We have a midsize production cluster with two NetApp 31xx Metrocluster as storage backend. In the last year everything seems to work fine, but in the last two months we have a most annoying issue. From time to time we get some dead pathes to the fc targets of the NetApp ( 2 out of 8 paths) of 20 seconds to 1 minute. After these events the system is working several hours without any disturbation and then the event occurs again on other targets.
We have several hosts with ESX 4.1 U1 on HP Proliant blades and switched Brocade FC SAN.
How can we get further information about these issue?
I appreciate any propositions.
