George,
Thank you for the excellent information, which I hope very much my health will allow me to examine tomorrow. I'm afraid my interest in your problem has been not because my knowledge of modern disk & file structures is excellent, but because they are poor; and this makes me want to learn about them - a problem is the best way. So, I feel as if I may be exploiting you rather than helping you.
RAID came out about when I retired, and I was unsuccessful in convincing a client to implement it on their internet server. Striped RAIDs & demand paging seem an elegant way of more closely 'balancing' the information traffic in a multitasking operating system, and I had always wanted to implement it.
Because of my delays and desire to study your problem as much as solve it, I should now offer you whatever advice I currently have. Your problem is: What's wrong with this disk? Why did it fail as part of a mirrored array? When I consulted, I recommended only the most boring applications I could, old applications. This is because mixing & matching parts & software always produces problems; and if thousands of people encountered these a year ago, they've likely all been fixed. Hence, my systems were dull, but reliable.
Your disk differs from the others in its model number, but it may have differed in other ways when first used. You then took it home & possibly partitioned it, certainly formatted it for an Apple computer. Then you, quite reasonably, used two (or three) applications that should have given you write permission to the disk's Microsoft partition (or directory). Unless each application anticipated your use of the others, and anticipated your using some disk structures that are artifacts of a RAID array on a 'plain' Mac, interactions could be unhappy.
Though I've not yet studied the details of RAIDs, certainly the controller handles the gritty details & the driver should accept normal input; except, it allows you to choose the type of array you want. One might expect data structures on a RAID disk to be unusual, even the partition table: for it would be simplest to use a partition table designed to work for both a striped array and a mirrored array as well - this means unusual. We also know that Microsoft partition tables are difficult to remove, and hfs_fsck didn't expect to see yours.
The first thing, of course, would be to examine your name, groups, & permissions accorded you; then use both Aqua & the terminal to look for any differences in ownerships or permissions in the files, directories, & volumes containing files you can access & files you can't. If there are different volumes, check where they are mounted and the permissions assigned the volumes.
The second thing would be to definitively identify your current partition table, which may be the same as that it had on the server. Next, definitively identify the byte-order used. Professional applications for Unix or Linux may easily assume big-endian order; and Microsoft would assuredly use little-endian.
Failing that, or skipping that, you want to be sure you can overwrite all data structures on your disk with those of your choice. If your disk is ATA, a free government security erasure program may be overkill, but it is a sure bet. Even if you find the identification of obscure disk structures a problem, recording the content of the first several blocks of the disk will allow you to assure yourself you have changed the disk's structures (boot code, partition table) to those of your choice.
If this problem isn't urgent, you can study it (as I shall); if it is, you may either need to stuff your data somewhere and be certain the disk is repartitioned & reformatted solely for the Mac of your choice; or pose your problem to manufacturers & forums of professional unix administrators.
Restructuring the disk for your Mac, you could then test the disk to assure yourself the original problem did not lie with a disk failure of any kind. If the disk's manufacturer assures you the two models are compatible in an array, the problem would then likely prove an intermittent one with your server. (This might cause you to change your backup strategy for a while.)
My apology for delays & the above 'arm-waving' strategy.
Bruce
