3 Replies Latest reply on Sep 6, 2011 2:48 PM by Flying_Kiwi

    RAID controller behaviour on DQ35JO with one bad block

    Clixel

      Hi all,

       

      A couple of quick searches didn't reveal anything, and apparently my board (DQ35JO) is beyond the support date.

       

      I'm supporting a small server with two WD1600AAJS drives configured in RAID1, and one of the drives has a single bad block (according to a read-only surface scan). Does anyone know how the RAID controller handles this circumstance? Will it just write around the bad block, or will this occassionally screw up the RAID?

       

      (As I type this, I realize I have been assuming the RAID is from the motherboard, but I have not checked for add-on cards and don't have physical access to the server. I may be asking this question in the wrong spot.)

       

      I ask because every now and then the verify on the backup goes nuts and tells me that hundreds of files are different or corrupt. I'm not sure how a single bad block could cause that, but I don't see any other indications of poor health on the server.

       

      Any insight would be greatly appreciated. Thanks!

      Cory

        • 1. Re: RAID controller behaviour on DQ35JO with one bad block
          Flying_Kiwi

          It'd be worth backing up what you can and then running a full HDD test on the suspect drive using software provided by the manufacturer. You can either take the drive out of the PC or revert its HDDs back to IDE via the BIOS settings. As there'll be a DOS version of the HDD utility software that you can boot from a bootable USB pendrive with, you'll be able to take it from there. Let us know how testing the drive(s) goes. Usually such software will automatically mark any dodgy blocks in a way that the drive won't attempt to use them anymore and thus reliability of the drive is restored.

          1 of 1 people found this helpful
          • 2. Re: RAID controller behaviour on DQ35JO with one bad block
            Clixel

            Thanks Flying_Kiwi! I'll definitely give that a try with the manufacturer's software. With an intermittent problem, I have no way of knowing if that's the correct answer, so I've flagged it as "helpful" so far.

             

            As a side note, I received a notification of a private message, but by the time I reset my password and was able to log in again the private message appeared to have vanished. If you are reading this and wondering why I haven't responded to your private message, it's because I was never given the opportunity to read it. Sorry!

            • 3. Re: RAID controller behaviour on DQ35JO with one bad block
              Flying_Kiwi

              Clixel wrote:

               

              Thanks Flying_Kiwi! I'll definitely give that a try with the manufacturer's software.

              Let us know how you get on. Manufacturers software is often very fussy when it comes to diagnosing problems so even though an error code may be generated on test, (if the drive is out of warranty) you MAY find that a low level format AKA zero fill resolves the issue for now and allows you to reinstall and get much more life out of the drive. Of course if the drive is still in warranty, you should contact the supplier or manufacturer.

               

              As long as you regularly back up your data, you should be OK. What does SMART say about the drive and are both drives in the array identical (with the same firmware)? If there are no SMART warnings, that's always a good sign!

               

              PS I've not sent you any PMs.