After one of the drives in our IBM RAID gone, we bought a new one and replaced it.

The rebuilding went fine, but now we see "Bad stripes" reported for the logical disk. Moreover, some data can not be read from the disk, and ServeRAID reports bad blocks.

CHKDSK just hangs on 22% of reading, closing it and re-running jumps to the 22% again.

Question - is it possible to resolve this situation by synchronizing the logical drive? Or it better leave the CHKDSK running until it hopefully completes?

Also, what the synchronization actually does and can it heal bad stripes?

The server model is x3400, RAID is ServeRAID 8k-l.


  • I don't have a solid answer for you, but my advice would be to contact IBM; support is one of the reasons you pay a premium for good server hardware. Also - You haven't said what model of server and RAID controller you have, that will probably make a difference for the answer as well. – mfinni Jan 27 '11 at 21:36
  • Correct - question updated. – SyRenity Jan 27 '11 at 21:50

Are you sure the new hard disk isn't "DOA"?

  • The disk is fine, the rebuild completed successfully. According to IBM, disks can be re-built and still considered with bad stripes - i.e. contain bad blocks. – SyRenity Jan 27 '11 at 23:16
  • I think you should get some answers from IBM. Maybe there's more than one faulty hard disk, or your RAID card is causing corruption. Your trouble wasn't caused by a single faulty disk. – DutchUncle Jan 28 '11 at 20:21
  • It seems BOTH disks became faulty at same time - just our luck it seems... – SyRenity Jan 31 '11 at 12:50
  • Welcome to RAID5 hell :-) – DutchUncle Jan 31 '11 at 21:39
  • 1
    RAID1 actually, but that was hell indeed :). – SyRenity Feb 10 '11 at 18:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.