Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-26388

Jenkins think the image cap is reached although it's not and cannot spawn new slaves

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Fixed
    • Icon: Major Major
    • docker-plugin
    • None

      Jenkins reports that the cap is reached and it's not possible to spawn new slaves:

      Jan 12, 2015 9:41:54 AM INFO com.nirima.jenkins.plugins.docker.DockerCloud provision

Excess workload after pending Spot instances: 4

Jan 12, 2015 9:41:54 AM INFO com.nirima.jenkins.plugins.docker.DockerCloud countCurrentDockerSlaves

Images found: [Image{repository=null, tag=null, id=58f7c5283c6d4d5df99c3594c99943aec38324da3734e327c9f0337dd48451e3, created=1418900954, size=0, virtualSize=2063934469}]

Jan 12, 2015 9:41:54 AM INFO com.nirima.jenkins.plugins.docker.DockerCloud addProvisionedSlave

Total container cap of 5 reached, not provisioning.

      Reality is that there is only one slave running and the /docker-plugin view tracks the active hosts correctly.

      My guess: The DockerCloud#provisioningAmis hashtable is not updated consistently and goes out of sync

          [JENKINS-26388] Jenkins think the image cap is reached although it's not and cannot spawn new slaves

          Kanstantsin Shautsou added a comment -

          Do you have any other running containers on this node?
          I'm patching locally to exclude not templated containers from total list.

          Kanstantsin Shautsou added a comment - Do you have any other running containers on this node? I'm patching locally to exclude not templated containers from total list.

          Antonio Sagliocco added a comment -

          Hi, thanks for looking.
          No, there's only one container running. The node is dedicated to Jenkins slaves

          Antonio Sagliocco added a comment - Hi, thanks for looking. No, there's only one container running. The node is dedicated to Jenkins slaves

          Kanstantsin Shautsou added a comment -

          Check `docker ps -a` and remove all unrelated nodes. Then when provisioning will start - capture some log.

          Kanstantsin Shautsou added a comment - Check `docker ps -a` and remove all unrelated nodes. Then when provisioning will start - capture some log.

          Kanstantsin Shautsou added a comment -

          Please try 0.11.0 and reopen if issue still exists.
          Also please specify information mentioned in https://github.com/jenkinsci/docker-plugin/blob/master/CONTRIBUTING.md

          Kanstantsin Shautsou added a comment - Please try 0.11.0 and reopen if issue still exists. Also please specify information mentioned in https://github.com/jenkinsci/docker-plugin/blob/master/CONTRIBUTING.md

            integer Kanstantsin Shautsou
            kalymero Antonio Sagliocco
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

              Created:
              Updated:
              Resolved: