Uploaded image for project: 'Couchbase Kubernetes'
  1. Couchbase Kubernetes
  2. K8S-3480

Prior to failover or graceful failover, we should verify expected state of the cluster before proceeding

    XMLWordPrintable

Details

    • Task
    • Resolution: Fixed
    • Major
    • 2.6.4
    • None
    • None
    • None
    • 10 - Path of Vengeance
    • 3

    Description

      Currently we do not check to see if it is safe to proceed with a graceful or hard failover immediately prior to executing any kind of failover.

      Our expectations should be:
      1) The nodes we know about are part of the cluster and marked as active via /pools/default/
      2) There are 0 inactiveFailed nodes
      3) There are 0 inactiveAdded nodes

      This check should be performed immediately prior to any failover or graceful failover of a node as closely as possible. In the future, we will only do this on versions that do not have MB-61888.

      If this checks fail, we should abort any upgrades and allow reconciliation to recover the cluster prior to continuing any upgrades.

      Attachments

        No reviews matched the request. Check your Options in the drop-down menu of this sections header.

        Activity

          People

            yusuf.ramzan Yusuf Ramzan
            justin.ashworth Justin Ashworth
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Gerrit Reviews

                There are no open Gerrit changes

                PagerDuty