• Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Major Major
    • integrity-plugin
    • None
    • Jenkins 10.9
      integrity plugin 2.0.1

      We got some pipeline jobs which are impossible to kill. After pressing the red 'x' Button we get the following output:

      [...]
      
      [Pipeline] }
      [Pipeline] // node
      [Pipeline] }
      [Pipeline] // stage
      [Pipeline] End of Pipeline
      Finished: SUCCESS
      Resuming build at Wed Feb 15 07:47:17 CET 2017 after Jenkins restart
      Ready to run at Wed Feb 15 07:47:19 CET 2017
      Resuming build at Wed Feb 15 07:50:34 CET 2017 after Jenkins restart
      Ready to run at Wed Feb 15 07:50:35 CET 2017
      Aborted by User
      Click here to forcibly terminate running steps
      

      The main problem is, that the build succeeded but tried to restart after a jenkins restart. Clicking the 'terminate' message does not work. Are they suggestions how to kill those 'zombie' pipelines?

          [JENKINS-42049] How to kill unstoppable zombie pipeline jobs

          Simon Beyer added a comment -

          Actually about 2-3 hours after posting this there was a Pipeline Plugin update. Now there are more options (and finally) any response after hitting the "kill" button. Now I can kill the process. The handling is kind of buggy but it works.

          Simon Beyer added a comment - Actually about 2-3 hours after posting this there was a Pipeline Plugin update. Now there are more options (and finally) any response after hitting the "kill" button. Now I can kill the process. The handling is kind of buggy but it works.

          Simon Beyer added a comment -

          Newest pipeline job fixed the inaction of the "kill" button

          Simon Beyer added a comment - Newest pipeline job fixed the inaction of the "kill" button

          Andrey Beletsky added a comment - - edited

          I have an unstoppable job that has been running for 7 days. I couldn't stop it via UI, by Jenkins restart.

          Job failed with " No space left on device" error and continue has status "running". 

          See   and 

          Jenkins 2.46.1

          Andrey Beletsky added a comment - - edited I have an unstoppable job that has been running for 7 days. I couldn't stop it via UI, by Jenkins restart. Job failed with " No space left on device" error and continue has status "running".  See   and  Jenkins 2.46.1

          Ilia Shakitko added a comment -

          Same with me, but "killing/stopping" a hanged job doesn't work - no reaction...

          Ilia Shakitko added a comment - Same with me, but "killing/stopping" a hanged job doesn't work - no reaction...

          I could remove this job only via "Manage Jenkins" > "Script Console". It easily can be googled how to do it.

          Andrey Beletsky added a comment - I could remove this job only via "Manage Jenkins" > "Script Console". It easily can be googled how to do it.

          Ilia Shakitko added a comment -

          abeletsky Removing (deleting) job folder can be done even physically connecting to Jenkins machine and deleting it with rm -rf

          But googling is not helping to understanding how to prevent the issue neither as kill/stop a hanged job. Especially if there is an ".nfs8947134" file is created and locked withing a job.

          There are various root causes for this issue and some of them are resolvable doing stuff in Script console, but some of them are not.

          Therefore I suggest to upvote this issue and get some attention to fix it, instead of "easily googling" advise to everyone.

          Ilia Shakitko added a comment - abeletsky  Removing (deleting) job folder can be done even physically connecting to Jenkins machine and deleting it with rm -rf .  But googling is not helping to understanding how to prevent the issue neither as kill/stop a hanged job. Especially if there is an ".nfs8947134" file is created and locked withing a job. There are various root causes for this issue and some of them are resolvable doing stuff in Script console, but some of them are not. Therefore I suggest to upvote this issue and get some attention to fix it, instead of "easily googling" advise to everyone.

          Agree, it is why I voted. I think it is the only thing I can do.

          Andrey Beletsky added a comment - Agree, it is why I voted. I think it is the only thing I can do.

          Voted, I have same recurrent issue  on my installation, and it's getting worse and worse...

          Frédéric Meyrou added a comment - Voted, I have same recurrent issue  on my installation, and it's getting worse and worse...

          Gail Stewart added a comment -

          I have two zombie jobs after the jenkins server ran out of disk space.  Nothing has been able to clear the jobs.  I am running Jenkins ver. 2.102 with these plugins

          Pipeline: Job  2.16
          Pipeline: Basic Steps: 2.6
          Pipeline: Declarative Extension Points API 1.2.6
          Docker Pipeline 1.14

          I have attempted to kill threads, but not of the "readily googled" methods have even been able to find the threads, Multiple restarts of jenkins have not solved the problem. Attempting to stop the job in the console (again) is not giving me the links to "force" stopping the pipeline job.

          Gail Stewart added a comment - I have two zombie jobs after the jenkins server ran out of disk space.  Nothing has been able to clear the jobs.  I am running Jenkins ver. 2.102 with these plugins Pipeline: Job  2.16 Pipeline: Basic Steps: 2.6 Pipeline: Declarative Extension Points API 1.2.6 Docker Pipeline 1.14 I have attempted to kill threads, but not of the "readily googled" methods have even been able to find the threads, Multiple restarts of jenkins have not solved the problem. Attempting to stop the job in the console (again) is not giving me the links to "force" stopping the pipeline job.

          Daniel Ruiz added a comment -

          you can try going to the jenkins script console and execute:

          Jenkins.instance.getItemByFullName("YOURJOBFULLNAME").getBuildByNumber(YOURJOBBUILDNUMBER).finish(hudson.model.Result.ABORTED, new java.io.IOException("Aborting build"));

           

          if it's not working, the only way I found it's deleting the build number folder inside the builds of the job, on master jobs folder. and restarting jenkins. It's a horrible solution, but ...

           

           

          Daniel Ruiz added a comment - you can try going to the jenkins script console and execute: Jenkins.instance.getItemByFullName("YOURJOBFULLNAME").getBuildByNumber(YOURJOBBUILDNUMBER).finish(hudson.model.Result.ABORTED, new java.io.IOException("Aborting build"));   if it's not working, the only way I found it's deleting the build number folder inside the builds of the job, on master jobs folder. and restarting jenkins. It's a horrible solution, but ...    

            swanh Simon Beyer
            swanh Simon Beyer
            Votes:
            18 Vote for this issue
            Watchers:
            17 Start watching this issue

              Created:
              Updated: