CRC error in RAID - which drive died?

edited September 2007 in Hardware
Hi Everyone,

I recently was asked to look at a server that was giving disk {CRC} read errors. It is setup as a Virtual Machine {VM ware} server.

I was instructed to break the raid (raid-0), perform a low-level format of the three Sata drives that were in the raid array, and then recreate the raid, reinstall Windows Server 2003 (R2) and the VM ware images. The instructions given to me to low-level format the drives should format around any bad sectors on the hard drive.

I did all of the above, and now we still get the CRC error.

So it seems one of the drives needs to be replaced, but I don't really want to have to buy 3 new drives when only one is down.

Can anyone recommend a method or application that will allow me to establish what drive or drives have the problem? Will a chkdisk command do it?

Thanks for any advice, I realise whatever I use will take a fair whack of time.

Hunter

Comments

  • TexTex Dallas/Ft. Worth
    edited September 2007
    Whoever made the disk... maxtor, seagate, wd etc.... will have tools to certify the drive.

    You know it may not even be the drives.

    Cowboy
  • edited September 2007
    SEAGATE Barracuda 7200.7
    120GIG Sata
    Model # ST3120827AS

    Thanks, will look for some tools from SeaGate....
  • edited September 2007
    Found a SeaGate Tool

    http://www.seagate.com/www/en-us/support/downloads/seatools/seatooldreg

    As I have SEAGATE Barracuda 7200.7 / 120GIG Sata / Model # ST3120827AS, so maybe this tool will work as well.

    Will post back my findings.

    Hunter
  • edited September 2007
    OK so the RAID manager has no option to check for errors or anything like that - it has Create Raid, Delete Raid, Repair Raid1, Low-Level Format, Resolve Conflicts (cant find any) and Logical Drive Info.

    The SeaTools utility says no hard drives found. Maybe it wont see a Raid array... might try breaking the Raid and running SeaTools again.
  • TexTex Dallas/Ft. Worth
    edited September 2007
    Won't see those drives in a array! Can you stick them on a normal controller or will yours let them be seen as non raided single drives??

    Cowboy
  • edited September 2007
    Looked at the motherboard, and the 3 SATA drives were plugged into SATA ports 3, 4 and 5. Hmmm... I unplugged them all and plugged one into SATA port 1.

    Now SeaTools sees this drive. I am now erasing, and will then do diagnostics. I have 3 drives to do this to so it may take a while!

    Will post result.

    Hunter
  • TexTex Dallas/Ft. Worth
    edited September 2007
    This will take a long time especially if you reformat using their tools and then recertify...

    I would almost bet its not a drive failure.... A bad cable or psu etc... when you have that many drives is just as likely.

    I wish ya the best of luck !

    Cowboy
  • edited September 2007
    Tex wrote:
    This will take a long time especially if you reformat using their tools and then recertify...

    I would almost bet its not a drive failure.... A bad cable or psu etc... when you have that many drives is just as likely.

    I wish ya the best of luck !

    Cowboy

    Cheers!

    I'll check each drive, and if still not finding the cause, try swappin' them cables, and try a different PSU....
  • edited September 2007
    There were 3 drives in a Raid0 array.

    Drive 0, Drive 1, Drive 2

    Drive 1 was the culprit - SeaTools has identified it as the dodgy disk.

    Thanks for all the advice.

    Hunter
Sign In or Register to comment.