Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-1583

Aborting a job in progress aborts the next build in queue after it starts

    • Icon: Bug Bug
    • Resolution: Fixed
    • Icon: Major Major
    • other
    • None
    • Platform: All, OS: All

      One job is running that I wish to cancel, and another is in the queue waiting
      to start. I cancel the running job, and the next job starts. It fails, moments
      later, with an aborted svn update as if I told Hudson to abort the second
      build. The initial aborted build was done with a single click and that build
      has a grey status orb and logs indicating it was aborted. The abort during the
      svn update of the next build causes the second build to fail. (evidenced by the
      subversion update stacktrace:
      org.tmatesoft.svn.core.SVNCancelException: svn: Operation cancelled
      at ... ... ...

      Reproducible in 1.200 and 1.210

          [JENKINS-1583] Aborting a job in progress aborts the next build in queue after it starts

          Can you please post the entire stack trace?

          Kohsuke Kawaguchi added a comment - Can you please post the entire stack trace?

          Code changed in hudson
          User: : kohsuke
          Path:
          trunk/www/changelog.html
          http://fisheye4.cenqua.com/changelog/hudson/?cs=8828
          Log:
          preventive fix for issue #1583

          SCM/JIRA link daemon added a comment - Code changed in hudson User: : kohsuke Path: trunk/www/changelog.html http://fisheye4.cenqua.com/changelog/hudson/?cs=8828 Log: preventive fix for issue #1583

          Code changed in hudson
          User: : kohsuke
          Path:
          trunk/hudson/main/core/src/main/java/hudson/model/Executor.java
          http://fisheye4.cenqua.com/changelog/hudson/?cs=8827
          Log:
          preventive fix for issue #1583

          SCM/JIRA link daemon added a comment - Code changed in hudson User: : kohsuke Path: trunk/hudson/main/core/src/main/java/hudson/model/Executor.java http://fisheye4.cenqua.com/changelog/hudson/?cs=8827 Log: preventive fix for issue #1583

          jmmckinnon added a comment -

          Sure - sorry about that, I thought it was just the standard stacktrace for if
          abort is pressed while subversion is doing its thing.

          org.tmatesoft.svn.core.SVNCancelException: svn: Operation cancelled
          at
          hudson.scm.SubversionUpdateEventHandler.checkCancelled(SubversionUpdateEventHandler.java:191)
          at
          org.tmatesoft.svn.core.wc.SVNBasicClient.checkCancelled(SVNBasicClient.java:334)
          at
          org.tmatesoft.svn.core.internal.wc.admin.SVNWCAccess.checkCancelled(SVNWCAccess.java:75)
          at
          org.tmatesoft.svn.core.internal.wc.admin.SVNWCAccess.doOpen(SVNWCAccess.java:359)
          at
          org.tmatesoft.svn.core.internal.wc.admin.SVNWCAccess.open(SVNWCAccess.java:261)
          at
          org.tmatesoft.svn.core.internal.wc.admin.SVNWCAccess.openAnchor(SVNWCAccess.java:153)
          at
          org.tmatesoft.svn.core.wc.SVNUpdateClient.doUpdate(SVNUpdateClient.java:145)
          at hudson.scm.SubversionSCM$CheckOutTask.invoke(SubversionSCM.java:380)
          at hudson.scm.SubversionSCM$CheckOutTask.invoke(SubversionSCM.java:352)
          at hudson.FilePath.act(FilePath.java:312)
          at hudson.scm.SubversionSCM.checkout(SubversionSCM.java:346)
          at hudson.scm.SubversionSCM.checkout(SubversionSCM.java:299)
          at hudson.model.AbstractProject.checkout(AbstractProject.java:566)
          at
          hudson.model.AbstractBuild$AbstractRunner.checkout(AbstractBuild.java:244)
          at hudson.model.AbstractBuild$AbstractRunner.run(AbstractBuild.java:218)
          at hudson.model.Run.run(Run.java:751)
          at hudson.model.Build.run(Build.java:102)
          at hudson.model.ResourceController.execute(ResourceController.java:70)
          at hudson.model.Executor.run(Executor.java:71)
          Recording test results
          finished: FAILURE

          I got familiar with that early on and then learned to not interrupt a build
          intentionally while it was still doing an update (svn cleanup can be a pain
          sometimes)

          jmmckinnon added a comment - Sure - sorry about that, I thought it was just the standard stacktrace for if abort is pressed while subversion is doing its thing. org.tmatesoft.svn.core.SVNCancelException: svn: Operation cancelled at hudson.scm.SubversionUpdateEventHandler.checkCancelled(SubversionUpdateEventHandler.java:191) at org.tmatesoft.svn.core.wc.SVNBasicClient.checkCancelled(SVNBasicClient.java:334) at org.tmatesoft.svn.core.internal.wc.admin.SVNWCAccess.checkCancelled(SVNWCAccess.java:75) at org.tmatesoft.svn.core.internal.wc.admin.SVNWCAccess.doOpen(SVNWCAccess.java:359) at org.tmatesoft.svn.core.internal.wc.admin.SVNWCAccess.open(SVNWCAccess.java:261) at org.tmatesoft.svn.core.internal.wc.admin.SVNWCAccess.openAnchor(SVNWCAccess.java:153) at org.tmatesoft.svn.core.wc.SVNUpdateClient.doUpdate(SVNUpdateClient.java:145) at hudson.scm.SubversionSCM$CheckOutTask.invoke(SubversionSCM.java:380) at hudson.scm.SubversionSCM$CheckOutTask.invoke(SubversionSCM.java:352) at hudson.FilePath.act(FilePath.java:312) at hudson.scm.SubversionSCM.checkout(SubversionSCM.java:346) at hudson.scm.SubversionSCM.checkout(SubversionSCM.java:299) at hudson.model.AbstractProject.checkout(AbstractProject.java:566) at hudson.model.AbstractBuild$AbstractRunner.checkout(AbstractBuild.java:244) at hudson.model.AbstractBuild$AbstractRunner.run(AbstractBuild.java:218) at hudson.model.Run.run(Run.java:751) at hudson.model.Build.run(Build.java:102) at hudson.model.ResourceController.execute(ResourceController.java:70) at hudson.model.Executor.run(Executor.java:71) Recording test results finished: FAILURE I got familiar with that early on and then learned to not interrupt a build intentionally while it was still doing an update (svn cleanup can be a pain sometimes)

          Alan Harder added a comment -

          I tried 2 cases:
          1) interrupting a build after svn update, while working.
          2) interrupting a build during svn update (actually failed with "svn: Svndiff
          data contains corrupt window" instead of getting "aborted")

          In both cases the pending job in the queue started and ran ok.. so, closing this
          issue.. reopen if you still see any problems.

          Alan Harder added a comment - I tried 2 cases: 1) interrupting a build after svn update, while working. 2) interrupting a build during svn update (actually failed with "svn: Svndiff data contains corrupt window" instead of getting "aborted") In both cases the pending job in the queue started and ran ok.. so, closing this issue.. reopen if you still see any problems.

            Unassigned Unassigned
            jmmckinnon jmmckinnon
            Votes:
            0 Vote for this issue
            Watchers:
            0 Start watching this issue

              Created:
              Updated:
              Resolved: