Ultimate Boot CD

Long story short: I have 2 internal hard disks, one WD, one Seagate, which were working perfectly, and I was looking to dual boot XP and Ubuntu. After Gparted failed abysmally to create the necessary partitions (crashed each time from the Ubuntu boot cd, and later from the Gparted live cd), I used the XP install disk to create the necessary XP partitions and left appropriate space for the linux ones. XP was working perfectly, as were both drives, so far so good.

I turned to some of the UBCD utilities to solve the problem, and used PTDD Super Fdisk in particular. Upon loading it told me that there were some errors on the Seagate I was trying to write the ext3 partitions too, and I assumed that they were the cause of my problems with gparted... I opted to correct them. I do not recall making any alterations to the WD drive with my XP installation on it.

I reboot, having just done something simple like creating a fat32 partition in the space instead of an ext3 (I think), and am having all kinds of problems. I think the program goofed around with the MBR, but seemingly for both drives (even though I'm 99% certain none of the options should have affected the WD with XP on it).

Any ideas how this happened? And what the fix might be? I do have most of the data backed up, but there's possibly a small proportion of valuable data left on there that I'd like to retrieve (or at least, I'd like to confirm I've got it all). I just don't understand it... I was very careful, and all of the operations seemed fairly straightforward... except for those "error corrections" PTDD performed on startup.

Thanks for any help,

Paraselene.

Update:

Fixmbr, fixboot, bootcfg /rebuild from XP's recovery console won't repair one of the drives.
Formatting won't repair one of the drives.
Low-level formatting (from Diskwizard, UBCD version) seemingly won't repair the drive. It crashes at some point during the process.

wtf has this thing done to my system?

Testdisk (UBCD) shows: "Incorrect number of heads/cylinder 16 (NTFS) != 255 (HD)"
Some of the messages from the prompts PTDD was showing me made alterations to this information, iirc. Shouldn't the low level format have wiped out all the changes PTDD might have made?

Could it have conceivably written new information to the bios? One of the first things I did was clear the CMOS, and that certainly hasn't helped. I'm starting to worry now... I can't see how to fix the problem.

The only bright side is that the disk surface is fine, according to the seatools on UBCD.

Please, desperately need your help... can't even get this thing to run a completely new installation of windows.

Paraselene.

My suggestion would be to disconnect 1 of the drives and see if you can get the other loaded to begin with. Before you begin loaded run the Manufacturer diagnostic program on the connected drive and see if it gives you any information that is useful and post back with detailed information.

I managed to both resolve the problem and identify what was causing it. I share this information now, and a recommendation to remove PTDD Super Fdisk from the UBCD for the reasons below.

PTDD's Super Fdisk "error corrections" were, in fact, the cause- they altered the drive geometry and subsequently caused my bios problems identifying the drive information correctly (oddly enough for both drives on the IDE channel, even though I can confirm that I did, in fact, only alter the geometry for one of them). The critical clue was the Testdisk prompt:

Testdisk (UBCD) shows: "Incorrect number of heads/cylinder 16 (NTFS) != 255 (HD)"

The particular way I resolved the problem was to first change my bios settings to LBA, and then run a battery of fixmbr, fixboot, chkdsk /r and formats via the XP recovery CD... annoyingly enough, I forget the exact sequence. An easier method of reparation may have been to use Testdisk's geometry correcting facility, as detailed on their wiki.

Once I'd resolved the problems with the Seagate drive, I plugged the WD back in, set its bios settings to LBA, replaced a corrupted .dll and it's been fine ever since.

I hope that's helpful if anyone else encounters the same problem. But yes, given the very deceptive "error correction" prompts of the PTDD software and the difficulties it can cause (you wouldn't believe how little information there is about this- the Seagate techie I spoke to hadn't ever heard of legacy software being able to do this), I would recommend removing it from the UBCD.

Paraselene.