Details

    • Bug
    • Status: Done
    • Medium
    • Resolution: Fixed
    • 1.10.0
    • 1.12.0
    • None
    • None
    • Yes
    • Yes
    • Yes

    Description

      Affected version 1.10.0 and main branch.

      If I take default cr.yaml and set setFCV flag to true:

         upgradeOptions:
          versionServiceEndpoint: https://check.percona.com
          apply: 4.4-recommended
          schedule: "0 2 * * *"
          setFCV: true

      I see the thread leak in a mongod container of the replicaset pods.

      As a result pods might get into a crash loop eventually.

       

      Steps to reproduce:

      1) I can reproduce on both GKE and minikube

      2) Deploy default bundle.yaml

      3) Change setFCV to true in default cr.yaml

      4) Deploy cr.yaml

       

      P.S. It does not matter if sharding is used or not.

       

      How to see the issue

      **Exec into mongod container of a replicaset pod:

      $ kubectl exec -ti my-cluster-name-rs0-0 bash -c mongod
      bash-4.4$ while true; do ps -eLf | wc -l && sleep 5 ; done
      174
      177
      180
      183
      186
      189
      192
      193
      196
      199
      202
      205
      208
      211
      214

      Expectation

      No thread leak

      Attachments

        Activity

          People

            dmytro.zghoba Dmytro Zghoba
            sergey.pronin Sergey Pronin
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Smart Checklist