msa1500cs "Consistency Check Failed"

[costin gusa]

hi guys,

On 13th (yeah, bad luck) a msa1500cs with a 6-disks sata raid6 array was connected through a qlogic hba with a proliant dl380 g4 and mounted as ext3 under rhel5.

a cron script started which executed a mv command, then the kernel spitted this error:

Sep 13 22:10:01 fileserver kernel: EXT3-fs warning (device dm-0): empty_dir: bad directory (dir #76219579) - no `.' or `..'

following this, a lot more errors appeared in kernel log (see attached file) finally ending with a kernel panic this morning.

ran hpaducli today and it says:

Consolidated Error Report:
Controller: MSA1500 CS in SGA0733024
Device: Logical Drive 3
Message: Parity Initialization Failed
Controller: MSA1500 CS in SGA0733024
Device: Logical Drive 4
Message: Parity Initialization Failed

however, neither disks nor controller shows any red light or unusual light activity. how can i troubleshoot further?
many thanks,
costin

[costin gusa]

ps: attached the hpaducli report

[Uwe Zessin]

Looks like a 'strip kill' has happened (those 13ths...). According to the ADU report, one disk drive has failed and others show issues, too, but I don't have the tools to analyze it easily and - I'm sorry, but I don't have time to decode the SCSI status information manually.

I suggest you log a call with HP and let them take a look at the ADU report.

[costin gusa]

the failed one is in the local smart array controller - ignore it (http://forums12.itrc.hp.com/service/forums/questionanswer.do?threadId=1245574), it is safely running on spare drive.