From: Brcobrem on
Hi Forum Members,

I have nine new drives to install to an Intel SC5200 server soon. In
preparation for this, I manually ran a "Parity Verify" on the SQL mdb
disk (actually a 4 drive + private hotspare RAID 5 "array" drive). The
"Parity Verify is done via "Intel Storage Console v2.16.6". The RAID
controller is an Intel SRCZCR. About 50% through I received an e-mail
message from the RAID Configuration Service:

Asynchronous controller event:
Severity level: critical
Event text: Adapter 0, Array Drive 2: parity error detected

I've looked in the Intel Integrated RAID Software User's Guide
(v1.2.4) pdf manual and web and can't find any information about how
to correct (or approach correcting) this situation. Perhaps I'm
missing something elementary or obvious. I realize that this 2004
software and hardware is not the latest technology, but it does run
like a race horse and works 24/7.

More info:
It looks like the drives need to be replaced using the raid
controller's bootup utility (or the hotswap capability of the Intel
Storage Console Plus (ie. the Windows version). Apparently the raid
controller makes note of hardware/scsi ID's of the individual drives
in an array. This seems like it could take a fairly long time.
Question if you care to reply: I do Live State Recovery bare metal
images of the host drives (ie. logical C: D: E: etc.). I've used Live
State Recovery to do a restore of host E: (RAID 5) before and it
worked perfectly. That said, the
restore I did was to the same array and it's same physical disks. Have
you every tried a bare metal restore to a new unitialized raid array?
Read on . . .
My thinking is to down the server, install the array disks, boot into
the raid utility, create the array, then blow the Live State Recovery
image back onto the array. Think that would work?

I appreciate your comments and suggestions.

Regards,
Brcobrem
From: Brcobrem on
Thanks. That was a great idea and it worked perfectly!

Regards,
Brcobrem