Can incorrect HD jumper settings cause corruption??

[Acoustic Jimmy]

Thanks. That's clear.

But I had forgotten all about the 120GB drive that was on the secondary channel for the past year, which doesn't have a limiting jumper on it. Somehow it was always detected. I don't recall how I did it (I installed that drive for my friend).

The 120GB drive is now installed on his new Win2000 machine, and I don't want to jeapordize the data on it by moving it back to this old machine just to test out what's going on here. I may change my mind if I can't figure this out. I at least have to verify that there is/isn't a limiting jumper. It's a Seagate 120GB, about a year old -- anyone know offhand?

And is there any other way other than the jumper that will allow a drive to be detected on a machine that sees only up to 32GB?

TIA.

 

[Nodsu]

The secondary, tertiary, etc drives don't need to be detected by the BIOS. Once the operating system has loaded its own drivers for the IDE controller(s) the BIOS limitations don't matter any more since all communication is done between the drivers and the controller directly.

 

[Acoustic Jimmy]

Maybe they don't need to be detected to allow communications, but 1) the secondary, tertiary, etc. drives will hang the BIOS at detection point if they are larger than 32GB, and 2) they won't show up as a drive in the OS (My Computer) if not detected.

The more important point here is #1... I can't even hit F4 and continue once the system hangs. So it will never get to the OS.

I guess I want to know if the Track 0 Sector 1 information is at all read during drive detection (pre-OS boot).

Or could the CMOS have been patched to allow the 120GB drive to be detected?

Thanks.

 

[Nodsu]

In BIOS setup, set the undetectable drives to "none" so BIOS won't even try to find them. Your OS might still pick 'em up.

No, disk contents are never read during the IDE device detection.

You should see the motherboard maker's website for an updated BIOS.

A separate PCI IDE controller card with support for big hard drives is a solution too - the controller has its own BIOS onboard with no limitations.

 

[Acoustic Jimmy]

I'll try setting them to "none" and see what happens.

I know I can update BIOS or use a controller card; the thing is, the 120GB drive did work in this machine!! Now I'm trying to figure out *how* it did without an updated BIOS or a controller card.

Is there ANY way that the BIOS could have reverted from being able to detected these drives over 32GB (without a limit jumper) to not being able to???

Thanks.

 

[Acoustic Jimmy]

Hey... that worked!

I set the D: drive back to normal "Slave" jumpering, set all else but the C: drive to "none" in BIOS detect, and the drives show up in NT and it's all working! (Picture files which would not open before now open.) So why is there BIOS detection at all if it works without it? Is this an NT versus Win9X difference?

Now the new E: drive does not have any drive overlay software on it because I formatted that myself on another computer... and it is working correctly and at its full capacity. What gives? Why flash BIOS then if the OS works on +32GB drives without the BIOS detection? Again, is this just NT4 that bypasses what BIOS says exists in the system, or do all the OS's do this?

Thanks a million for the help! - I'm extremely happy to have all of my friend's data back (and he will be too)!!

Now if you can clear up my misconceptions, I won't have this confusion again...

 

[Nodsu]

All OSes that have the IDE controller drivers installed can use drives not detected by the BIOS. This leaves only very simple things like DOS, Syllable, Menuet etc that rely on BIOS interfaces. Windows 9x sits on top of DOS, so you cannot install it on a non-BIOS disk nor can you see non-BIOS disks in safe mode AFAIK.

For all OSes, the BIOS detection is needed because you need to boot from a disk and the drivers need to get loaded from somewhere too.

 

[patio]

Which is the basis of the theory in Nodsu's sig...LOL

 

[Acoustic Jimmy]

Well, it's making sense then... except why in so many posts I've read (on various sites) people are concerned (like I was) about detecting their second hard disk, and no one says "Hey, just turn off detection and let the OS find it". All this talk about BIOS upgrading. That is only valid if they have to BOOT from the disk, right?

Now this also brings up a point that makes no sense.... if a drive overlay isn't read during detection and doesn't make an undetectable drive to be detectable, then it is useless on a boot drive, where the BIOS has to first detect the drive to boot from it. True? The BIOS will fail to detect the drive, overlay or not, and that's that. The machine hangs. No boot.

And if you say that a drive overlay is only for secondary, non-boot drives, then I would say it isn't needed there either, since the OS will see the drive fine without it. Like my system.

How do you explain that??? When would one ever need an overlay program like EZ-BIOS or EZ-Drive?


And as for my NT4 D: drive looking corrupted when read on other systems (non-NT systems), please read this:

"Also, if you lost your password on NT - install a new instance of NT, not Windows 2000, as doing so will ruin your old NT installation (because of the difference between the NTFS versions). Same goes for W2K, XP and Windows Server 2003. Always install the same OS."

(emphasis mine)

I haven't had time to investigate this yet, but could this have caused the D: drive to *look* corrupted because the NTFS is different on NT than 200 or XP? Or am I interpreting this wrong?

Thanks.

 

[Nodsu]

Now this also brings up a point that makes no sense.... if a drive overlay isn't read during detection and doesn't make an undetectable drive to be detectable, then it is useless on a boot drive, where the BIOS has to first detect the drive to boot from it. True? The BIOS will fail to detect the drive, overlay or not, and that's that. The machine hangs. No boot.

That's why you have the 32GB jumper And not all BIOSes hang when they find a drive they don't like. More often than not you get a drive detected with some wrong size and that's all you need really - once the BIOS manages to start the boot process, your overlay will kick in and fix everything.

And as for my NT4 D: drive looking corrupted when read on other systems (non-NT systems), please read this:

"Also, if you lost your password on NT - install a new instance of NT, not Windows 2000, as doing so will ruin your old NT installation (because of the difference between the NTFS versions). Same goes for W2K, XP and Windows Server 2003. Always install the same OS."

(emphasis mine)

The NTFS version used in Windows NT is different from the one used in 2k and upwards. Windows 2000 and newer can read the older NTFS just fine. You might want to read this about NTFS compatibility: http://www.pcguide.com/ref/hdd/file/ntfs/verCompat-c.html

 

[Acoustic Jimmy]

So then the overlay is for the boot process only... to access the disk and let the disk drivers be found and loaded. Is that correct?

My experience is that if one uses the 32GB jumper they don't have access to the whole drive-- only 32GB of it. Would an overlay be able to overcome that? I don't think so; it seemed to be that the rest of the drive was "unreachable" from the OS. True, one could boot from it. But I think you'd lose the other part of the drive space. Is this true?

As far as my machine, if Win2000 and XP read NT's NTFS, then that's not my problem. I don't know why it looked corrupted. Would an overlay on the drive be read and implemented even though it's not being used to boot the machine??

I ask a lot of questions, but there's nowhere on the 'Net (that I found) that explains the whole sequence in detail.

Thanks.

 

[Nodsu]

Yes, the 32GB jumper will limit the drive to 32GB no matter what (AFAIK). Apparently there are software solutions to make the hard drive to report itself to BIOS as less than 32GB, but still allow to use its full capacity. Never seen/used one though.

The overlay is loaded only when executing the boot sector, so unless you have told your BIOS to boot from secondary disks, it is not loaded otherwise. You can get weird results from partitioning programs etc if you work on drives (secondary or not) with overlays installed though.

 

[Acoustic Jimmy]

Thanks for the info. That helps.

I'm going to move the secondary drive to another machine and I'll report back what happens. I suspect it will not work in the other machine. However, one of the two machines that indicated that the drive was corrupted was running the partitioning/formatting part of the Win2K installation; so as you were saying about partitioning programs and overlays, that might explain why the drive was seen as corrupted by the W2K install program... or maybe the Win2K install just isn't sophisticated enough to understand the NT4 NTFS. That difference in the NTFS's might also explain why the file system was coming up as HPFS, not NTFS, in certain utilities.

Hey, is there any way I can tell (verify) if an overlay really is on a drive? I mean without any prior knowledge of it... like a program which will detect that an overlay is in place.

 

[Acoustic Jimmy]

It's been a long time since I haven't followed up, so here is what I know and would like to know:

The hard disk in question can be read on my NT4 computer, no problem.

It cannot be read on any other computer; it is seen as not being formatted.

I suspect that it is in HPFS format. A disk utility read the file system as HPFS, not NTFS. It is a secondary drive, not the boot drive, so HPFS is not out of the question. It is an IBM disk, which may be more than just a coincidence (HPFS is an IBM file system).

Any guesses as to what the problem really is, or if an HPFS disk would behave like that?

Thanks for any help anyone can give me.

 

[Nodsu]

Linux can read and write HPFS partitions no problem. So just boot Knoppix or INSERT and see if the drive works there.

 

[Acoustic Jimmy]

Nodsu.... I just tried that now... neither OS can mount the drive. But NT4 happily reads the drive as if it were an ordinary file system.

There must be a more direct way to know what the file system is. Isn't there a utility out there that can tell me this?

Or is there a command (or GUI action) in NT that will reveal the filesystem?

 

[Acoustic Jimmy]

OK, Nodsu, I finally have a real explanation for this mystery....

What is true in fact is that the file format IS indeed NTFS, however the whole drive is an extension of a Volume started on the first physical disk. That first disk has two partitions, one FAT 16 (#1) and one NTFS (#2). The NTFS (#2) partition is the first part of logical drive D: and the whole second physical drive is the rest of drive D:.

The second physical drive is the one that could not be read outside of its home on the NT4 machine. I suppose that there is a reason that the "Partition Type" byte in the partition table is not 07 for NTFS (it is 05) and that the reason has something to do with the drive being part of an Extended Volume Set. But I'm not sure how it works/why it is that way. Maybe you can tell me.

But in any case, when the drive is married to it's mate on the NT4 system, the OS has enough info to be able to read everything fine. And when the drive is separated from that necessary info (almost certainly on the first drive) then the Format Type has no meaning to the OS that's trying to read it. So apparently the OS then declares the drive space "Unformatted".

Any thoughts/comments, anyone??