You receive a ticket from a database group which is complaining about slow access to storage on
only one particular server, as compared to other servers in an active-active cluster. You verify
zoning and LUN masking which are fine. On the FC switch, you verify link speed, errors, and
throughput for the FC switch ports used by this server. There are no errors on the ports and the
link speed is the same as on the switch ports used by other servers in the cluster. You notice that
on one port on the FC switch used by this server, that there is no throughput at all.
What is causing the problem?
The switch has obsolete firmware.
There are too many ports used on the FC switch.
Multipathing was not configured for load-balancing.
The switch is over-subscribed.