How to Replace a Fibre Channel Component in a Running Cluster
Use this procedure to replace the following node-to-switch components in a running cluster.
Node-to-switch fiber-optic cable
Node-to-storage fiber-optic cable
GBIC or an SFP on an FC switch, storage array, or node
Determine the resource groups and device groups that are running on the node that is connected to the component that you are replacing.
Record this information because you will use it in Step 4 of this procedure to return resource groups and device groups to these nodes.
# scstat
Move all resource groups and device groups to another node.
# scswitch -S -h nodename
Replace the node-to-switch component.
For the procedure on replacing a fiber-optic cable between a node and an FC switch, see the documentation that shipped with your switch hardware.
For the procedure on replacing a GBIC or an SFP on an FC switch, see the documentation that shipped with your FC switch hardware.
Return the resource groups and device groups that you identified in Step 1 to the node that is connected to the component that you replaced.
# scswitch -z -g resource-group -h nodename # scswitch -z -D device-group-name -h nodename
How to Replace a Chassis in a Running Cluster
Use this procedure to replace a storage array chassis in a running cluster. This procedure assumes that you want to retain all FRUs other than the chassis and the backplane.
To stop all I/O activity to this storage array, detach the submirrors that are connected to the chassis you are replacing.
For more information, see your Solstice DiskSuite or VERITAS Volume Manager documentation.
Is this storage array made redundant by host-based mirroring?
If yes, proceed to Step 3.
If no, shut down the cluster.
For the full procedure on shutting down a cluster, see the Sun Cluster system administration documentation.
Replace the chassis and backplane.
For the procedure on replacing a chassis, see the Sun StorEdge 3000 Family Installation, Operation, and Service Manual.
Did you shut down the cluster in Step 2?
If no, proceed to Step 5.
If yes, boot it back into cluster mode.
For the full procedure on booting a cluster, see the Sun Cluster system administration documentation.
Reattach the submirrors that you detached in Step 1 to resynchronize them.
Caution - The world wide numbers (WWNs) might change as a result of this procedure. If the WWNs change, and you must reconfigure your volume manager software to recognize the new WWNs.For more information, see your Solstice DiskSuite or VERITAS Volume Manager documentation.