I read from gpfs implementation guide about node quorum with tie-breaker disks:
Quote:
Node quorum with tiebreaker disks: Node quorum with tiebreaker disks allows you to run with as little as one quorum node available as long as you have access to a majority of the quorum disks
In my testing environment, I have 4 AIX as quorum nodes, each node connected to 3 GPFS disks. All the 3 disks are configured to be tie-breaker disks.
Then, I disconnect the network of 2 nodes, and then all gpfs file systems failed.
According to the quoted text, the file system should work as long as only 1 quorum is available. Can anyone kindly help to teach me the behavior of node quorum with tie-breaker disks?
Thanks!