Replacing Faulty Storage in a Storage Server
Generally, if you have more than one path to your application server, it is not necessary to stop I/O operations on the application server before changing faulty storage on the storage server.
Change the mapping instead. You can do substitutions on the storage server with non standard mirrors and application servers will simply failover. Indeed if you substitute primary volumes, you can do this with “all” mirrors as I/O is passed over the mirror channel and the application server remains on the primary storage server as if nothing had happened. It is all transparent.
It is always a good idea to check new storage added to an application server to verify that it matches the changes you have made on a storage server.