performance degradation after cold reboot - CR3500 on CL380

[David Abbott_1]

Our CL380 cluster was power cycled a few weeks ago, since when we seem to have noticed a general degradation in performance of disk I/O. We believe we might be affected by a known firmware problem fixed by SoftPaq SP19633, and I know the solution is to just install the firmware - but can anyone advise of a tool or method, possibly querying disk geometry settings, etc. to really confirm that this is the problem ?

It's the standard issue - trying to convince others for the need for downtime to investigate, and needing to confirm that the firmware issue described in EC011115_CW01 is the likely culprit, rather than other possibilities with a growing database on RAID5. We're on Red Hat Linux as well, which appears to have reduced our options for RAID controller management to going to the data centre and plugging in a serial line...

Thanks in advance for any responses.