slightly more verbose) syntax.
Requiring newest version is not a good solution because
Linux distributions may have older version of attrs (17.4.0 in Fedora 28),
and requiring to build (and package)
newer version just to use newer syntactic sugar in only one test
is just too much.
It's much better to fix that test to use older syntax.

Signed-off-by: Oleg Girko <ol@infoserver.lv>
2018-07-11 04:22:46 +10:00
Amber Brown 85354bb18e changelog entry 2018-07-11 03:27:03 +10:00
Erik Johnston 6ccefef07a Use 'is not None' and add comments 2018-07-10 18:12:39 +01:00
Matthew Hodgson ea752bdd99 s/becuase/because/g 2018-07-10 17:58:18 +01:00
Erik Johnston bb3d536087 Newsfile 2018-07-10 17:28:31 +01:00
Erik Johnston 05f5dabc10 Use stream cache in get_linearized_receipts_for_room
This avoids us from uncessarily hitting the database when there has been
no change for the room
2018-07-10 17:22:42 +01:00
Richard van der Hoff c3c29aa196
Attempt to include db threads in cpu usage stats (#3496)
Let's try to include time spent in the DB threads in the per-request/block cpu
usage metrics.
2018-07-10 16:12:36 +01:00
Richard van der Hoff 55370331da
Refactor logcontext resource usage tracking (#3501)
Factor out the resource usage tracking out to a separate object, which can be
passed around and copied independently of the logcontext itself.
2018-07-10 13:56:07 +01:00
Matthew Hodgson 16b10666e7 another typo 2018-07-10 12:28:42 +01:00
Matthew Hodgson 4ea391a6ae typo (i think) 2018-07-10 12:08:09 +01:00
Richard van der Hoff b1fe697b3c
Merge pull request #3497 from matrix-org/rav/measure_fetch_event_loop
Add CPU metrics for _fetch_event_list
2018-07-10 10:00:24 +01:00
Oleg Girko 6c1ec5a1bd Use more portable syntax using attrs package.
Newer syntax

    attr.ib(factory=dict)

is just a syntactic sugar for

    attr.ib(default=attr.Factory(dict))

It was introduced in newest version of attrs package (18.1.0)
and doesn't work with older versions.

We should either require minimum version of attrs to be 18.1.0,
or use older (slightly more verbose) syntax.
Requiring newest version is not a good solution because
Linux distributions may have older version of attrs (17.4.0 in Fedora 28),
and requiring to build (and package)
newer version just to use newer syntactic sugar in only one test
is just too much.
It's much better to fix that test to use older syntax.

Signed-off-by: Oleg Girko <ol@infoserver.lv>
2018-07-10 00:38:49 +01:00
Richard van der Hoff f3b3b9dd8f changelog 2018-07-09 18:16:52 +01:00
Richard van der Hoff e31e5dee38 Add CPU metrics for _fetch_event_list
add a Measure block on _fetch_event_list, in the hope that we can better
measure CPU usage here.
2018-07-09 18:15:54 +01:00
Richard van der Hoff 395fa8d1fd
Merge pull request #3464 from matrix-org/hawkowl/isort-run
Run isort on Synapse
2018-07-09 10:24:43 +01:00
Amber Brown 09477bd884 changelog 2018-07-09 16:09:37 +10:00