Today we have an onprime cluster with physical index servers.
Usually our disks fail, leaving the cluster performance down.
Any query that would help catch disk error?
