On 7 December 2013 18:09, Hervé BOUTEMY <herve.bout...@free.fr> wrote: > Olivier did a full reboot a few days ago when I reported the same issues, and > it stuck once again
I restarted only Jenkins instance on aegis.a.o. I believe Dan says restarting slaves boxes. Could be a good idea (but I don't have karma for that) > > so rebooting will be short term solution, but it seems the actual problem > stays > > Does anybody have a stack-trace of a stuck Jenkins instance, to report to > Jenkins developpers and look for a real solution? > > I really want this to be fixed, because ASF Jenkins buildfarm is pretty > useless > these days :( > > Regards, > > Hervé > > Le vendredi 6 décembre 2013 17:18:13 Daniel Kulp a écrit : >> On Dec 6, 2013, at 5:14 PM, Dave Brondsema <d...@brondsema.net> wrote: >> Looks like we still have a problem, and the ubuntu backlog has been at 25+ >> jobs >> > for 3 days. A specific example is >> > https://builds.apache.org/job/CXF-2.7-deploy/231/ currently running for >> > 7hr so far. It has an absolute timeout of 187 minutes. I'm sure there >> > are more cases, but that was the first I found. >> >> Personally, I think all of the unbuntu boxes should just be rebooted and see >> if a fresh start would help. The builds that are succeeding are taking >> almost double the normal amount of time which leads me think there are all >> kinds of stuck processes or something consuming resources. >> >> Dan >> >> > -Dave >> > >> > On 11/18/13 6:48 PM, Gavin McDonald wrote: >> >> Olivier Lamy upgraded the Jenkins instance, he seemed to recall a fix for >> >> this. >> >> >> >> Let's see how it runs now. >> >> >> >> Thanks >> >> >> >> Gav... >> >> >> >>> -----Original Message----- >> >>> From: Dave Brondsema [mailto:d...@brondsema.net] >> >>> Sent: Tuesday, 19 November 2013 8:19 AM >> >>> To: builds@apache.org >> >>> Subject: Re: Timeout limit doesn't work >> >>> >> >>> On 11/14/13 2:14 PM, Lukasz Lenart wrote: >> >>>> Why the Absolute Timeout limit doesn't work? For example, the build >> >>>> [1] has timeout limit set to 187 minutes, but that job [1] is running >> >>>> 15h as for now and blocks other builds :\ >> >>>> >> >>>> https://builds.apache.org/job/CXF-2.7-deploy/configure >> >>>> https://builds.apache.org/job/CXF-2.7-deploy/219/ >> >>> >> >>> I'm seeing this as a real problem, too. Looking at jobs currently >> >>> running on the ubuntu hosts, these seem very very long (and still >> >>> going, as of this writing): >> >>> >> >>> https://builds.apache.org/job/Camel.trunk.notest/2041/ >> >>> 16hr, but timeout is 180 min >> >>> >> >>> https://builds.apache.org/job/Qpid-Java-Java-MMS-TestMatrix/1473/ >> >>> 11hr, but timeout is 180 min >> >>> >> >>> https://builds.apache.org/job/cloudstack-master-maven/3348/ >> >>> 6hr, timeout is elastic 150% >> >>> >> >>> https://builds.apache.org/job/Camel.trunk.fulltest/1615/ >> >>> 11hr, but timeout is 365 min >> >>> >> >>> https://builds.apache.org/job/river-qa-refactor-jdk7/113/ >> >>> 1.5hr (previous took 16hr), timeout is 2253 min -- really? >> >>> >> >>> https://builds.apache.org/job/Camel.2.12.x.fulltest/102/ >> >>> 11hr, but timeout is 365 min >> >>> >> >>> This ends up backlogging other jobs: >> >>> https://builds.apache.org/label/ubuntu/load-statistics?type=hour >> >>> >> >>> -- >> >>> Dave Brondsema : d...@brondsema.net >> >>> http://www.brondsema.net : personal >> >>> http://www.splike.com : programming >> >>> >> >>> <>< > -- Olivier Lamy Ecetera: http://ecetera.com.au http://twitter.com/olamy | http://linkedin.com/in/olamy
