RocketRaid 2310 - Intermittent Failure

Associate
Joined
3 Sep 2006
Posts
2,048
Location
London
My RR2310 card has been beeping from time to time (took me a while to figure out what was beeping, tbh).

It appeared initially to be related to just the cables, given a wiggle the beeps would stop and my raid array (D: drive) would reappear on reboot.

Anyway, not ruling cable-failure out just yet, going (eventually) into the card's bios it shows that the array is status: critical. Sometimes it shows three cards initialized, sometimes just two. But its very intermittent, albeit increasingly frequent.

I'm surprised that any of the 3*320 western digitals are in failure, they're only about 2 years old, and hardly pushed very hard. About 30 days at 24/7 during that time, in about six batches of five spread equally over that time - highly project-specific. I ran windows chkdisk on the D: drive and, while it found some indexing failures, it reported the drive (array) as all good.

Other than that, used for analysis several hours a day. Fwiw, my 10k 160gb raptor has been heavily in use as a system drive, every day in the same period without error.

So I'd like to recover whatever cards remain salvable, but given the intermittent nature of the problem, I was wondering if anyone knows how I can tell which of the physical drives are still capable?

I'm thinking I should just remove all drives from the array, attach them to the mobo (IP35), and reformat and test them individually, but I'm hoping this is just a false alarm. Couldn't it just be that the cables are twitching out of place, and reporting a missing drive from time-to-time?

Any help would be greatly appreciated.

Running Vista 64, and sorry for the long post, btw.

TIA.

mk17
 
quick update. opened the array Highpoint's RAID Management Console, one of the discs was highlighted, rescanned the array, and all is well!

the rescan took a little over four hours, but all appears well again. As I had mentioned, chkdisk had found some file based anomalies - maybe this had contributed to the array corruption?!

Anyway, all's looking good so no complaints.

bloody hardware.
 
Back
Top Bottom