1121589 - Schedule Mn-e10s jobs on inbound, central, and try

Status: RESOLVED FIXED

(Release Engineering :: General, defect)

Description

[Chris Manchester (limited bugmail, email directly)]

I think we can start these off on whichever platform is most amenable to more test jobs. My impression is that Linux32 would be appropriate.

Comment 1

[Chris Manchester (limited bugmail, email directly)]

Attachment: Run marionette-e10s on trunk branches

jlund, perhaps you can advise on comment 0 as well?

Builders added:
+ Ubuntu VM 12.04 ash opt test marionette-e10s
+ Ubuntu VM 12.04 b2g-inbound opt test marionette-e10s
+ Ubuntu VM 12.04 b2g-inbound pgo test marionette-e10s
+ Ubuntu VM 12.04 cedar opt test marionette-e10s
+ Ubuntu VM 12.04 cypress opt test marionette-e10s
+ Ubuntu VM 12.04 elm opt test marionette-e10s
+ Ubuntu VM 12.04 fig opt test marionette-e10s
+ Ubuntu VM 12.04 fx-team opt test marionette-e10s
+ Ubuntu VM 12.04 fx-team pgo test marionette-e10s
+ Ubuntu VM 12.04 jamun opt test marionette-e10s
+ Ubuntu VM 12.04 maple opt test marionette-e10s
+ Ubuntu VM 12.04 mozilla-central opt test marionette-e10s
+ Ubuntu VM 12.04 mozilla-central pgo test marionette-e10s
+ Ubuntu VM 12.04 mozilla-inbound opt test marionette-e10s
+ Ubuntu VM 12.04 mozilla-inbound pgo test marionette-e10s
+ Ubuntu VM 12.04 oak opt test marionette-e10s
+ Ubuntu VM 12.04 try opt test marionette-e10s
+ Ubuntu VM 12.04 try pgo test marionette-e10s
+ Ubuntu VM 12.04 ux opt test marionette-e10s

Comment 2

[Jordan Lund (:jlund)]

Comment on attachment 8549267 [details] [diff] [review]
Run marionette-e10s on trunk branches

Review of attachment 8549267 [details] [diff] [review]:
-----------------------------------------------------------------

lgtm

test coverage: even though this enables across trunk all at once, Chris, over irc, confirmed that he has done a few try pushes (via forcing --e10s in in-tree options). Both linux32 and linux64 was tested. He also tested the --e10s mozharness option locally.

confirmed this will cause mozharness to be called across gecko >= m-c + try with the following cmd:
        e.g. MockCommand {'command': ['/tools/buildbot/bin/python',  'scripts/scripts/marionette.py',  '--cfg',  'marionette/prod_config.py',  '--e10s',  '--blob-upload-branch',  'ash'], 'description': None, 'descriptionDone': None, 'env': {'CCACHE_DIR': '/builds/ccache',  'CCACHE_UMASK': '002',  'DISPLAY': ':0',  'MOZ_HIDE_RESULTS_TABLE': '1',  'MOZ_NODE_PATH': '/usr/bin/node',  'MOZ_NO_REMOTE': '1',  'NO_FAIL_ON_TEST_ERRORS': '1',  'PROPERTIES_FILE': <WithProperties "%(basedir)s/buildprops.json">}, 'haltOnFailure': True, 'log_eval_func': <function eval_func>, 'logfiles': {}, 'maxTime': 7200, 'mock': False, 'mock_args': ['--unpriv'], 'mock_login': 'mock_mozilla', 'mock_workdir_mutator': <function <lambda>>, 'mock_workdir_prefix': '%(basedir)s/', 'name': 'run_script', 'target': None, 'timeout': 1800, 'usePTY': 'slave-config', 'warnOnWarnings': True, 'workdir': '.'} {}

coop feedback request: before landing, I'd like confirmation from coop that tst-linux32 has less load/capacity issues than tst-linux64. cmanchester confirmed that either arch works from a test perspective.

Comment 3

[Chris Cooper [:coop] (he/him)]

Comment on attachment 8549267 [details] [diff] [review]
Run marionette-e10s on trunk branches

Review of attachment 8549267 [details] [diff] [review]:
-----------------------------------------------------------------

(In reply to Jordan Lund (:jlund) from comment #2)
> coop feedback request: before landing, I'd like confirmation from coop that
> tst-linux32 has less load/capacity issues than tst-linux64. cmanchester
> confirmed that either arch works from a test perspective.

Correct. If you have a choice, tst-linux32 is a better pool to target right now.

Comment 4

[Jordan Lund (:jlund)]

(In reply to Chris Cooper [:coop] from comment #3)
> Comment on attachment 8549267 [details] [diff] [review]
> Run marionette-e10s on trunk branches
> 
> Review of attachment 8549267 [details] [diff] [review]:
> -----------------------------------------------------------------
> 
> (In reply to Jordan Lund (:jlund) from comment #2)
> > coop feedback request: before landing, I'd like confirmation from coop that
> > tst-linux32 has less load/capacity issues than tst-linux64. cmanchester
> > confirmed that either arch works from a test perspective.
> 
> Correct. If you have a choice, tst-linux32 is a better pool to target right
> now.

ship it :)

Comment 5

[Chris Manchester (limited bugmail, email directly)]

https://hg.mozilla.org/build/buildbot-configs/rev/ca4c43dc230c

Comment 6

[Chris Cooper [:coop] (he/him)]

In production: https://hg.mozilla.org/build/buildbot-configs/rev/ca4c43dc230c

Comment 7

[Chris Manchester (limited bugmail, email directly)]

Everything looks ok for these on treeherder and in logs, thanks everyone!