1160845 - No Firefox linux nightly l10n builds since April 30

Status: RESOLVED FIXED

(Release Engineering :: General, defect)

Description

[Pascal Chevrel:pascalc]

Last Firefox linux l10n build in http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/latest-mozilla-central-l10n/ has been built on April 30

There were en-US nightly builds in the last days.

Comment 1

[Phil Ringnalda (:philor)]

https://treeherder.mozilla.org/logviewer.html#?job_id=1426982&repo=mozilla-central

Comment 2

[Dustin J. Mitchell [:dustin] (he/him)]

Picking a random spot instance:

[[e-mail] ~]# ls -al /builds/relengapi.tok
-rw-r--r-- 1 root root 115 May  3 01:20 /builds/relengapi.tok

So either (1) this is done in mock without reflecting that file into the mock, or (b) this is done on an old AMI for some reason.

Comment 3

[Dustin J. Mitchell [:dustin] (he/him)]

Not run in mock:

 argv: ['/tools/buildbot/bin/python', 'scripts/scripts/desktop_l10n.py', '--branch-config', 'single_locale/mozilla-central.py', '--platform-config', 'single_locale/linux64.py', '--environment-config', 'single_locale/production.py', '--balrog-config', 'balrog/production.py', '--total-chunks', '3', '--this-chunk', '1'] 

the instance is long gone, so I can't see the AMI used.  It would seem odd to revert to an old AMI just for l10n's though.

More matching logs might help.

Comment 4

[Dustin J. Mitchell [:dustin] (he/him)]

The failure is in mock, though -- maybe I'm misreading the logs and this is run from mock.

Notably, the mock runs mirror /builds/gapi.data into the mock environment, so this may be necessary for relengapi.tok too.

Comment 5

[Dustin J. Mitchell [:dustin] (he/him)]

Attachment: MozReview Request: bz://1160845/dustin

/r/8079 - Bug 1160845: reflect relengapi.tok into mock environments; r=mgerva

Pull down this commit:

hg pull -r 941bd99e080e6f73eaf95120492b602f0b7d676f https://reviewboard-hg.mozilla.org/build-mozharness

Comment 6

[Dustin J. Mitchell [:dustin] (he/him)]

Comment on attachment 8600879 [details]
MozReview Request: bz://1160845/dustin

remote:   https://hg.mozilla.org/build/mozharness/rev/941bd99e080e

Comment 7

[Dustin J. Mitchell [:dustin] (he/him)]

Comment on attachment 8600879 [details]
MozReview Request: bz://1160845/dustin

https://reviewboard.mozilla.org/r/8077/#review6859

Ship It!

Comment 8

[Dustin J. Mitchell [:dustin] (he/him)]

I think this is finished, but I think that running a rebuild will use the old code, so we need to wait for another fresh nightly.

Comment 9

[Massimo Gervasini [:massimo]]

Attachment: [mozharness] Bug 1160845 - add relengapi.tok to l10n mock_files for linux.patch

last l10n log: http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/2015/05/2015-05-05-03-02-06-mozilla-central-l10n/mozilla-central-linux-l10n-nightly-1-unknown-bm77-build1-build6.txt.gz

We need relengapi.tok for linux 32 as well.

Comment 10

[Dustin J. Mitchell [:dustin] (he/him)]

Comment on attachment 8601501 [details] [diff] [review]
[mozharness] Bug 1160845 - add relengapi.tok to l10n mock_files for linux.patch

I'm not sure how my grep missed that

Comment 11

[Massimo Gervasini [:massimo]]

Comment on attachment 8601501 [details] [diff] [review]
[mozharness] Bug 1160845 - add relengapi.tok to l10n mock_files for linux.patch

landed: https://hg.mozilla.org/build/mozharness/rev/e47abb2ed057

Comment 12

[Massimo Gervasini [:massimo]]

landed on production branch: https://hg.mozilla.org/build/mozharness/rev/565ea3ec1d4d

Comment 13

[Dustin J. Mitchell [:dustin] (he/him)]

I haven't heard anything, so let's call this good.

Comment 15

[Dustin J. Mitchell [:dustin] (he/him)]

Attachment: MozReview Request: Bug 1160845: reflect relengapi.tok into mock environments; r=mgerva