Unsolved
This post is more than 5 years old
4 Posts
0
12424
January 31st, 2005 23:00
Virtual Disk Problem
Hi All,
I have a rather annoying problem with a RAID 1 mirror on a PE4400 - PERC3/di system. The system has, or should have rather:
One RAID 1 virtual disk consisting of 2 x9gb drives.
One RAID 5 virtual disk consisting of 6 x36gb drives.
The problem is when I open the Dell Array Manager Console, I see that the RAID 1 virtual disk has a yellow exclaimation mark against it and failed redundancy alerts are obvious for that virtual disk. I looked for the options to "rebuild" and/or "remove orphan" and then reconfigure a hot spare as suggested in other posts and FAQ's, but I do not see these options anywhere in the array manager GUI? Only the alerts mentioned above.
What I suspect is that a disk in the RAID 1 container perhaps went faulty and was replaced with a new disk. However, it doesn't look like the procedure was carried out properly. The container that should be RAID 1 is just coming up as a normal volume when I hit "CTRL + A" for the RAID config menu for the PERC controller when the server is booting.
The RAID controller seems to be able to "see" the new disk in slot 4 on the backplane, but I can't find a way to add this new disk back into the RAID 1 virtual disk container. Any suggestions would be greatly appreciated. This is a fairly important server and that RAID 1 container held the NT4.0 system partition, so rebuilding from scratch is most certainly not an option I'm looking forward to!
The server is up-to-date with the latest embedded firmware and perc3 drivers etc.
Thanks!
No Events found!


barhampa
718 Posts
0
February 1st, 2005 09:00
Desparate Dan
4 Posts
0
February 1st, 2005 10:00
Desparate Dan
4 Posts
0
February 1st, 2005 18:00
I’ve managed to uncover some more information regarding the problem I'm having.
Some server details:
Poweredge 4400 (ESM Firmware v5.50)
Adaptec PERC 3/Di Controller (Newest firmware revision installed)
Server OS = Windows NT 4.0 sp6
Newest PERC3/Di NT Drivers installed.
Dell OpenManage Array Utility v3.1.1
My problem is:
I have failed redundancy on a RAID-1 virtual disk that did have 2x 9gb SCSI disks included in the container. I am nearly certain now that someone removed one of the disks in this container in order to replace it with a new disk. However, I don’t think they did the operation using the proper procedures.
From what I can gather, this is what happened:
A suspected faulty disk was removed without first being prepared for removal or being marked offline. The server was then shutdown, the old disk removed and the new disk installed. The whole procedure being quite contrary to the Dell recommended method; (Dell support doc: TT1070984).
What I have now inherited is a failed redundancy volume and I don’t know how, or if, I can re-establish the redundancy without needing to rebuild the virtual disk from scratch? As the virtual disk in question holds the NT system partition, am I right in assuming that if I have to redo the RAID-1 mirror, I’ll need to reinstall the NT OS afterwards?
None of the options that are detailed in other posts or on the Dell support site are available to me when I run the Array Manager GUI, such as “Rebuild” or “Remove Orphans” etc. I just see a yellow exclamation mark against the array in question.
Is there a way to re-establish the redundancy without loss of system data or am I up the creek?
Any help or pointers would be appreciated. Thanks!!
Message Edited by Desparate Dan on 02-01-2005 02:19 PM
Message Edited by Desparate Dan on 02-01-2005 02:19 PM
barhampa
718 Posts
0
February 2nd, 2005 05:00
Desparate Dan
4 Posts
0
February 6th, 2005 11:00