Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-48387

No job should be taken as a reference if it is triggered by Gerrit

XMLWordPrintable

      The plugin is configured the following way:

      • 'Compute new warnings' is activated
      • 'Only use stable builds as reference' is activated

      The behaviour of the plugin is incorrect when following development workflow is used:

      • developer A uploads a change 'a', patch set 1 to Gerrit that fixes N warnings.
      • the job for PS1 passes in Jenkins, since we have less warnings, and this build becomes the new reference
      • developer A upload for change 'a' a patch set 2 to Gerrit where the initial N warnings are present again (he had removed them by 'mistake').
      • the job for PS2 fails, since it refers to PS1 as a reference

       

      In general, the same effect can be seen in various combinations: developer A and developer B upload at roughly the same time and can be blocked since e.g. developer B's job will reference non yet merged change of developer A.

      An option should be added so that no passing build that has GERRIT_CHANGE_NUMBER parameter (or similar) is taken as a reference.

          [JENKINS-48387] No job should be taken as a reference if it is triggered by Gerrit

          chanti vlad created issue -

          Ulli Hafner added a comment -

          This is already possible: set checkbox 'Use previous build as reference'.

          Ulli Hafner added a comment - This is already possible: set checkbox 'Use previous build as reference'.
          Ulli Hafner made changes -
          Resolution New: Fixed [ 1 ]
          Status Original: Open [ 1 ] New: Resolved [ 5 ]
          chanti vlad made changes -
          Description Original: I would like to have a job which sends a warning mail whenever a new checkstyle warning appears in a project. However, this job should not be unstable or failing after this one new build because of the warning. With the current implementation this is not possible as the build will be failing or unstable until that warning is fixed again. A simple way to achieve this behavior is to always use the latest build as the reference build, so that old warning are bascially ignored in the new running build. New: The plugin is configured the following way:
           * 'Compute new warnings' is activated
           * 'Only use stable builds as reference' is activated

          The behaviour of the plugin is incorrect when following development workflow is used:
           * developer A uploads a change 'a', patch set 1 to Gerrit that fixes N warnings.
           * the job for PS1 passes in Jenkins, since we have less warnings, and this build becomes the new reference
           * developer A upload for change 'a' a patch set 2 to Gerrit where the initial N warnings are present again (he had removed them by 'mistake').
           * the job for PS2 fails, since it refers to PS1 as a reference

           

          In general, the same effect can be seen in various combinations: developer A and developer B upload at roughly the same time and can be blocked since e.g. developer B's job will reference non yet merged change of developer A.

          An option should be added so that no passing build that has GERRIT_CHANGE_NUMBER parameter (or similar) is taken as a reference.
          Labels New: gerrit gerrit-trigger
          Priority Original: Minor [ 4 ] New: Major [ 3 ]
          Summary Original: CLONE - Add switch to set reference build always as the latest build New: No job should be taken as a reference if it is triggered by Gerrit

          chanti vlad added a comment -

          Well the closing was fast 

          PLease see my updated description.

          chanti vlad added a comment - Well the closing was fast  PLease see my updated description.
          chanti vlad made changes -
          Resolution Original: Fixed [ 1 ]
          Status Original: Resolved [ 5 ] New: Reopened [ 4 ]
          Ulli Hafner made changes -
          Link New: This issue duplicates JENKINS-13056 [ JENKINS-13056 ]

          Ulli Hafner added a comment -

          Can you please check if this is different from JENKINS-13056?

          Ulli Hafner added a comment - Can you please check if this is different from JENKINS-13056 ?

          chanti vlad added a comment -

          Yes this sounds about right. Thanks for the pointer, i did not find this similar issue in my search.

          chanti vlad added a comment - Yes this sounds about right. Thanks for the pointer, i did not find this similar issue in my search.
          chanti vlad made changes -
          Resolution New: Duplicate [ 3 ]
          Status Original: Reopened [ 4 ] New: Closed [ 6 ]

            drulli Ulli Hafner
            chantivlad chanti vlad
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

              Created:
              Updated:
              Resolved: