Iomega StorCenter RAID Failure

Be specific, "see" the data?  Does that mean you can read 1s and 0s from every block in all HDDs, or that R-Studio has determined the block sizes, drive ordering, but don't know specifics of file system parameters and how to mount them?

What do you mean by "IOMEGA" format.  There is no such thing as an IOMEGA format, or more correctly, that terminology is rather unique, as IOMEGA does not have a proprietary file system.

Two things make me believe the data exist:

First, I can still see the shares when the drives are in the Iomega StorCenter.  However, I cannot access them as they have permission issues.  I am unable to adjust permissions since the firmware is not allowing the OS to load all the way.

Second, when I scan the disk with R-Studio (http://www.r-studio.com/) it shows that there is data on the map.  However, it doesn't recognize the format which is leading me to believe it doesn't understand how to read whatever format it is that Iomega uses.

See attached for Scan Information and Partition layout.  I have reason to believe the first set of data is the firmware partition (20gb) and the second is the user data.  The user data is what I'm attempting to retrieve.  The drives are the ST31000520ASCC38
drives.jpg

Seeing the shares can mean as little as the first 64KB of the filesystem is good on the first drive in the RAID5 set, and the rest of the disks are scrambled in wrong order, or even the block size is too large.

Can you mount read/only and read the entire contents of all files and are they corrupted?  IF not, then you haven't reconstructed properly.

So pick a few files > (256KB * (nDiskDrives - 1)).   Maybe jpg, pdf, movie, something that you can pull up and easily confirm if it works.   Having a file this large insures that stripes at least are correct.  Mount read-only to avoid permission problems.

Looked at the image.  you do NOT have it reconstructed properly.  Could be any or all of wrong stripe size, incorrect drive ordering.   Starting ending blocks for partitions are going to be wrong because they can't be right if you don't get the first two things correct.

P.S. Exactly what model do you have?

I understand what you saying.  If I go to the Parents tab, seen in the screenshot, the alignment appears correct.(on the left).  The model I'm working with is IX4-200d.

Please forgive my ignorance on this topic as it is not my specially.

In regards to your comment "So pick a few files > (256KB * (nDiskDrives - 1)).   " I am a bit lost.  

I am seeing some files, some of which open ok and some are scrambled.  I am looking for directories and not finding it in the search.  Furthermore, I'm not sure if it is a limitation to the current software, but I do not see most of the .pub files that I know exist.

I do not know how to mount these drives without messing them up.  I believe they are of a Linux format ant I'm on windows at the moment.

Here was the drive list I tried to attach
drives-list.jpg

Well, bottom line, I can't walk you through this.  First, you need to be using LINUX.  You need to have scratch disks and image copies of all drives.  You'll have to run diagnostics to assess health and potentially root cause.  You may have had a file system failure and the crappy kernel wrote a new file system on top of what you had.

I'd have to look at hex dumps of partitioning, and even look at the raw data to learn what is really there.

R-Studio is a pretty decent consumer product, but if it just doesn't automatically work out-of-the-box then you'll just have to pay somebody to look at it, as this takes a lot of experience and software that does things RStudio does not.

Sorry.

I have tried running knoppix live and looking at the partition.  However, I didn't know how to mount them up.  I suppose it is time to read up some more.  Thanks for your assistance

Don't even *think* of doing this unless you make image backups.  You can use mdadm to reconstruct, but you'll have to figure out the ordering and stripe size by looking at the raw data itself.

Own answer