Closed Bug 963197 (b-2008-ix-0001) Opened 10 years ago Closed 9 years ago

b-2008-ix-0001 problem tracking

Categories

(Infrastructure & Operations Graveyard :: CIDuty, task)

task
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: slaveapi, Unassigned)

References

Details

      No description provided.
This shouldn't have been opened because SlaveAPI managed to reboot through IPMI. I filed bug 96204 to fix this.
Status: NEW → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
Assignee: nobody → armenzg
Depends on: 933768
Assignee: armenzg → nobody
QA Contact: armenzg → bugspam.Callek
Status: RESOLVED → REOPENED
Depends on: 1063018
Resolution: FIXED → ---
seems slave is still taking jobs while disabled
https://treeherder.mozilla.org/ui/logviewer.html#?job_id=193348&repo=mozilla-beta

 LINK : fatal error LNK1123: failure during conversion to COFF: file invalid or corrupt
Re-imaged and returned to production.
Status: REOPENED → RESOLVED
Closed: 10 years ago9 years ago
Resolution: --- → FIXED
Depends on: 1122975
Disabled for the various ways it falls under bug 1122975.
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Depends on: 1125887
Reenabled with 8GB.
Status: REOPENED → RESOLVED
Closed: 9 years ago9 years ago
Resolution: --- → FIXED
Nice work, slave, timed out in the "set props: purge_target" step.

Disabled.
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Interesting, its partner in crime, b-2008-ix-0005, also failed purging on its first job, but managed to show some output so I could see it was stuck having to remove dozens of objdirs, so I let it try again, and after one more failure it has managed several successful builds.

Entirely possible that's saying we have some which not only had too little RAM but also have too slow disk, but it might just be saying that we cannot leave a Windows build slave disabled for two weeks and then put it back in production without either reimaging or manually removing all the expired objdirs, because they just aren't fast enough at removing them to get through several dozen during a production job.
I think it's the latter in that all our builddirs got 'old' and purge is trying to collect them all. I believe there is an effort to change purge behaviour but for now, I'll re-image this slave
Flags: needinfo?(jlund)
Re-imaged and returned to production.
Status: REOPENED → RESOLVED
Closed: 9 years ago9 years ago
Flags: needinfo?(jlund)
Resolution: --- → FIXED
Status: RESOLVED → REOPENED
Depends on: 1165457
Resolution: FIXED → ---
No longer depends on: 1165457
Depends on: 1165771
Re-imaged slave, did not enable it in slavealloc.
Status: REOPENED → RESOLVED
Closed: 9 years ago9 years ago
Resolution: --- → FIXED
Product: Release Engineering → Infrastructure & Operations
Product: Infrastructure & Operations → Infrastructure & Operations Graveyard
You need to log in before you can comment on or make changes to this bug.