Symptom:
A Gls resource failure occurs, and then cluster applications are switched or stopped.
Corrective action:
During operation, due to the effect of maintenance work such as replacing or restarting a HUB, changing a cable connection, and shutting down a network to take security measures, the GLS monitoring function detects a failure, and then cluster applications may be switched or stopped. In this case, perform the following procedure to recover all the failed nodes.
Recovering the transfer path where a failure has occurred
Restore the failed transfer path to the normal status (perform necessary work such as reconnecting the cable, powering on the switch/HUB again, and replacing the failed switch/HUB).
Checking the GLS status
Fast switching mode
Make sure that the status of the communication target is Active by using the dsphanet command with the -o option.
GS linkage mode
Make sure that the status of the communication target is Active by using the dspobserv command.
NIC switching mode
Make sure that the status of HUB monitoring is ON by using the dsppol command.
Also, make sure that the status of the standby patrol is ON by using the dsphanet command.
Virtual NIC mode
Make sure that the status of HUB monitoring is ON and the status of the standby patrol is Active by using the dsppathmon command.
Clearing the Gls resource failure status
Clear the Gls resource failure status by using "Cluster Admin" in Web-Based Admin View.
Restoring the original operation status
Restore the original operation status by restarting the cluster applications or performing fail-back operation for the cluster applications, from "Cluster Admin" in Web-Based Admin View.
