On Tue, Feb 6, 2018 at 10:27 AM, Robert Varga <n...@hq.sk> wrote: > On 06/02/18 16:09, Tom Pantelis wrote: > > It's not just merge jobs with random failures. Can anyone explain why > > this AR job > > failed: https://jenkins.opendaylight.org/releng/job/ > openflowplugin-validate-autorelease-oxygen/227/console? > > > > 13:49:24 [INFO] > > ------------------------------------------------------------------------ > > 13:49:24 [INFO] Reactor Summary: > > 13:49:24 [INFO] > > 13:49:24 [INFO] mdsal-artifacts .................................... > > SUCCESS [ 35.062 s] > > 13:49:24 [INFO] mdsal-model-artifacts .............................. > > SUCCESS [ 34.991 s] > > 13:49:24 [INFO] config-artifacts ................................... > > SUCCESS [ 42.971 s] > > 13:49:24 [INFO] yang-test-plugin ................................... > > SUCCESS [01:13 min] > > 13:49:24 [INFO] mdsal-artifacts .................................... > > SUCCESS [ 43.020 s] > > 13:49:24 [INFO] autorelease-validate-projects ...................... > > SUCCESS [ 35.985 s] > > 13:49:24 [INFO] > > ------------------------------------------------------------------------ > > 13:49:24 [INFO] BUILD SUCCESS > > 13:49:24 [INFO] > > ------------------------------------------------------------------------ > > 13:49:24 [INFO] Total time: 01:41 min (Wall Clock) > > 13:49:24 [INFO] Finished at: 2018-02-06T13:49:24+00:00 > > 13:49:24 [INFO] Final Memory: 51M/309M > > > > > > It all indicates success but still mysteriously fails. I don't see any > > reason why in the output. > > 14:49:24 [openflowplugin-validate-autorelease-oxygen] $ > /w/tools/hudson.tasks.Maven_MavenInstallation/apache-maven-3.3.9/bin/mvn > -f pom.xml > [...] > 14:49:25 [INFO] Scanning for projects... > 14:50:36 Build step 'Invoke top-level Maven targets' marked build as > failure > 14:50:36 $ ssh-agent -k > 14:50:37 unset SSH_AUTH_SOCK; > 14:50:37 unset SSH_AGENT_PID; > 14:50:37 echo Agent pid 8247 killed; > 14:50:37 [ssh-agent] Stopped. > 14:50:37 [PostBuildScript] - Execution post build scripts. > > Jenkins slave node died? > > I have seen overcommit errors (and hence JVM aborts) in other > validate-autorelease jobs, so now I am just ignoring their results. >
That's a subtle failure. I agree - AR jobs are pretty much useless these days with the instability - I think we should just disable them. Maybe dist-check jobs too. It's lucky enough just to get a verify job to succeed these days :) > > Bye, > Robert > >
_______________________________________________ controller-dev mailing list controller-dev@lists.opendaylight.org https://lists.opendaylight.org/mailman/listinfo/controller-dev