Port bug 1524358, Remove slugid and push comment from .taskcluster.yml
Categories
(Thunderbird :: Build Config, defect)
Tracking
(thunderbird_esr6066+ fixed, thunderbird66 fixed, thunderbird67 fixed)
People
(Reporter: rjl, Assigned: rjl)
References
Details
Attachments
(3 files, 1 obsolete file)
4.22 KB,
patch
|
jorgk-bmo
:
approval-comm-beta+
jorgk-bmo
:
approval-comm-esr60+
|
Details | Diff | Splinter Review |
47 bytes,
text/x-phabricator-request
|
Details | Review | |
47 bytes,
text/x-phabricator-request
|
jorgk-bmo
:
approval-comm-beta+
jorgk-bmo
:
approval-comm-esr60+
|
Details | Review |
Port Bug 1524358 to C-C.
Stop relying on push comment, as_slugid from mozilla-taskcluster
Pushed by thunderbird@calypsoblue.org:
https://hg.mozilla.org/comm-central/rev/f8e238ce124e
Port bug 1524358: remove slugid and commit message from .taskcluster.yml. rs=bustage-fix
Assignee | ||
Comment 2•5 years ago
|
||
[COMM-BETA Approval Request Comment]
Regression caused by (bug #): 1524358
User impact if declined: Builds fail at decision task
Testing completed (on c-c, etc.): yes
Risk to taking this patch (and alternatives if risky):
No risk.
Mozilla-central has already uplifted their changes to beta.
Assignee | ||
Comment 3•5 years ago
|
||
Comment on attachment 9041236 [details] [diff] [review] .taskcluster.yml bustage fix [COMM-ESR60 Approval Request Comment] Regression caused by (bug #): 1524358 User impact if declined: Builds fail at decision task Testing completed (on c-c, etc.): yes Risk to taking this patch (and alternatives if risky): No risk. Mozilla-central plans to uplift this to esr60.
Assignee | ||
Updated•5 years ago
|
Comment 4•5 years ago
|
||
Comment on attachment 9041236 [details] [diff] [review] .taskcluster.yml bustage fix I'll check ESR uplift.
Updated•5 years ago
|
Comment hidden (obsolete) |
Assignee | ||
Comment 6•5 years ago
|
||
Work on bug 1520874 led to failures when pushing to try-comm-central using
the "try:" syntax in a commit message.
This commit changes the call to get_hg_commit_message. Instead of passing
the root path of the Gecko checkout, the full path of "product-dir" is passed.
There's no need to find the actual root as "hg log" will work no matter where
in the tree it is executed.
Updated•5 years ago
|
Assignee | ||
Comment 7•5 years ago
|
||
Pushed by thunderbird@calypsoblue.org: https://hg.mozilla.org/integration/autoland/rev/00f00bf8cdf1 Support retrieval of comm repo commit messages. r=aki
Comment 9•5 years ago
|
||
bugherder |
Comment 11•5 years ago
|
||
Hi! I think this is almost done. I still see
https://hg.mozilla.org/comm-central/file/tip/.taskcluster.yml#l20
_pushComment: {$if: 'tasks_for == "cron" || tasks_for == "action"', then: '', else: {$eval: 'push.comment'}}
but _pushComment
doesn't appear elsewhere in the file. So, I think we can just remove that line. I'll make a patch in a sec.
Comment 12•5 years ago
|
||
Comment is particularly critical, since it is based on push.comment which soon
won't exist.
This also copies some explanatory comments about the context in which this file
is interpreted from mozilla-central.
Comment 13•5 years ago
|
||
Comment 14•5 years ago
|
||
Comment 15•5 years ago
|
||
Pushed by thunderbird@calypsoblue.org:
https://hg.mozilla.org/comm-central/rev/d7cd1486000d
remove unused context variables r=rjl
Assignee | ||
Comment 16•5 years ago
|
||
Comment on attachment 9042238 [details]
Bug 1525072 - remove unused context variables
[COMM-BETA Approval Request Comment]
Regression caused by (bug #): 1524358
User impact if declined: Builds fail at decision task
Testing completed (on c-c, etc.): yes
Risk to taking this patch (and alternatives if risky):
No risk.
This is an additional update to the previous patches, a couple of lines that should have been removed were not.
[COMM-ESR60 Approval Request Comment]
Regression caused by (bug #): 1524358
User impact if declined: Builds fail at decision task
Testing completed (on c-c, etc.): yes
Risk to taking this patch (and alternatives if risky):
No risk.
This is an additional update to the previous patches, a couple of lines that should have been removed were not.
Comment 17•5 years ago
|
||
Comment on attachment 9042238 [details]
Bug 1525072 - remove unused context variables
OK, we check ESR when the time comes, OK?
Comment 18•5 years ago
|
||
Comment 19•5 years ago
|
||
Bug 1524358 landed on ESR 60:
https://hg.mozilla.org/releases/mozilla-esr60/rev/7afa591a19b4
https://hg.mozilla.org/releases/mozilla-esr60/rev/83a4237d5aa1
https://hg.mozilla.org/releases/mozilla-esr60/rev/1180e2a8f401
Updated•5 years ago
|
Updated•5 years ago
|
Comment 20•5 years ago
|
||
Rob, I ported the two patches here (comment #14, comment #18) to ESR and at least the D is green:
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=d04633b2b37fc4e892fe9000fb70fae9bab28009
https://hg.mozilla.org/try-comm-central/graph/d04633b2b37fc4e892fe9000fb70fae9bab28009
Can you please check it. I've also started some builds to see:
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=61701b4c41ba0781338cf48b548de12e8c8ccc54
Comment 21•5 years ago
|
||
If the decision ran correctly, that's pretty much evidence that it worked!
Comment 22•5 years ago
|
||
Thanks for looking, Dustin. I asked because both patches needed rebasing, for example if you compare
beta: https://hg.mozilla.org/releases/comm-beta/rev/7b3232d39834
with
ESR: https://hg.mozilla.org/try-comm-central/rev/044f908fc99b85ba1f46bbd77399dbaa0d53a27c
The second changeset became a "comment only" since the _pushDate:
and _pushComment:
wasn't in the ESR to start with. Strangely the second push:
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=61701b4c41ba0781338cf48b548de12e8c8ccc54
sadly didn't build anything, not even when redone:
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=b893a357341b9c6dd2d4589de68ba3e5f21d572d
Let's see if I get feedback from Rob before m-esr60 at 60.6 becomes available which is when I need to push this.
Comment 23•5 years ago
|
||
Hm, it's possible that's because get_hg_commit_message
gets the commit message for the wrong repo.
taskcluster/taskgraph/decision.py
parameters['message'] = try_syntax_from_message(
get_hg_commit_message(os.path.join(GECKO, product_dir)))
looks like it does the right thing, though.
Comment 24•5 years ago
|
||
Hmm, what you quoted there is for trunk from here, no?
https://hg.mozilla.org/mozilla-central/rev/7d289fe21aec#l1.40
I was pushing from comm-esr60 to try-comm-central. That has worked in the past, I could push from comm-central, comm-beta and comm-esr60, which was very handy. Today I noticed that the builds don't start, see comment #22.
It has already not worked for beta here, 6th of March:
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=e108217840869de16a92eed31603ea3f9b0ad8a1
Last time I did an ESR build that way was here, 9th of Feb:
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=bec21e91a896bec6f5dd515ed0a4634b94262012
I should file a new bug for that.
Comment 25•5 years ago
|
||
Filed bug 1534204.
Comment 26•5 years ago
|
||
Ah, I misunderstood the comments above -- it seems that this is not yet ported to esr60. At least, I still see as_slugid and push.comment in https://hg.mozilla.org/releases/comm-esr60/file/tip/.taskcluster.yml.
Comment 27•5 years ago
|
||
Dustin, yes, the changes haven't been committed to c-esr60 yet. There was only a try run so far:
https://hg.mozilla.org/try-comm-central/graph/d04633b2b37fc4e892fe9000fb70fae9bab28009
Once landed on c-esr60 in the next couple of days, there will only be as_slugid
left which is also still used on M-C trunk:
https://searchfox.org/mozilla-central/search?q=as_slugid&case=false®exp=false&path=
Comment 28•5 years ago
|
||
TB 60.6 ESR:
https://hg.mozilla.org/releases/comm-esr60/rev/e009d8709076af4255db5939102d5043b3be118e
https://hg.mozilla.org/releases/comm-esr60/rev/c3b3a117ac21d24883488375f4e41f12c8ac888a
Updated•5 years ago
|
Description
•