| Summary: | Please kill e4 build #106 | ||
|---|---|---|---|
| Product: | Community | Reporter: | Paul Webster <pwebster> |
| Component: | CI-Jenkins | Assignee: | CI Admin Inbox <ci.admin-inbox> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | P3 | CC: | david_williams, d_a_carver, kim.moir |
| Version: | unspecified | ||
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| Whiteboard: | |||
|
Description
Paul Webster
From what I can see, there are no other jobs running under 'hudsonbuild' id (Except the one, main one for the hudson service itself). Therefore I'm assuming hudson is hung "internally" somehow, and needs to be restarted. I'll restart in a few minutes, if no one else speaks up. (In reply to comment #1) > > I'll restart in a few minutes, if no one else speaks up. Thanx David, PW hudson restarted. I see your next e4 test job was queued and started, as expected. I don't know what caused (causes) the error, but if you didn't notice, the last thing it said in the log was something to the effect of "parsing log file for errors". If that is a special function/plugin from hudson that you use specifically to find a a certain type of error, you may want to consider disabling that part of your hudson job configuration, and doing that "scan for errors" using your own ant task ... or similar. I'm just guessing. (In reply to comment #0) > I've tried to stop/kill build 106 (it died late yesterday) a couple of times > but it won't stop, and my new build can't proceed until it does. > > https://build.eclipse.org/hudson/job/eclipse-e4-test/106/ > > Please kill it! :-) > > Kim, any idea why our builds hang like that (for 13 hours) instead of > dying/completing in about 4? > > PW Paul, there is an option in your job's config called: Abort the build if it's stuck. If you check that option and then set the amount of time that the job should take to run, Hudson will try and kill the job if it goes over that threshold. I noticed that the e4-test doesn't have this option set. (In reply to comment #4) > Paul, there is an option in your job's config called: > > Abort the build if it's stuck. > > If you check that option and then set the amount of time that the job should > take to run, Hudson will try and kill the job if it goes over that threshold. > I noticed that the e4-test doesn't have this option set. thanx Dave. We change the configuration to stop parsing the logs (but it might not have taken effect for #107). I'll add the abort if stuck for my next build. PW |