Server 2012 R2 PERC error / drives missing, boot problems
Server shutdown with the following error: "Foreign config found on adapter" Drive 1 was blinking green / amber. "PERC adapter: Virtual disks, ID: 0, virtual disk 0"
RAID5 four drives total.
2 drives originally showed missing.
After reboot first and second drive showed online and foreign
Server prior to first and second drive showing up would not boot, as it sits now only drive we're waiting on is the first drive, says it's rebuilding. I thought the server would boot with three drives however now we're stuck on this screen
RAID5 four drives total.
2 drives originally showed missing.
After reboot first and second drive showed online and foreign
Server prior to first and second drive showing up would not boot, as it sits now only drive we're waiting on is the first drive, says it's rebuilding. I thought the server would boot with three drives however now we're stuck on this screen
Do more with
EXPERT OFFICE® is a registered trademark of EXPERTS EXCHANGE®
We haven't imported the foreign config yet, we get the following screen 
Perform the following steps to import or clear foreign configurations.
1. During bootup, press <Ctrl><R> when prompted by the BIOS banner. The VD Mgmt screen appears by default.
2. On the VD Mgmt screen, highlight the Controller #. The controller number is the only item that appears until you import the foreign configuration.
3. Press <F2> to display the available actions.
4. Press the right arrow key to display the available actions, Import and Clear.
NOTE: Ensure that your virtual disk has all the physical disks by verifying that there are no physical disks marked as Missing in the foreign view page and that all the disks appear as expected before importing them.
5. Select Import to import the foreign configuration or Clear to delete the foreign configuration and then press <Enter>.If you import the configuration, the VD Mgmt displays detailed configuration information. This includes information about the disk groups, virtual disks, physical disks, space allocation, and hot spares
Regarding the NOTE above looking for any information how to best proceed since it's irreversible.
Perform the following steps to import or clear foreign configurations.
1. During bootup, press <Ctrl><R> when prompted by the BIOS banner. The VD Mgmt screen appears by default.
2. On the VD Mgmt screen, highlight the Controller #. The controller number is the only item that appears until you import the foreign configuration.
3. Press <F2> to display the available actions.
4. Press the right arrow key to display the available actions, Import and Clear.
NOTE: Ensure that your virtual disk has all the physical disks by verifying that there are no physical disks marked as Missing in the foreign view page and that all the disks appear as expected before importing them.
5. Select Import to import the foreign configuration or Clear to delete the foreign configuration and then press <Enter>.If you import the configuration, the VD Mgmt displays detailed configuration information. This includes information about the disk groups, virtual disks, physical disks, space allocation, and hot spares
Regarding the NOTE above looking for any information how to best proceed since it's irreversible.
With 3 disks online, you don't want to import. You want to clear the foreign configuration on the remaining drive, then assign it as a hot-spare to start the rebuild. However, it appears as though you have OS damage that needs repairing. You can try bootrec /fixboot and/or bootrec /fixmbr from the command line recovery option (Troubleshoot, Advanced ... I think).
With 3 disks online, you don't want to import. You want to clear the foreign configuration on the remaining drive, then assign it as a hot-spare to start the rebuild.three online I thought the server would boot up, so it's not the drive array but the OS? Please clarify? Thank you.
You want to clear the foreign configuration on the remaining drive, then assign it as a hot-spare to start the rebuild.the remaining drive is already rebuilding after I pulled it, it showed missing and then came online, current status is rebuilding? I'm guessing I don't have to do anything with the config like clearing.
it appears as though you have OS damage that needs repairingshould I wait on the OS repairs after the rebuild?
Last question, would it be possible to install the OS on the current RAID configuration without losing data? I don't believe so but at this time of night doesn't hurt to ask. I do have a backup as well, would be nice to boot up the server though.
Sorry, I'm not sure I looked at the pictures closely enough - disks 0 and 1 are still showing 'foreign', meaning they are NOT participating in the array. YES, you want to import the foreign configuration. It will likely bring disk 0 online and leave disk 1 offline, at which point you can assign it as a hotspare. It LOOKS like disk 1 had failed and started rebuilding when disk 0 failed (or went 'foreign').
Interesting there's amber lights on disks 0 and 1 however VD Mgmt shows all four physical disks are good and the virtual disks.
Operating system is not doing well, and something interesting while going to the command prompt under the repair utility it says, "drive is corrupt or data is not recognizable"
Operating system is not doing well, and something interesting while going to the command prompt under the repair utility it says, "drive is corrupt or data is not recognizable"
Before importing anything you have to be sure what it will do.
I think others have covered mentioned, but I like to be sure and clear.
A foreign config means that while the system was off drives from another system were connected.
You have to look through the log to see which drive was kicked out last as that is the drive that brought down the raid volume.
Forcing the last drive ejected online will be restoring the raid volume with least possible data loss.
If the wrong drive is forced online, the state of data could/would likely be compromised.
I think others have covered mentioned, but I like to be sure and clear.
A foreign config means that while the system was off drives from another system were connected.
You have to look through the log to see which drive was kicked out last as that is the drive that brought down the raid volume.
Forcing the last drive ejected online will be restoring the raid volume with least possible data loss.
If the wrong drive is forced online, the state of data could/would likely be compromised.
The lights are amber because the drives are offline.
A foreign config does not always mean that "while the system was off drives from another system were connected". It certainly can mean that, but most of the time, the drive experienced a timeout/hiccup and the controller forced it offline, keeping its configuration intact while the operator decides what to do with it. If the array is online, the foreign configuration should be cleared. If the array is offline, the foreign configuration should be imported.
On controllers with foreign configs, you don't normally need to actually force online offline disks like on older controllers.
however VD Mgmt shows all four physical disks are good and the virtual disksThis is not what your pictures show. The first shows the VD as offline with two missing drives. The second shows the two missing drives with a foreign status.
A foreign config does not always mean that "while the system was off drives from another system were connected". It certainly can mean that, but most of the time, the drive experienced a timeout/hiccup and the controller forced it offline, keeping its configuration intact while the operator decides what to do with it. If the array is online, the foreign configuration should be cleared. If the array is offline, the foreign configuration should be imported.
On controllers with foreign configs, you don't normally need to actually force online offline disks like on older controllers.
So you just need to repair the OS then. Boot to 2012 installation media and run a chkdsk /r on the C: drive. You may need to run it twice. Then try the bootrec fix commands.
does chkdsk /r always look like it's replacing the entire OS or does this installation really have issues?
Sorry, I went to bed last night :) What do you mean "replacing the entire OS"? You may have extensive damage to the OS. How did it turn out?
Didn't go well. One of the drives was forced online, which now we believe pushed corrupt data into the rest of the array. We can get into the operating system partition and see files / folders. We can't get into the E: partition where the main data is.
We ran chkdsk /r and it appears this may have created more corruption...these things are always 20/20 hind site though.
If there was a way to get into E: from the command prompt we would be better off however it reports the data is corrupt and inaccessible.
To make things worse Acronis (who've I've had issues with in the past) is having a hard time pulling up backups. It's working but support found a bug and is trying to find the fix. We're experiencing errors mounting images, this appears with the Acronis software though so other than the time data is good.
Appears the server is done and we're recovering from backups unless anyone anything to add (Hail Marry pass) situation.
We ran chkdsk /r and it appears this may have created more corruption...these things are always 20/20 hind site though.
If there was a way to get into E: from the command prompt we would be better off however it reports the data is corrupt and inaccessible.
To make things worse Acronis (who've I've had issues with in the past) is having a hard time pulling up backups. It's working but support found a bug and is trying to find the fix. We're experiencing errors mounting images, this appears with the Acronis software though so other than the time data is good.
Appears the server is done and we're recovering from backups unless anyone anything to add (Hail Marry pass) situation.
Absolutely, forcing online drives should only be done in certain situations. This was likely the cause of the corruption. If you ran chkdsk on only the C: drive, it should not have affected the E: drive, but the damage was probably already irreparable.
Forcing a drive online in the order LOFI is the least impactful. Once the wrong drive is brought back the OS portion of data is one that does not change frequently will be sustained but anything that has changed since the drive was ejected will be in an unknown state and will make it hard to recover. The metadata on the two remaining drives recovering from this is ........
In this stage, hopefully you have a backup of data, etc. that would minimize your data loss, do you have a system on which you can establish the roles this system performed, restore the data,,, ,,,
Having reached the current situation, what is it you want to end up with?
I.e. Reconfigure the raid as appropriate, reinstall tge OS, setup the roles, etc and return the system to....
IMHO if Youcan have a raud1 OS and or raid 10 instead of the raid5 setup.
Is the perc firmware up to date?
In this stage, hopefully you have a backup of data, etc. that would minimize your data loss, do you have a system on which you can establish the roles this system performed, restore the data,,, ,,,
Having reached the current situation, what is it you want to end up with?
I.e. Reconfigure the raid as appropriate, reinstall tge OS, setup the roles, etc and return the system to....
IMHO if Youcan have a raud1 OS and or raid 10 instead of the raid5 setup.
Is the perc firmware up to date?
Arnold,
Thank you
In this stage, hopefully you have a backup of data, etc. that would minimize your data loss,yes, working with Acronis as mounting the data is having an issue.
do you have a system on which you can establish the roles this system performed, restore the data,,, ,,,please provide more information regarding this. I have another 2012 R2 server, are you saying I could create a role and get to the data of the failed server?
Thank you
The other server will perform the roles this is supposed to do while you reinstall the oS/reconfigure the failed.
Merely a faster way to get your setup operational.
How is the data backed up? Is it acronisimage,
Has the backup ever tested to make sure?
Since you've ran chdsk, usually of the wrong disk is forced online, and provided the other was not reinserted, to fail the wrongfully added drive, and forcing the last ejected drive in hopes that the metadata was not flushed.
If the data is extremely important and no backup available, your only remedy is to send the two drives that remained online with the last ejected before your... To data recovery firm they will try to reconstitute the raid group/metadata and underlying files/data.
Merely a faster way to get your setup operational.
How is the data backed up? Is it acronisimage,
Has the backup ever tested to make sure?
Since you've ran chdsk, usually of the wrong disk is forced online, and provided the other was not reinserted, to fail the wrongfully added drive, and forcing the last ejected drive in hopes that the metadata was not flushed.
If the data is extremely important and no backup available, your only remedy is to send the two drives that remained online with the last ejected before your... To data recovery firm they will try to reconstitute the raid group/metadata and underlying files/data.
Premium Content
You need an Expert Office subscription to comment.Start Free Trial