Disk connectivity failure and recovery

The GPFS has certain error messages defined for local connection failure from NSD servers.

If a disk is defined to have a local connection and to be connected to defined NSD servers, and the local connection fails, GPFS bypasses the broken local connection and uses the NSD servers to maintain disk access. The following error message appears in the GPFS log:

6027-361 [E]: Local access to disk failed with EIO, switching to access the disk remotely.

This is the default behavior, and can be changed with the useNSDserver file system mount option. NSD server considerations.

For a file system using the default mount option useNSDserver=asneeded, disk access fails over from local access to remote NSD access. Once local access is restored, GPFS detects this fact and switches back to local access. The detection and switch over are not instantaneous, but occur at approximately five minute intervals.

Note: In general, after fixing the path to a disk, you must run the mmnsddiscover command on the server that lost the path to the NSD. (Until the mmnsddiscover command is run, the reconnected node will see its local disks and start using them by itself, but it will not act as the NSD server.)

After that, you must run the command on all client nodes that need to access the NSD on that server; or you can achieve the same effect with a single mmnsddiscover invocation if you utilize the -N option to specify a node list that contains all the NSD servers and clients that need to rediscover paths.