After upgrading the Storage Foundation HA / Veritas Cluster Server product from 5.1 to 5.1SP1, various resources fail to probe.
Article ID: 100024665
Updated On:
Cause
An outdated types.cf file will not be properly parsed by the VCS engine on startup.
To verify the types.cf file is outdated, perform the following:
# ls -l /etc/VRTSvcs/conf/types.cf
# ls -l /etc/VRTSvcs/conf/config/types.cf
If the file in /etc/VRTSvcs/conf is larger in size that the file residing in config, the file in use in the cluster is an older revision.
Resolution
Replace the types.cf file to resolve the issue.
This will require the VCS engine to be stopped.
VCS can be stopped while leaving all applications online.
An outage is not required for this action, but may be desired.
To stop VCS leaving all applications / filesystems online:
# hastop -all -force
On each node, perform the following commands in parallel:
(meaning run each command on all nodes prior to executing the next command)
# vxfenconfig -U
# gabconfig -U
# lltconfig -U
(a prompt will appear requesting reset LLT, answer "y" to the prompt)
Now that VCS is stopped, copy the correct types.cf into place on all nodes:
# cp /etc/VRTSvcs/conf/types.cf /etc/VRTSvcs/conf/config/types.cf
Now, restart VCS on each node (again, these commands should be done in parallel on all the nodes):
# lltconfig -c
# gabconfig -c -n# (where # is the number of nodes in this cluster)
# vxfenconfig -c
# hastart
Execute “hastatus –sum” in 2 to 5 minutes to verify all resources have successfully probed.
Issue/Introduction
The output from 'hastatus -sum' command will show this "UNKNOWN" status.
Resource types exhibiting the behavior may include, and may not be limited to: mountpoints
disk groups
applications
IP
network interfaces
