Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1221
  • Last Modified:

Data restore after failed SAS controller Dell PowerEdge 840

Server has a failed RAID drive and the controller is not working properly as I'm getting I/O error messages in Windows Event Log and can't copy files into external USB, it also gives I/O errors and stops.  

We want to transfer data into the new server but are not able to at this point.

Is it possible to plug the hard drive to a regular computer to access the data using a SAS controller?

For example: install one of these SAS controllers on a Dell Optiplex
http://www.newegg.com/Product/Product.aspx?Item=N82E16816115100

I'm just not sure if that gets installed if the computer tries to boot from it instead of the SATA drives onboard?

My other option is to get a used SAS5i/r controller to replace the defective one, but since I have a failed drive I'm not sure if the controller will allow to initialize the array for boot?
0
Armin57
Asked:
Armin57
  • 5
  • 3
  • 2
  • +1
1 Solution
 
djcanterCommented:
you will want to be extremely cautious connecting the drive to any other controller as the array config may get wiped from the disk. At this point, i would log into the controller and get array specs, stripe size, etc.  There are other tools ilke raid reconstructor that can reassemble the data. First step will be imaging the drives and recovering the raid from from images.
0
 
andyalderSaggar makers bottom knockerCommented:
You've probablt got a bad block on a remaining drive hence the errors. Sometimes if you replace the failed disk and allow the array to rebuild it will get past the bad block by retrying it a few times and the data will become usable again. Downside is that all the time the remaining drives are running they're under more load than normal and may fail.
0
 
Armin57Author Commented:
The server was in a RAID 1 configuration, if this was a SATA drive, I could just connect it to an external enclosure and will have immediate access to the data.  I'm trying to accomplish something similar.  

If I connect the drive to a different RAID controller, do I have to configure it as a RAID before I'm able to access the data? I just want to boot into a windows using another machine and access the data on this drive without making any RAID configurations to it.
0
Independent Software Vendors: We Want Your Opinion

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
djcanterCommented:
if it was a RAID 1, you can attach it to any sas controller to read data.
0
 
DavidCommented:
If this has the LSI/PERC controller, then there is metadata at physical block 0.  Hook that disk up directly to an ATA controller or USB,  and it will say that the disk is unformatted.

You need to have a controller in the same product family.   Or you can buy something that strips of the metadata, like runtime.org reconstructor ... but that is overkill.

If you are good with C and a binary editor, you can write a program to discover the beginning and end and then manually repair the partition table.  If you have the proper skills, then this paragraph is all you need to know to move forward. (Don't feel bad if you don't, this  is a test).

My suggestion (Assuming this is a Dell/PERC controller)
1. Go to a used equipment dealer locally and bring the busted controller with you.  The Dell/PERC controllers were quite popular, and I would be surprised if you couldn't pick one up locally for $20 or so.

2. Put that same controller and drives in a different computer, boot to whatever you booted to before, and then post what the working controller says about everything.  If this is a redundant array and just degraded, then no need to put in a replacement drive, you can just image the degraded LUN onto a physical disk, or another raid volume using a different controller.

3. If data is valuable, and there is no backup, then consider hiring a pro.  You have not assessed the health of the surviving disks, so there is risk that other drive(s) are dead/dying and every moment they are powered up could be the last.  REBUILDS ARE STRESSFUL!
0
 
Armin57Author Commented:
It's a SAS5i/r controller.

I went ahead and purchased a new controller online, I will have it tomorrow.

Based on this thread:

http://www.experts-exchange.com/Networking/Network_Management/Disaster_Recovery/Q_27815136.html

Once the new controller is installed, it will want to "Activate Array" before it can boot, question is if I only have one drive attached if it will allow me to activate and boot from it?
0
 
DavidCommented:
Since controller is new, it should not have a legacy config in the NVRAM.  It *should*  learn the configuration that is in the metadata, and discover that it was degraded.  If not, you can import and then activate the degraded array.  Activating will be required no matter what, but that is painless and can be done through the BIOS.

If it was me, and since you are ordering anyway, I'd get a replacement drive, or better yet, get two replacement higher density drives.   Chances are the disks are as old as the controller, and that other drive was same batch, same data, same duty cycle, and probably is on it's last leg of life anyway.

 put in 2 new disks in that same controller, build that into a new RAID1.  Then use image or file backup software to migrate the data from the degraded single-disk RAID1 onto the new & improved RAID1.  

(Well, actually I would run full diagnostic suite to assess health of that disk first to see if it is in stress, but you don't have what you need to do that, so just play it safe and plan for the worst and hope for the best)
0
 
Armin57Author Commented:
Put in the new raid controller, went into bios and did an activate on array, came right up.
0
 
Armin57Author Commented:
I've requested that this question be closed as follows:

Accepted answer: 0 points for Armin57's comment #a38313620

for the following reason:

Final solution
0
 
DavidCommented:
Seems to me the author did exactly as I suggested, "Since controller is new, it should not have a legacy config in the NVRAM.  It *should*  learn the configuration that is in the metadata, and discover that it was degraded.  If not, you can import and then activate the degraded array.  Activating will be required no matter what, but that is painless and can be done through the BIOS."

Author put in replacement controller, and activated, and it worked.
0
 
Armin57Author Commented:
Sorry about the confusion.
0

Featured Post

Concerto's Cloud Advisory Services

Want to avoid the missteps to gaining all the benefits of the cloud? Learn more about the different assessment options from our Cloud Advisory team.

  • 5
  • 3
  • 2
  • +1
Tackle projects and never again get stuck behind a technical roadblock.
Join Now