Insight Diagnostics says a drive is bad, Array Diagnostics says the same drive is fine.

[Down and Out]

We have a Proliant ML370 G4 server that is receiving conflicting reports. The System Management homepage is showing that everything is ok, the Array Diagnostic Report is showing everything is fine, but Insight Diagnostics is showing one of the drive as needing to be replaced.

I don't want to replace this drive if I don't need to. I saw online that someone had this same scenario and when they replaced the drive all three drives started flashing amber.

Has anyone seen this before? Or might possibly know why we would be receiving conflicting reports?

I am attaching the Array Diagnostic Report and here is the Insight Diagnostics error:

Controller has reported a critical performance threshold error on this drive.
This drive has experienced/recorded error conditions reported by diagnosis and requires replacement.

Any help would be appreciated.

[cnb · Accepted solution]

Looks like the SA641 Firmware is out of date. Several error reporting issues were corrected in later versions. It could have something to do with any of these or is really a hardware issue (Drive 0 reported a Time-out and could be triggering the Diagnostic analysis threshold).

I would make sure all system drivers, support packs, firmware, ADU and ACU are up to date before attempting any h/w replacement. As you noted, some RAID rebuilds with out of date drivers and firmware had a history of failing to rebuild volumes properly after disk replacement.

641 2.84 F/W for W2k3/8 is here:
http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareDescription.jsp?swItem=MTX-24affc031c56442181448eb5e4〈=en&cc=us&mode=5&


hth,

[Roy Main]

Your Array Diagnostic Utility is fairly outdated. That needs to be upgraded to v8.2.

Also, make sure you are testing with Insight Diagnostics 8.2.

Did it say Port 1 Drive ID 2 is the failed drive? That drive has a fair amount of ECC errors. I'd be concerned.

Make sure you have a good backup. Now. Then it's your choice whether to continue to watch it or replace it.

[Down and Out]

Thank you both for the help. The firmware was causing the false positive.