From the CVMCluster_A.log:
2022/07/17 12:07:18 VCS ERROR V-16-20006-1005 CVMCluster:cvm_clus:monitor:node - state: out of cluster
reason: error in processing members
The following error may be present in that log:
ERROR: VxVM vxclustadm ERROR V-5-1-9743 errno 3
In addition, the following error may be present in the OS message log:
vxio: NOTICE: VxVM vxio V-5-0-1379 Node failed to join because of protocol version mismatch with the running cluster
During the upgrade, the cluster protocol version may not have changed. This mismatch will ultimately prevent the cluster from communicating and functioning properly. The mismatch can be verified by a comparison between the protocol versions from the volboot file on the problem node and the working node.
|
In the following example, the CVM resource is online on node1 but failing to online on node2. Note the difference between the protocol version numbers in the volboot files on both nodes:
|
1. Stop the cluster locally:
# hastop -local -force
2. Update cluster protocol version on problem node to match working node:
# vxdctl setversion 200
3. Start cluster.
# hastart
4. Check both the volboot file version and the cluster protocol version itself to ensure they have been updated successfully after cluster is back online locally.
# cat /etc/vx/volboot
# vxdctl support
Check cluster status to ensure the CVM service group has come online.
# hastatus -sum
W_ONLINE) status for extended period of time.