Hard drive starts rebuild then stays in queue status

[bballlal]

ProLiant ML350 G5

Smart Array E200i controller with latest windows driver and firmware

Windows Server 2003

I've got a RAID 5 with four 450GB disks. One of the drives went bad so I replaced it. When I reseat the drive the rebuild process gets to 2 or 3 % and then goes to the "Code 776 - (Ready for Recovery) Logical Drive 1 (1.2 TB, RAID 5) is queued for rebuilding" error message. It stayed in that status overnight. I set the rebuild prioroity to high and then resat the drive, it started and got to 2 or 3% and then stopped and went back to that queued status. I've attached a copy of the ADU report. Can anyone help me determine why the rebuild stops after 3% and then sits in the queued status? HP support is being pretty non-responsive and unhelpful. Thanks!

[bballlal]

I need some serious help. HP is saying that it will never rebuild becuase of all the read errors on physical drive 1. I can't replace it because drive 3 hasn't rebuilt yet! A problem that also seemed to start after replacing drive 3 (the one that hasn't rebuilt yet) is that when windows restarts it hangs on "applying computer settings" and never goes past that point unless I reboot it with the ethernet cable unplugged. This database server has to be rebooted at least once a week and it makes the problems even more inconvenient without being able to reboot it remotely since the ethernet cable has to be unplugged. So frustrating.

[Matti_Kurkela]

A RAID 5 set can recover from one disk failure, but if a second disk fails before the recovery of the first one is completed, the RAID set will fail. Looks like this is what has happened to your RAID set.

At this point, the easiest way to recover would be to replace the failed drive 1, delete and re-create the RAID set, and restore data from backup.

If you don't have a backup, and the data is valuable to you, stop trying things on your own and contact a data recovery specialist (like Kroll Ontrack for example) ASAP.

[bballlal]

Thanks for your reply. I'm aware that if more than one disk went bad we'd lose data. HP said only one drive was bad but it turns out that another drive had hard read errors that wasn't noticed so the array won't rebuild. Now the OS is running like crap. We've got the data now so we're migrating our massive database to a virtual environment before we lose everything.