{mosimage}

If hdisk numbers have changed in an HACMP cluster where any heartbeat disks are used you may need to amend the existing HACMP heartbeat resource. It’s not obvious in SMITTY HACMP, and when you find the screen, you can’t change the device value.

Here I’ve re-constructed the command line from the F6 output from SMIT.


In this example, we have two heartbeat disks on each node in the cluster. There are two nodes in the cluster.

  • node_a did have hdisk111 and hdisk191 as heartbeat disks, but removal of redundant disks, followed by a cfgmgr invoked by a reboot has reassigned these disks (check the PVID to confirm) as hdisk109 and hdisk189.
  • node_b’s heartbeat disks have not changed.
  • The cluster will not start and shows errors indicating the PVIDs of the heartbeat disks on both nodes do not match.
  • Our HACMP device resources for node_a are node_a_diskhb03 and node_a_diskhb04.
  • The HACMP network resources for these device resources are net_diskhb03 and net_diskhb04.

 

The commands we can use to change the HACMP Communication Interfaces/Devices from hdisk111/hdisk191 to hdisk109/hdisk189 are: –

/usr/es/bin/cluster/utilities/clchnode -n node_a -a node_a_diskhb03 : node_a_diskhb03 : diskhb : net_diskhb_03 : :
service : /dev/hdisk109

/usr/es/bin/cluster/utilities/clchnode -n node_a -a node_a_diskhb04 : node_a_diskhb04 : diskhb : net_diskhb_04 : :
service : /dev/hdisk189

Now the cluster starts as the PVID’s match once again.

 

About troyski

I'm a freelance UNIX engineer working in the UK. I'm married to Tina and between us we have six children. I'm a bit of an Apple fan boy, and all the Windows machines in the house are a thing of the past now.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Post navigation