Closed
Bug 630904
Opened 14 years ago
Closed 14 years ago
release_sanity should wait for reconfig to finish successfully before doing sendchange
Categories
(Release Engineering :: General, defect, P2)
Release Engineering
General
Tracking
(Not tracked)
RESOLVED
FIXED
People
(Reporter: jhford, Assigned: rail)
References
Details
(Whiteboard: [releases][automation])
Attachments
(2 files)
13.50 KB,
patch
|
catlee
:
review+
rail
:
checked-in+
|
Details | Diff | Splinter Review |
15.20 KB,
patch
|
catlee
:
review+
rail
:
checked-in+
|
Details | Diff | Splinter Review |
or it shouldn't do the reconfig at all.
Assignee | ||
Comment 1•14 years ago
|
||
I'd add --reconfig (default=False).
Comment 2•14 years ago
|
||
Personally, I'd like to see it doing reconfigs still...it's a good safeguard to have. But if that's too difficult to do immediately, disabling them is definitely good.
Assignee | ||
Comment 3•14 years ago
|
||
Could we use http://hg.mozilla.org/build/braindump/file/391f2a2e2d74/buildbot-related/buildbot-wrangler.py instead of 'buildbot reconfig' here?
Comment 4•14 years ago
|
||
If it doesn't do the reconfig, it shouldn't do the sendchange either.
Comment 5•14 years ago
|
||
(In reply to comment #4) > If it doesn't do the reconfig, it shouldn't do the sendchange either. I'm not sure I follow....having it do the sendchange is still good IMHO, because it makes it impossible for us to typo in it.
Comment 6•14 years ago
|
||
(In reply to comment #5) > (In reply to comment #4) > > If it doesn't do the reconfig, it shouldn't do the sendchange either. > > I'm not sure I follow....having it do the sendchange is still good IMHO, > because it makes it impossible for us to typo in it. The whole point of release_sanity is to make sure the master is in a good state before kicking things off. If it doesn't do the reconfig, it doesn't know if the files on disk have any relation to what the master is running.
Comment 7•14 years ago
|
||
I thought the point was to reduce opportunity for manual error. In any case, you seem to feel strongly about this so I'm not going to push back. If it's not doing the sendchange it should at least spit out the "buildbot sendchange" line for you to copy and paste from.
Comment 8•14 years ago
|
||
(In reply to comment #7) > I thought the point was to reduce opportunity for manual error. In any case, > you seem to feel strongly about this so I'm not going to push back. If it's not > doing the sendchange it should at least spit out the "buildbot sendchange" line > for you to copy and paste from. I think it should be doing both the reconfig and the sendchange. We just need to add smarts to make sure it waits for the reconfig to finish.
Updated•14 years ago
|
Priority: -- → P3
Whiteboard: [releases][automation]
Assignee | ||
Updated•14 years ago
|
Assignee: nobody → rail
Priority: P3 → P2
Updated•14 years ago
|
Severity: normal → critical
Assignee | ||
Comment 9•14 years ago
|
||
Attachment #516842 -
Flags: review?(catlee)
Assignee | ||
Comment 10•14 years ago
|
||
Attachment #516843 -
Flags: review?(catlee)
Assignee | ||
Comment 11•14 years ago
|
||
Worked fine in staging.
Updated•14 years ago
|
Attachment #516843 -
Flags: review?(catlee) → review+
Updated•14 years ago
|
Attachment #516842 -
Flags: review?(catlee) → review+
Assignee | ||
Comment 12•14 years ago
|
||
Comment on attachment 516843 [details] [diff] [review] Use buildbot-wrangler.py for reconfig http://hg.mozilla.org/build/tools/rev/a1acec433bbb
Attachment #516843 -
Flags: checked-in+
Assignee | ||
Comment 13•14 years ago
|
||
Comment on attachment 516842 [details] [diff] [review] Move buildbot-wrangler.py manage_masters.py master_fabric.py to tools http://hg.mozilla.org/build/braindump/rev/1b5befc79b85
Attachment #516842 -
Flags: checked-in+
Assignee | ||
Updated•14 years ago
|
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
Updated•11 years ago
|
Product: mozilla.org → Release Engineering
You need to log in
before you can comment on or make changes to this bug.
Description
•