Hello,

      The wipe workspace plugin is no longer removing old data, what happens is that it just renames the workspace to random folder name, however this renamed folder is not getting deleted, instead it's just kept there on the Disk, which resulted in a huge workspace consumption.

      This is happening pretty much across all of our agents

          [JENKINS-73662] The workspace not getting cleaned up

          Markus Winter added a comment -

          You're on windows. Are the directories empty or are there still files inside?

          Markus Winter added a comment - You're on windows. Are the directories empty or are there still files inside?

          All directories still contain all files in them, hogging all available disk space.

          Its happing across all slaves we have connected to the master, the picture is just an example from a random machine. 

          Mohamed Khamis added a comment - All directories still contain all files in them, hogging all available disk space. Its happing across all slaves we have connected to the master, the picture is just an example from a random machine. 

          To add more info, we are using Jenkins version 2.440.3 and the version of the plugin is 0.45

          Mohamed Khamis added a comment - To add more info, we are using Jenkins version 2.440.3 and the version of the plugin is 0.45

          Markus Winter added a comment -

          Anything in the Jenkins logs or agent logs that it's not able to delete the files?

          Markus Winter added a comment - Anything in the Jenkins logs or agent logs that it's not able to delete the files?

          In the Jenkins console, it says that succeeded

          Mohamed Khamis added a comment - In the Jenkins console, it says that succeeded

          Markus Winter added a comment -

          It says Deferred wipeout is used. So all that happens during the build is that the folder is renamed and then a background job should start that takes care of deleting things.

          You need to look in the logs of Jenkins itself, not the job log. I think there is also an admin monitor that tells you if there are problems with the async deletion of files

          Markus Winter added a comment - It says Deferred wipeout is used. So all that happens during the build is that the folder is renamed and then a background job should start that takes care of deleting things. You need to look in the logs of Jenkins itself, not the job log. I think there is also an admin monitor that tells you if there are problems with the async deletion of files

          Hello,

           

          After further investigation, It was determined that the issue is not in the plugin ws-cleanup-plugin

          But rather in the plugin: resource-disposer-plugin

          The <Jenkins_Url>/administrativeMonitor/AsyncResourceDisposer 

          Showes a significant amount of recourse that are to be disposed, yet they are not getting removed.

          Any Idea how to fix this ?

          Mohamed Khamis added a comment - Hello,   After further investigation, It was determined that the issue is not in the plugin ws-cleanup-plugin But rather in the plugin: resource-disposer-plugin The <Jenkins_Url>/administrativeMonitor/AsyncResourceDisposer  Showes a significant amount of recourse that are to be disposed, yet they are not getting removed. Any Idea how to fix this ?

          Just to add more information to my previous comment,

          Since the last Jenkins restart (which was carried out 2 days ago) The resource disposal has logged almost 1005 workspace entry, that its yet to dispose, however it's not removing any.

          Mohamed Khamis added a comment - Just to add more information to my previous comment, Since the last Jenkins restart (which was carried out 2 days ago) The resource disposal has logged almost 1005 workspace entry, that its yet to dispose, however it's not removing any.

          Mohamed Khamis added a comment - - edited

          Hello,

          Any support here please? 

          Mohamed Khamis added a comment - - edited Hello, Any support here please? 

          Emil Śniadach added a comment - - edited

          similar behaviour on our production jenkins.
          unable to reproduce on testing setup  [ the same JK version and the same plugins versions]. 
          jenkins/jenkins:2.426.2-lts-jdk11, ws-cleanup plugin 0.46, Resource Disposer: 0.23As WA I setup disable deferred wipeout on job level and in some cases on node level.

          I restarted JK 2 weeks ago and issue faced up about 1 week ago after reach "no space left on device".

          No idea what to check at the moment.
          mkhamisg  did you fix it somehow?

          Emil Śniadach added a comment - - edited similar behaviour on our production jenkins. unable to reproduce on testing setup  [ the same JK version and the same plugins versions].  jenkins/jenkins:2.426.2-lts-jdk11, ws-cleanup plugin 0.46, Resource Disposer: 0.23 As WA I setup disable deferred wipeout on job level and in some cases on node level. I restarted JK 2 weeks ago and issue faced up about 1 week ago after reach "no space left on device". No idea what to check at the moment. mkhamisg   did you fix it somehow?

            olivergondza Oliver Gondža
            mkhamisg Mohamed Khamis
            Votes:
            2 Vote for this issue
            Watchers:
            5 Start watching this issue

              Created:
              Updated: