Re: [PATCH v2] xmlrpc/patch_list: only fetch required fields

2017-10-19 Thread Stephen Finucane
On Thu, 2017-10-19 at 02:53 +1100, Daniel Axtens wrote: > OzLabs noticed *massive* slowdowns in queries like this one: > > SELECT "patchwork_submission"."id", "patchwork_submission"."msgid", > "patchwork_submission"."date", "patchwork_submission"."headers", > "patchwork_submission"."submitter_id",

[PATCH v2 3/4] Support testing with PostgreSQL

2017-10-19 Thread Daniel Axtens
This allows us to easily test against PostgreSQL using the same tooling we normally use. This is helpful in (for example) shaking out the test failures that were observed on ozlabs.org To use it: docker-compose -f docker-compose-pg.yml (You may find in necessary to do a 'docker-compose down' f

[PATCH v2 1/4] REST: Specify default ordering fields

2017-10-19 Thread Daniel Axtens
From: Stephen Finucane This hides warnings likes the following: UnorderedObjectListWarning: Pagination may yield inconsistent results with an unordered object_list: ]> Signed-off-by: Stephen Finucane [dja: fix merge conflict in patch.py] Signed-off-by: Daniel Axtens --- patchwork/api

[PATCH v2 2/4] Simplify docker-compose MySQL setup

2017-10-19 Thread Daniel Axtens
This moves the config from a separate Dockerfile to something integrated into the docker-compose.yml file. Signed-off-by: Daniel Axtens --- Having this set up with an image rather than a build makes it much easier for the postgres file to come. --- docker-compose.yml| 9 +++--

[PATCH v2 0/4] PostgreSQL fixes and test support

2017-10-19 Thread Daniel Axtens
Hi all, Production deployments of Patchwork often use postgres instead of mysql. So far we haven't been testing that very well, and so when OzLabs was upgraded some bugs were discovered. jk has kindly sent patches to fix the fuzztest failures. We also saw errors due to ordering - one of Stephen's

[PATCH v2 4/4] Test postgresql support in Travis CI

2017-10-19 Thread Daniel Axtens
Use the most recent version of postgres (9.6) because we need a more recent version than the default to show up the broken bundle behaviour. This should prevent us from causing any further regressions. Signed-off-by: Daniel Axtens --- v2: MySQL has already been moved to the root user --- .tra