Uploaded image for project: 'PBS Pro'
  1. PP-35

Failover file locks are broken if file timestamps don't move for a long enough period (STONITH)

    Details

    • Type: Bug
    • Status: In Progress
    • Priority: High
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: 18.1.1
    • Component/s: Datastore
    • Labels:
      None

      Description

      This has been asked by Alexis for a long time now. Despite our best efforts in ensuring that there is no split brain scenario, the fact that we use NFS both as a datastore as well as a quorum server makes it impossible to completely negate the possibility of a split-brain (ie, primary/secondary both decide to get active).

      One possible solution often found in other software that have active/pasive failover implement a STONITH mechanism. STONITH = Shoot The Other Node In The Head. This is a callout/hook/external script, that the pbs_server needs to call when it has *already* decided to become active. This script can be customized by an admin to call some site-specific tools/actions that 'shoot' the primary dead, if any alive, partially.

        Attachments

          Activity

            People

            • Assignee:
              siddhs Siddharth Sahay
              Reporter:
              pbsadmin pbsadmin
            • Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

              Dates

              • Created:
                Updated: