Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-6576

Perforce plugin: Continuous jobs when using new View Mask feature.

    XMLWordPrintable

    Details

    • Similar Issues:

      Description

      ------------------
      My view is:
      //depot/IT_Engineering/experimental/user/cpatrick/...

      My View Mask is:
      //depot/IT_Engineering/experimental/user/cpatrick/ami/...

      I've checked "Use mask when syncing".

      I'm polling once per minute. Every minute the job trips again.
      ------------------

      Actually, when I unchecked both polling options for the mask, the job stopped getting tripped. But when I checked "Use mask when polling" the job started running again every minute without my even checking anything in. The file I had checked in originally should not even have been in the mask for polling the second time around. So I'm pretty stumped what might be causing this behavior.

      I've tried this a few different times now. It seems that, if either checkbox is checked the job runs every minute whether I check anything in or not. If the masking options is unchecked the job seems to be launched only when something gets checked in.

      Here is the polling log:
      -------------
      Started on May 19, 2010 6:24:28 PM
      Looking for changes...
      Using master perforce client: testPerforcePlugin
      [workspace] $ p4 workspace -o testPerforcePlugin
      Saving modified client testPerforcePlugin
      [workspace] $ p4 -s client -i
      Last sync'd change was 370843
      [workspace] $ p4 counter change
      [workspace] $ p4 -s changes //depot/IT_Engineering/experimental/user/cpatrick/ami/...@370843,@370859
      Triggering a build.
      Done. Took 0.31 sec
      Changes found
      ----------------

        Attachments

          Activity

          Hide
          akmartin akmartin added a comment -

          In one case it appears to be reporting the last changeset for the view and not for the view mask. (In all cases view mask is a subdir of the view.)

          In another, they are the same it correctly reports no changes found.

          In another and confusing case it is reporting the wrong last synced changeset. 23051 is the last synced changeset for the view and the view mask:

          Started on Jun 8, 2010 2:01:33 PM
          Looking for changes...
          Using master perforce client: dev_amartin_hudson_lms-4.4.0-release_deploy
          [workspace] $ "C:\Program Files\Perforce\p4.exe" workspace -o dev_amartin_hudson_lms-4.4.0-release_deploy
          Saving modified client dev_amartin_hudson_lms-4.4.0-release_deploy
          [workspace] $ "C:\Program Files\Perforce\p4.exe" -s client -i
          Last sync'd change was 23030
          [workspace] $ "C:\Program Files\Perforce\p4.exe" counter change
          [workspace] $ "C:\Program Files\Perforce\p4.exe" -s changes //apg/prod/sam/lms/4.4.0/release/allsrc/...@23030,@23058
          Latest submitted change selected by workspace is 23051
          Triggering a build.
          Done. Took 0.26 sec
          Changes found

          And in another interesting case it is reporting a changelist that isn't in the view or the mask (actual view: 22491, actual mask: 22379):

          Started on Jun 8, 2010 2:06:32 PM
          Looking for changes...
          Using master perforce client: dev_amartin_hudson_lms-4.5.1-main_test
          [workspace] $ "C:\Program Files\Perforce\p4.exe" workspace -o dev_amartin_hudson_lms-4.5.1-main_test
          Saving modified client dev_amartin_hudson_lms-4.5.1-main_test
          [workspace] $ "C:\Program Files\Perforce\p4.exe" -s client -i
          Last sync'd change was 22491
          [workspace] $ "C:\Program Files\Perforce\p4.exe" counter change
          [workspace] $ "C:\Program Files\Perforce\p4.exe" -s changes //apg/prod/sam/lms/4.5.1/main/allsrc/...@22491,@23059
          Triggering a build.
          Done. Took 0.28 sec
          Changes found

          I hope I'm not totally overloading you with data, it just seems inconsistent in ways only the developer would understand so I wanted to give you the best data I could.

          Show
          akmartin akmartin added a comment - In one case it appears to be reporting the last changeset for the view and not for the view mask. (In all cases view mask is a subdir of the view.) In another, they are the same it correctly reports no changes found. In another and confusing case it is reporting the wrong last synced changeset. 23051 is the last synced changeset for the view and the view mask: Started on Jun 8, 2010 2:01:33 PM Looking for changes... Using master perforce client: dev_amartin_hudson_lms-4.4.0-release_deploy [workspace] $ "C:\Program Files\Perforce\p4.exe" workspace -o dev_amartin_hudson_lms-4.4.0-release_deploy Saving modified client dev_amartin_hudson_lms-4.4.0-release_deploy [workspace] $ "C:\Program Files\Perforce\p4.exe" -s client -i Last sync'd change was 23030 [workspace] $ "C:\Program Files\Perforce\p4.exe" counter change [workspace] $ "C:\Program Files\Perforce\p4.exe" -s changes //apg/prod/sam/lms/4.4.0/release/allsrc/...@23030,@23058 Latest submitted change selected by workspace is 23051 Triggering a build. Done. Took 0.26 sec Changes found And in another interesting case it is reporting a changelist that isn't in the view or the mask (actual view: 22491, actual mask: 22379): Started on Jun 8, 2010 2:06:32 PM Looking for changes... Using master perforce client: dev_amartin_hudson_lms-4.5.1-main_test [workspace] $ "C:\Program Files\Perforce\p4.exe" workspace -o dev_amartin_hudson_lms-4.5.1-main_test Saving modified client dev_amartin_hudson_lms-4.5.1-main_test [workspace] $ "C:\Program Files\Perforce\p4.exe" -s client -i Last sync'd change was 22491 [workspace] $ "C:\Program Files\Perforce\p4.exe" counter change [workspace] $ "C:\Program Files\Perforce\p4.exe" -s changes //apg/prod/sam/lms/4.5.1/main/allsrc/...@22491,@23059 Triggering a build. Done. Took 0.28 sec Changes found I hope I'm not totally overloading you with data, it just seems inconsistent in ways only the developer would understand so I wanted to give you the best data I could.
          Hide
          rpetti Rob Petti added a comment -

          I've got a fix for the second case, but the first case still eludes me. It's possible that it may simply be the result of a polling interval that is too short.

          Show
          rpetti Rob Petti added a comment - I've got a fix for the second case, but the first case still eludes me. It's possible that it may simply be the result of a polling interval that is too short.
          Hide
          scm_issue_link SCM/JIRA link daemon added a comment -

          Code changed in hudson
          User: : rpetti
          Path:
          trunk/hudson/plugins/perforce/src/main/java/hudson/plugins/perforce/PerforceSCM.java
          http://jenkins-ci.org/commit/31857
          Log:
          JENKINS-6576 fixing a case when using view masks for polling that may cause builds to be triggered indefinitely.

          Show
          scm_issue_link SCM/JIRA link daemon added a comment - Code changed in hudson User: : rpetti Path: trunk/hudson/plugins/perforce/src/main/java/hudson/plugins/perforce/PerforceSCM.java http://jenkins-ci.org/commit/31857 Log: JENKINS-6576 fixing a case when using view masks for polling that may cause builds to be triggered indefinitely.
          Hide
          akmartin akmartin added a comment -

          Built 1.1.3 tag and installed into Hudson 1.362. I forced a change within the view but outside the view mask and it did not trigger a build, and so far all is quiet--no unexpected builds have popped up. I will continue to monitor for the next day or so and report back.

          Show
          akmartin akmartin added a comment - Built 1.1.3 tag and installed into Hudson 1.362. I forced a change within the view but outside the view mask and it did not trigger a build, and so far all is quiet--no unexpected builds have popped up. I will continue to monitor for the next day or so and report back.
          Hide
          akmartin akmartin added a comment -

          Polling has been working correctly with v1.1.3.

          Show
          akmartin akmartin added a comment - Polling has been working correctly with v1.1.3.

            People

            • Assignee:
              rpetti Rob Petti
              Reporter:
              cpatrick_netflix cpatrick_netflix
            • Votes:
              1 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: