Start a Conversation

Unsolved

This post is more than 5 years old

60985

October 23rd, 2014 15:00

MD1000 failed disk

I have an MD1000 that had a disk fail, and the hot-spare has a failure predicted. My RAID5 is off-line at the moment.

Is it possible to force the failure predicted drive to go online until I can replace the failed drive(s)?

Richard

Moderator

 • 

6.9K Posts

October 24th, 2014 10:00

Hello Richard,

What I would do is that if you know which drive that failed second which i figure is your hot spare drive then I would try to see if the drive can go online or try reseating that drive & see if you can get it to start rebuilding. Also what you can do is to remove the failed drive first and see if that makes the rebuild start as well as I have seen in some cases that a failed drive can cause the array to go offline and once removed the array will come back online. Here is a link to the user guide if you don’t already have it. ftp://ftp.dell.com/Manuals/all-products/esuprt_ser_stor_net/esuprt_powervault/powervault-md1000_Owner%27s%20Manual_en-us.pdf

Please let us know if you have any other questions.

15 Posts

October 30th, 2014 15:00

It turns out that my problem is much more severe. I lost 2 drives in close timing to each other. One listed as failed (D11), the other as failure predicted (D12). I incorrectly thought that the failure predicted drive was the HS. It wasn't. It was part of the RAID. The HS (D6) never kicked in, and never rebuilt.

In trying to get the drive I thought was a HS to kick in, I cleared the drive. My RAID5 now has 2 missing drives (D11 and D12).

Is it possible to replace the cleared D12 and get the failed D11 to stay online long enough to get the HS to rebuild? Is there any way to force this to happen?

15 Posts

November 3rd, 2014 09:00

Anyone, or am I just up the creek on this one?

15 Posts

November 3rd, 2014 10:00

I'm not familiar with the retag procedure. Can you explain? Also, by your addressing the cleared drive and not the failed drive, are you suggesting that there is no option to force the failed drive back online?

Moderator

 • 

6.9K Posts

November 3rd, 2014 10:00

Hello Richard,

So since you cleared the drive what you can try to do is a retag on the drive & see if it can rebuild the meta data on the drive.  If that doesn’t work then what I would suggest if you are in need of getting the data back is to contact a drive recovery company that can see if they can get the data back for you.  When doing a retag is the last option that other than going to a drive recovery company to get the data back.  

Please let us know if you have any other questions.

Moderator

 • 

6.9K Posts

November 3rd, 2014 11:00

Hello Richard,

You can try to reseat the failed drive & see if it comes back online, but there isn’t a way to force the failed drive back online.  Now when you do a retag what you are doing is recreating the virtual disk exactly as it was created the first time with all the drives that were in the original VD.  When you do this you can sometimes get your virtual disk to come back to a degraded state.  If you can get the VD to a degraded state then you should be able to replace the failed drive and get the VD optimal.  Here are the steps that you will need to do if you were to try & do a retag:

1.Create a new array using the same stripe, RAID level, and member disks as before. The caching options and size don't matter.

2.Do NOT intialize the array when you create it. An initialization is a low level format that will wipe the data

3.Before you leave the controller BIOS offline the first drive that went foreign. If you boot up the system with that drive online it will corrupt the array. If you can't remember which drive went offline/foreign first then review the logs. Remember that the drive numbering starts with 0.

4.Boot up the server and check to see if the data is present.

5.If the data is not present then stop & contact a data recovery company to get the data convered. Do not attempt to recover the data. If anything is written to the drive it will start overwriting the old data and corrupt the old data.

Please let us know if you have any other questions.

15 Posts

November 3rd, 2014 12:00

Thanks so much for the detailed description. Now, my only concern before beginning these steps is that the message I received when clearing disk 12 indicated that this is a data destructive procedure. As I thought it was the hot spare drive, no big deal, but now learning that the drive was actually a part of the RAID, it seems to me that the procedures you outline would not actually recreate the missing data. It seems that my only option is if the failed drive will come back online for long enough to rebuild either the cleared drive or the failed drive to one of the spares. Is my thinking correct?

Moderator

 • 

6.9K Posts

November 5th, 2014 08:00

Hello Richard,

So since you cleared the data on the drive there is a chance that it maybe able to rebuild the data when you do a retag but it is hard to say.  The best option would be to see if you could get the drive that failed to come online & see if it can rebuild.  If you can’t get the drive to come online then I would contact a drive recovery company to recovery the data if it is mission critical data but once again that is your call to make as only you know how critical your data is.

Please let us know if you have any other questions.

No Events found!

Top