Need help recovering data from a Raid 10

[eddnyc]

The computer is a Power edge 2900 using Vmware Server 3.0.5. on a riad 10.
My friend ask me to recover data from this computer. ( I never heard of Vmwear or raid 10, I thought it was on Windows XP)
I did not know it was in a raid 10 or on Vmwear, so I pulled the first disk out (1 0f 4) then I slaved it to my computer but could not see any data for obvious reason. Now I read up on raids and I'm a little more familiar with it but not enough to recover data. The raid is broken and Vmwear does not boot up all the way. Can some body please help me recover the data?

[BlackST]

The raid is broken logically (structure) or phisically (1 or more drives fail to start or bad) ? In case 1 need some soft tool to do the job, in case 2 first repair the drive(s) then the array will be functional or you'll still need logical reconstruction.

[eddnyc]

How do I repair the drives???

[BlackST]

Er... Brand and model? Can you check them individually for basic functionality?

[eddnyc]

Seagate, Baracuda Es 250 gb

[eddnyc]

let me tell you what I've done. Maybe you can help me this way. I remove hard drive1 in position 0. I disconeted the other 3 but did no remove then, remember the last one is a hot spear. I got a new hard drive installed it in position 0 with Vmwear ESX Server 3i (60 day trail), with raid 0. What I want to do is import disk 2 and 3 (in posotion 1 and 2) wich are in raid 10 to the new raid 0 with out messing up the data.
1. is this safe?? Will it work??
2. how do i do it, what are the steps??

[cgallery]

First you have to test each drive individually to determine whether the drive has any mechanical or firmware problems.

If they work, then you have a logical problem. In that case, you disconnect the RAID from the machine and install a separate boot drive and install the O/S. After you have that working, you add the RAID back into the machine and load the drivers. Now use some logical recovery software to recover the data.

Basically, you can't start recovering the data until you figure out what is wrong.

[wiseleo]

RAID 10 can survive the failure of one drive with no ill effects. If this were in my lab, my course of action would be to image the drives and reassemble them using a software program that is capable of reconstructing RAIDs. UFS explorer is one such program and it's one of the cheapest.

RAID 10 can actually survive the failure of two drives as well, assuming that they do not belong to the same mirror set.

I can treat a RAID 10 array of 4 disks as a RAID 0 with multiple copies of each drive. Since the array actually broke, I think you may be missing an entire mirror. That is basically a RAID 0 failure scenario, which usually means very bad things.

VMware has nothing to do with this. The array is likely to have been built using Dell's RAID card (PERC or CERC) and that absolutely has nothing to do with the ESX. However, since you don't know such basic things, that tells me that you are in over your head. I do not believe that you are qualified to help your friend with this matter.

If you have 4 disks, of which one is a hot spare, that is not a RAID 10 array. It can be a RAID 1, 0, or 5. To have a RAID 10 with a hot spare would require a minimum of 5 disks.

You do not have enough information. It was not a good idea to reconfigure the original array controller. This kind of work is normally done with disk images and software capable of reconstructing RAIDs virtually.

[hhddrec]

I try to recover it with any RAID reconstrucctor too.