This post is more than 5 years old
3 Posts
0
25916
May 25th, 2017 14:00
HD in predictive failure in RAID 5 array - how to replace?
I have a PowerEdge T610 with a PERC H700 RAID controller. Running Windows Server 2008 R2 Enterprise.
All hard drives are connected to the H700 with two drives mirrored for the boot volume, 4 drives in a RAID 5 array for the data volume and 1 drive as a hot spare.
One of the drives in the RAID 5 array is in predictive failure status. I have a drive on hand to replace it but I'm not clear as to the procedure for doing that.
Do I just physically pull the bad drive and let the controller automatically recover using the hot spare, then replce the bad drive? (And if the drive fails completely before I do anything, will the controller automatically swap in the hot spare?)
Is there some more "graceful" procedure? Should I use Open Manage to "Offline" the bad drive?
DELL-Chris H
Moderator
•
9.6K Posts
0
May 26th, 2017 07:00
3chars,
With an ONLINE Predicted Failure the process would be to FORCE OFFLINE that drive, then when it is Offline to remove the drive and replace it. The replacement should start to rebuild upon adding it to the server, if not then assign as a Global Hot Spare and it should start.
If the Predicted Failure drive is already in a Offline state, then you can go ahead and replace it.
Let me know how it goes and if this helps.
Thank you.
3chars
3 Posts
0
May 26th, 2017 11:00
Thanks Chris.
I will probably wait until next week to do this as the boss has a bias against making system changes of any sort on Friday afternoons... and he'll be in here with me if things turn ugly in the mean time. ;)
theflash1932
9 Legend
•
16.3K Posts
0
May 26th, 2017 17:00
Nope, never Friday afternoon. Go home early and remote in to do it Sunday night ... that is how it should be done :)
3chars
3 Posts
0
June 2nd, 2017 08:00
This worked more or less as you described. This system had not previously had the OpenManage software installed - once that was in place I could see what was going on much more clearly.
I forced the failing drive offline but it never showed as "offline". It went into a "ready" state instead.
At that time the array automatically "rebuilt" using the hot spare. When that finished, an hour or so later, I swapped out the failing drive for the new one. The array then automatically "replaced" onto the new drive and returned the hot spare to "ready" status as a hot spare.
Thank you very much for your assistance/advice/hand-holding. :)
mboharakbc
1 Message
0
June 7th, 2018 11:00
I have a slightly different situation. PERC H310 RAID 5. Two disk showed failure predicted. One of my guys just pulled the drive out without forcing it offline first then slapped the new drive in and then executed the task convert to RAID Disk. Now that drive shows Ready but not online and a drive, not even replaced is now rebuilding. I do not want to make a move here because I am afraid that we might lose the server.
AtYourService_NL
1 Message
0
October 28th, 2019 06:00
"I do not want to make a move here because I am afraid that we might lose the server. "
I have exactly the same issue here, what was the fix ?
Thanks!