Closed Bug 727626 Opened 13 years ago Closed 13 years ago

jetpack hg bridge is broken

Categories

(Add-on SDK Graveyard :: General, defect, P1)

Product:
Add-on SDK Graveyard
Component:
General
Platform:
Other
All
Type:
defect

Tracking

(Not tracked)

Status:
RESOLVED FIXED

People

(Reporter: KWierso, Assigned: warner)

Details

Nothing committed has made its way over to the hg mirror since Monday. Irakli just landed a change that should fix the oranges/reds on tbpl, but we can't get it running until the mirror pulls in that change. I'm guessing the machine that does the mirroring just needs to be smacked or looked at disapprovingly to get it to start doing its job again.
Assignee: nobody → warner-bugzilla
Priority: -- → P1
Uh oh. I'll take a look. I'm in London at the moment but I think I can get into that machine remotely, unless it's really wedged.
Ok, I think I got it unstuck. It was hung in 'git fetch', pulling from github to the local copy. I don't have debug symbols on /usr/bin/git on that machine, so I was unable to find out what exactly it was waiting for (just __read_nocancel() and __kernel_vsyscall() ). But I killed the process and then the cronjob was able to re-acquire the lock and resume. http://hg.mozilla.org/projects/addon-sdk/ seems to match https://github.com/mozilla/addon-sdk now, so I think this should be fixed. Wes: could you keep an eye on this and ping me again if it looks stuck? For reference, it's driven by a */10 cronjob, so it runs every 10 minutes, on the hour, and it usually finishes in few seconds, so if it's not updated by 11 or 12 minutes past the hour, then something is wrong. Also for reference, it was the fetch that occurred at 04:00 AM (pacific) on 14-Feb-2012 that got stuck: things were probably ok until that point. And then I got it to resume at 10:34:43 16-Feb-2012.
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → FIXED
(In reply to Brian Warner [:warner :bwarner] from comment #2) > Wes: could you keep an eye on this and ping me again if it looks stuck? For > reference, it's driven by a */10 cronjob, so it runs every 10 minutes, on > the hour, and it usually finishes in few seconds, so if it's not updated by > 11 or 12 minutes past the hour, then something is wrong. Can do, thanks for the fix.
You need to log in before you can comment on or make changes to this bug.