my motherboard won't let me have a master without a slave

[hurtstotalktoyou]

THE LATEST PROBLEM:

I just got a new (used, but new to me) setup based on the MSI KT3 Ultra-ARU motherboard.

I tried installing my hard disk on the primary IDE channel as a master with no slave. However, the BIOS persists in reading it as a slave. I have tried both compatible jumper settings ("master" and "cable select"), two different IDE cables, and the secondary IDE channel. Nothing works.

Strangely, though, although my hard disk will not, my CD drive works just fine as a master without a slave.

Any suggestions?


THE OLD PROBLEM (HAS BEEN SOLVED):

I just got a new (used, but new to me) setup based on the MSI KT3 Ultra-ARU motherboard.

I immediately attempted to install Windows XP. However, I keep getting the same error messages. At "33 minutes remaining" the installation Wizard fails to install multiple devices (beginning with "display," I think, but including many other non-related devices such as the hard disk, audio and "computer," whatever that means), prompting me each time. After 8-10 of these prompts, it pauses for a moment, then gives me a blue screen of death, which reads:

KERNEL_DATA_INPAGE_ERROR

...along with a bunch of other probably useless information, of course. I have tried disabling RAID, BIOS cache and shadow.

I don't know what else to do. Any help would be much appreciated.

 

[conlan]

Did you zero out, partiton and format the hard drive?

 

[hurtstotalktoyou]

Originally posted by: conlan
Did you zero out, partiton and format the hard drive?

I know I partitioned and formatted the drive. I do believe I zeroed it out a while back, though I can try it again. I will report back the results.

 

[hurtstotalktoyou]

Hmm. I tried loading a hard disk utility via the floppy drive. It reports a "fatal" error in "absolute sector" something or other. So, to isolate the problem, I swapped out a different IDE cable and connected it to a different IDE channel on the board. The same error occurred.

I am now trying scandisk (through an ME startup floppy). I will report the surface scan results soon.

 

[conlan]

yup, the hard drive maker will have a utility in the download section of their website, ie. Maxblast, Seatools etc.

 

[Jack03]

This error message indicates that the requested page kernel data from the paging file cannot be read into memory. This is usually caused by a bad sector in a paging file or a virus, disk controller error or may be failing RAM. In most cases it is caused when nonpaged pool resources run out. It is also caused by defective hardware.

 

[hurtstotalktoyou]

Supposing my "nonpaged pool resources" are running out, how would I fix that?

 

[Markbnj]

What did your disk test say? It sounds a lot like a bad hard drive to me.

 

[Roguestar]

Run memtest.

 

[KGB]

Originally posted by: Roguestar
Run memtest.

Agreed.

"I just got a new (used, but new to me) setup based on the MSI KT3 Ultra-ARU motherboard. "

Are you certain the HD controller is working correctly if the MB is used?
Are any of the capacitors bulging or leaking? This MB would be of a vintage common to that problem.

 

[hurtstotalktoyou]

I formatted the hard disk to FAT32 using an ME startup disk, then ran a full surface scan using the scandisk utility. No errors were found. I then tried the Samsung diagnostic (shdiag.exe), and all tests (except the surface scan, which I instead ran with the ME disk) passed with flying colors.

But here's an oddity: I have my hard disk connected to the IDE1 master, with the jumper set to cable select. Yet the BIOS tells me it is on the slave.

 

[KGB]

Try setting the jumper to Master or Master Only.

Again, have you inspected the caps for any leakage?

 

[hurtstotalktoyou]

I just tried different jumper settings as well as swapping the IDE channel and cable--nothing works. So now I just have it connected as a slave, with the CS jumper and no master on the channel.

I did a cursory glance for leaking or bulging caps and found none. It is possible one has escaped my eye, though. Next time I open the case I will do a more thorough inspection.

 

[gsellis]

Slow down here folks. Do what Roguestar says and find a different driver if memtest is clean. This is mostlikely device driver/memory problem and no amount of formatting, repartitioning, or drive scans will fix it.

 

[hurtstotalktoyou]

I'm about to run memtest, but I have not just yet.

If it's a driver issue, how do I fix it since I can't even get through a fresh XP installation?

 

[hurtstotalktoyou]

MemTest passed just fine.

 

[KGB]

Do you have a PCI IDE controller card you can try instead of the on-board controller?

 

[conlan]

Originally posted by: KGBMAN
Do you have a PCI IDE controller card you can try instead of the on-board controller?

Agreed, if the HDD passes all tests, you may have an IDE issue

This is not a driver issue

I would set the jumper to Master, hook up the cable as Master as well

 

[hurtstotalktoyou]

Okay, the problem has been solved...sort of.

Apparently the "kernel data inpage error" was caused by an attempt to install Windows XP on a slave drive which had no master. That is, IDE1 had the hard disk as a slave, and nothing as a master, which is not compatible with a Windows XP installation. So I put a DVD drive on the master channel and the installation is going fine, now.

However...

Why can't I have my hard disk as a master with no slave? No jumper setting, cable or IDE channel will allow me to do that. It has me concerned.

 

[conlan]

Have you tried another cable?
Are you sure you have the jumper set to Master?

 

[hurtstotalktoyou]

Yes and yes.