611092 - Trigger working mobile unit tests jobs on the Talos pool of slaves

Status: RESOLVED FIXED

(Release Engineering :: General, defect)

Description

[John Ford [:jhford] CET/CEST Berlin Time]

Attachment: [wip] buildbotcustom v1

Now that we have infrastructure set up from 511282, we should start triggering these tests.

Things that need to happen in this bug include:
-sendchanges
-ensuring results end up on the correct tinderbox tree

The attached patch includes the buildbot factory changes required to get this going.

Comment 1

[John Ford [:jhford] CET/CEST Berlin Time]

Attachment: [wip] buildbot-configs v1

Comment 2

[Mark Finkle (:mfinkle) (use needinfo?)]

Running reliable tests on desktop will be crucial for the next Fennec release.
The current device based tests are far to unreliable. Until we get more
reliable testing infra (from hardware issues to flakey tests themselves) we'll
need the desktop-based tests. They can, and do, go green and will help catch
real issues.

Comment 3

[John Ford [:jhford] CET/CEST Berlin Time]

Attachment: buildbot-configs v1

Comment 4

[John Ford [:jhford] CET/CEST Berlin Time]

Attachment: buildbotcustom v1

Comment 5

[John Ford [:jhford] CET/CEST Berlin Time]

Attachment: dump.diff

I slightly changed the logic of the sendchange stuff, this patch shows that there are no regressions

Comment 6

[Aki Sasaki (not active)]

Comment on attachment 527570 [details] [diff] [review]
buildbot-configs v1

>diff --git a/mozilla-tests/config.py b/mozilla-tests/config.py
>--- a/mozilla-tests/config.py
>+++ b/mozilla-tests/config.py
>@@ -198,9 +198,9 @@ UNITTEST_SUITES = {
>         #                     'mochitest-ipcplugins']),
>         ('mobile-mochitest-browser-chrome', ['mobile-mochitest-browser-chrome']),
>         #('reftest', ['reftest']),
>-        ('crashtest', ['crashtest']),
>+        #('crashtest', ['crashtest']),
>         #('xpcshell', ['xpcshell']),
>-        ('jsreftest', ['jsreftest']),
>+        #('jsreftest', ['jsreftest']),
>     ],
> 
> }

I'm going to guess this is cruft?
r=me with this backed out (only r+ the mozilla/config.py changes).

Comment 7

[John Ford [:jhford] CET/CEST Berlin Time]

> I'm going to guess this is cruft?
> r=me with this backed out (only r+ the mozilla/config.py changes).

Actually, that is needed to make sure that crashtest and jsreftest don't get run.

Comment 8

[Aki Sasaki (not active)]

Comment on attachment 527571 [details] [diff] [review]
buildbotcustom v1

This will get cleaner once John Ford fixes bug 557260.

Comment 9

[Aki Sasaki (not active)]

Comment on attachment 527570 [details] [diff] [review]
buildbot-configs v1

n/m, r+ for the whole patch.

Comment 10

[Aki Sasaki (not active)]

Comment on attachment 527573 [details] [diff] [review]
dump.diff

I don't know we need these on all these branches, and talos-staging-master02 has gone away, but other than that I think this looks good.

Comment 11

[John Ford [:jhford] CET/CEST Berlin Time]

This landed last week.

http://hg.mozilla.org/build/buildbot-configs/rev/4324165089ee
http://hg.mozilla.org/build/buildbotcustom/rev/aafb13808c01

Comment 12

[John Ford [:jhford] CET/CEST Berlin Time]

Does anyone remember what the issues was the last time we tried this?

Comment 13

[John Ford [:jhford] CET/CEST Berlin Time]

(In reply to comment #12)
> Does anyone remember what the issues was the last time we tried this?

sorry, wrong bug.