Data Recovery and Disk Utilities Forum @R-TT

It surely would, but my point is little bit different. Better consider that head swap right now, especially look at at its price.

Alt wrote:I don't think the heads will cure themselves in the course of time.

Wouldn't it be nice if they did though? But to be clear, I meant hold on to it for a bit before deciding whether to do the head swap with the blank donor drive that I do have.

I don't think the heads will cure themselves in the course of time.

Data-Medics wrote:

gat0rjay wrote:The failed drive was a 3TB with 700GB used. It would spin up, make a few rather awful sounds (seems like head seek errors), then stop - can't be recognized in BIOS, much less in an OS. I do have a matching possible donor drive that is blank - but I'm not sure it's worth paying someone to do the head swap.

Let me guess, ST3000DM001 model drive?

Haha, yes, it sure is. Sorry for the delayed reply, I had set this aside for a bit. I think I'm going to hold on to the drive/data for a little longer and maybe see if an image can be made of it for recovery within RTT. It may be worth it if it allows me to recover a lot of the media I have accumulated over the years.

gat0rjay wrote:The failed drive was a 3TB with 700GB used. It would spin up, make a few rather awful sounds (seems like head seek errors), then stop - can't be recognized in BIOS, much less in an OS. I do have a matching possible donor drive that is blank - but I'm not sure it's worth paying someone to do the head swap.

Let me guess, ST3000DM001 model drive?

Hi Alt - thanks very much for your input. I didn't know WSS worked like that. When I originally set up and created the WSS pool in question I actually added drives sequentially - (the drives were all in a WHSv2 system previously). So I started with 2 empty drives in the WSS pool, then I dumped all the data from a drive on my WHS system to the pool, and after that was done I pulled the drive and moved it over to the WSS pool (so now there were 3 drives in the pool). I continued doing that until I had migrated all data/drives from the WHS machine to the new WSS pool.

The failed drive was the last, or second to last, drive I added (so I assume a majority of my data should be intact since that drive wasn't present for most of the file copies). I'm not sure if WSS writes data the same way with so many drives involved (it seems like it would sometimes pick a drive that was less full and use it exclusively until it reached a certain used space threshold). But there's probably no way to tell now.

@Data-Medics - thank you once more for your input. Yes, the drives in the WSS are of varying size (2TB - 4TB). The failed drive was a 3TB with 700GB used. It would spin up, make a few rather awful sounds (seems like head seek errors), then stop - can't be recognized in BIOS, much less in an OS. I do have a matching possible donor drive that is blank - but I'm not sure it's worth paying someone to do the head swap.

@Alt again - and on that note, I appreciate your link to the imaging link, but with the current state of the drive I don't think that's possible.

Thanks to both of you for all the input, it's been helpful.

If it's possible to image the failed drive somehow, you'll be able to add the image to the WSS in R-Studio. See R-Studio's help for details: Windows Storage Spaces.

Based on what Alt just mentioned, is it safe to assume that the drives were of varying sizes?

Another question, how badly did the one drive fail? Is it clicking, grinding, not spinning, etc.? Perhaps it's still possible to get the data extracted from it and have a 100% recovery.

Actually, a "simple" WSS is a kind of RAID0, but a kind, not a pure RAID0. For example, you have 3 disks, one 700 GB, two 2TB. You create a simple WSS from those disks. Windows will distribute stripes to the first, second, and the third disks. When space on the first disk is over, it distributes stripes to the second and third disks. When the first disk fails, you loose the data which stripes were distributed over the 3 disks, whereas the data which stripes distributed over the second and third disks is intact. Overall, you loose 2.1 GB of data, and 2.6GB of data will be good.

Hi, first of all thank you for your reply. Your first question is also my first question...

I do not know how a Microsoft Simple Storage Space is implemented. When I originally set it up (it's just a home server to store non-critical media) I ignorantly assumed it was similar to JBOD or WHSv1 drive pool (where one drive could fail, but other drives and their data would still be accessible - even if only by pulling them and reading the data off each drive individually). When I realized this may not be the case, I started migrating data, but the pool failed prior to this being realized - Murphy's law and all...

I too assumed if it were a completely striped RAID0 style array that recovering any data over a few KB would fail. However, since I've recovered thousands of multi-MB files (many in the few hundred MB range), and even a few multi-GB files, I was beginning to think that RAID0 structure wasn't the case. Like I said, there's approximately 20TB of "good" data versus 0.7TB of data on the failed drive in the MS simple storage space.

What I really want to know is if there is any way for R-Studio to determine if a large media file is broken - without having to open/test each file individually. I had a couple files that alerted me they were not entirely readable during recovery - so I'm wondering why R-Studio doesn't give me a similar warning on broken media files?

With the amount of media data I have it's certainly worth a bit of effort on my part if there's a chance I can recover a good portion of it. However, while recovery would be great, if I know that's not a possibility I'd rather know now so I can just wipe the drives and start over.

I'll may just have to do some trial and error testing.

Thanks in advance for any additional input.