How do I restart the quorum disk daemon (qdiskd)?

Environment

• Red Hat Enterprise Linux Server 5 (with the High Availability or Resilient Storage Add Ons)
• Red Hat Enterprise Linux Server 6 (with the High Availability or Resilient Storage Add Ons)
• Red Hat Cluster Suite

Issue

• How do I restart the quorum disk daemon (qdiskd)?
• I want to restart the quorum disk daemon, but RHEL 6 does not have a qdiskd init script.

Resolution

WARNING: Restarting qdiskd will result in a temporary loss of the quorum device votes. This should only be performed while enough node-votes are available to maintain quorum without the quorum device.

1. Check the output of cman_tool status to confirm that loss of the Quorum device votes will not result in Total votes dropping below Quorum. The following is an example of when it is safe to restart qdiskd, taken from a two node cluster with quorum device and two nodes with one vote each, all online.

        # cman_tool status |grep -ie quorum -e votes
        Expected votes: 3
        Quorum device votes: 1
        Total votes: 3
        Node votes: 1
        Quorum: 2
   

   The following example is from the same two node cluster with one node offline. Restarting qdiskd under these circumstances is not safe because it would result in quorum loss:

        # cman_tool status |grep -ie quorum -e votes
        Expected votes: 3
        Quorum device votes: 1
        Total votes: 2
        Node votes: 1
        Quorum: 2
   

2. Restart qdiskd.

   • Red Hat Enterprise Linux 5:

     1. Perform service qdiskd restart

         # service qdiskd restart
        

        Or altnernatively, in some cases it may be necessary to stop qdiskd on all nodes, perform some activity or maintenance, and only then start it back up.

         # service qdiskd stop   ### on all nodes
         # ### perform activity
         # service qdiskd start  ### on all nodes
        

   • Red Hat Enterprise Linux 6:

     1. Manually stop qdiskd with signal 15:

         # killall -15 qdiskd
        

     2. Manually start qdiskd:

         # qdiskd
        

Quorum disk should show as Online in the output of clustat within a few minutes, depending on how the cluster heartbeat timeout is configured. You can monitor /var/log/messages to watch qdiskd initialize:

        # tail -f /var/log/messages

Root Cause

• Red Hat Enterprise Linux (RHEL) 5 shipped with a separate qdiskd init script /etc/init.d/qdiskd, which allowed users to restart qdiskd without restarting the cluster. In RHEL 6, the functionality of /etc/init.d/qdiskd was moved into the cman init script, so qdiskd must be restarted manually.

• More information on quorum and how it is calculated can be found in the High Availability Administration Guide.

Diagnostic Steps

• This article may apply to you if you are interesting in the procedure for and implications of restarting the quorum disk daemon.

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.