Windows, so of course the first thing I did when I saw 161 pending jobs and only 6 slaves that had done jobs in the past 5 hours was reboot, but no joy. Can't be a coincidence that the 6 working slaves are the 5 slaves from bug 1164214 and the one from bug 989531, but since runner-on-Windows apparently didn't roll out today, I don't have any other good theory about what we rolled out that only works for try slaves which have been reimaged in the last month, and causes all others to stop doing work.
Runner was deployed to the hosts in question (b-2008-ix-**** in try). The problem appears to be my failure yesterday to disable mozilla-build/start-builbot.bat which has now been corrected. Working on getting the slaves to start accepting queued jobs now.
Runner roll-out reverted for all of wintry. Slaves are picking up jobs and states are changing ('broken' -> 'working').
Seems to have been resolved by roll-back.
Status: NEW → RESOLVED
Last Resolved: 3 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.