Uploaded image for project: 'Apache Ozone'
  1. Apache Ozone
  2. HDDS-8299

Disk full situation on a leader DN may result in followers getting stuck in a retry loop

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • 1.4.0
    • Ozone Datanode
    • None

    Description

      This issue was seen in a large-scale Ozone deployment where a few of the DNs ran out of disk space.

      The follower DNs in the same Ratis pipeline as the leader would get stuck in a loop. We have some initial analysis and the jira will be updated with details soon.

      Attachments

        Issue Links

          is a parent of

          Improvement - An improvement or enhancement to an existing feature or task. HDDS-8417 Cap on the queue of commands at DN

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved
          is caused by

          Bug - A problem which impairs or prevents the functions of the product. HDDS-8324 Remove chunk cache entry only for the lower index value used for apply transaction

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. RATIS-1761 If LeaderStateImpl is not running, it should not restart a LogAppender.

          • Major - Major loss of function.
          • Resolved

          Improvement - An improvement or enhancement to an existing feature or task. RATIS-1836 Reduce retry frequency by leader on failure path for appendEntry

          • Major - Major loss of function.
          • Resolved

          Activity

            People

              sumitagrawl Sumit Agrawal
              arp Arpit Agarwal
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: