Distributed Data Guard shows status of "Ready for Rebuild"

[Megan Kielman]

We recently replaced a drive in our RAID 5 set on a Smart Array P400 Controller and the System Management Page is showing a warning regarding the "Distributed Data Guarding". The only thing that looks awry is the stat is "Ready for Rebuild". Shouldn't the rebuild have kicked off automatically? What needs to be done to resolve this alert.

[Gary Benavides Meza]

Open the HP array configuration utility, higlight the storage controller. Then at the right there is an option for common tasks, click on controller settings and check for the rebuild priority. If the rebuild priority is set to low or medium change it to high. Reseat the hard drive and reboot the server the rebuilding task should start immediately, If you can't restart the server just save the chances and wait a couple of minutes. If after that the rebuilding process doesn't start check the ADU and make sure there is no issue with other hard drives in the array, that might cause the hard drive not to rebuild.

[Megan Kielman]

In the ACU all disks and array look fine. The Rebuild Priority is set to Medium. Is it possible the System Management Homepage is showing an incorrect status? I restarted the Foundation service but it didn't resolve the alert.

[Chad Beutler]

I had this problem and found that my disks looked healthy, but they actually had unreadable files on them. I even tried to boot from CD and do a Ghost disk to disk to another array and that failed with unable to read certain files. We were forced to backup the SQL data and rebuild the entire machine. Good luck.

[Megan Kielman]

How did you determine files were unreadable?

[Chad Beutler]

We used Norton Ghost to attempt a disk to disk copy of the data from the array that would not rebuild to a different new array that was created on the same server. When Ghost got a few files in, the copy failed. It would not copy any additional files, unable to read XXX. I can't remember the exact error. The server seemed to run fine otherwise, except that it was missing a disk and would not rebuild the array. We also tried to run a disk bad sectors check which did not resolve the issue and tried moving the drives to a different server hoping that the controller was bad, but we had the same issue still.