XP SP3 crashing with Event Category: (102), Event ID: 1003

Another full dump file.

0x8E errors on both and nothing definitive, only the OS driver win32k.sys.

As before, 0x8E errors are usually caused by hardware. What is your system's BIOS, ie., Phoenix Award, etc.?

BIOS is System BIOS and Setup Utility, 1982-2008 from HP, 786B2 version 2.44.

I did update the NIC card driver, Broadcom Netxtreme Digital Audio version.

Nuts. Often companies like HP, DELL, and Gateway will limit any access to the BIOS. I went back a re-read some our earlier discussion and I noted that you were not able to access the BIOS to determine the voltage setting for your memory, or am I wrong?

That's right there was no setting at all and any reference to voltage.

Another dump, this time with a 0X00000007F and it's crashing in Safe Mode whenever I copy files either to a USB or via the network. Any hope?

0x0000007F: UNEXPECTED_KERNEL_MODE_TRAP
One of three types of problems occurred in kernel-mode: (1) Hardware failures. (2) Software problems. (3) A bound trap (i.e., a condition that the kernel is not allowed to have or intercept). Hardware failures are the most common cause and, of these, memory hardware failures are the most common.

Nothing definitive as way of drivers, etc. However, within both dumps EXCETION_GP_FAILURE was noted and one possibility is that this can be due to faulty memory.

I went back and reread this thread. You ran memtest once for 3 Passes and then again for 4 Passes for a total of 7 Passes. I missed this earlier and that is not how it needs to work.

Memtest needs to run a mininimum of 7 Passes in one testing. Many people think that after running the minimum of 8 individual tests that constitute one Pass that the test is only repeating itself. Nothing could be further from the truth because the test throws at the memory differeing patterns all the time that is why a minimum of 7 Passes is essential.

I have run this test myself and my RAM together failed in the first Pass but running individually per stick I didn't get errors until the 7th or even 8th Pass. I know those who didn't get fails until the 10th, 15th, 19th, and one person until the 26th.

I strongly suggest you start this test again 2 hours before going to bed and check the next morning. In your case I would not stop this test until at least 10 Passes have occured.

I have run the MEMTEST overnight twice and never did an error show. Not sure how many times it ran but it was for over 12 hours.

I went back to the original memory and right away got this even in Safe Mode:
The computer has rebooted from a bugcheck. The bugcheck was: 0x00000050 (0x80f19dc1, 0x00000001, 0xf73fcac1, 0x00000000). A dump was saved in: C:\WINDOWS\MEMORY.DMP. And here it is.

You are going to find this hard to believe but the 0x50 error specifically cited corrupted memory are the cause of your system crashes. On the rare occasions bad RAM has been known to pass Memtest.


The other 0x50 error cited Mup.sys which is a Windows driver. See this link to gain more information: http://www.file.net/process/mup.sys.html

In my research the advice was to run a Repair but one person mentioned he got this this error and Mup.sys and came to be a bad memory slot on his motherboard and once he switched the memory to another slot all was well.

How many memory slots does your motherboard have?

4 slots, 2 blue, 2 black. That's what I keep thinking that it's a MB problem. But if it is a bad slot would MEMTEST show which?

The latest crash had 4 512 MB RAM's from HP, the originals.

The earlier crashes had replacement Kingston 1 GB chips.

I think this is going to be the "switch around and see what works" diagnostic step. Not very sleek but it works. :blush:

I wish I could see how the manual designates the memory configuration for your board. I take it the Kingston is in a 2x1gig set, correct?

If so where are they placed in the motherboard, both blue or both black?

Can you link me to your system on HP's website?

HP D530 CMT drivers link.

Yes they come in pairs and need to be installed in the corresponding blue & black slots respectively, but it doesn't much matter since the 4 are all the same.

See attached pic of the slots w old mem installed.

I have MEMTEST running now on the old memory, and I'll run it all night.

I believe I found the correct manual for your system (not 100% sure though). Can you link me to your Kingston RAM?

KTH-D530/512

but on Kingston's site they have KTH-D530/512S but my chips do not have the last "S" on them.

Note MEMTEST ran 39 times over night and passed with no errors no the OLD HP memory.

And you are getting crashes even with your old memory in, correct?

Yes already 2 today after running MEMTEST 39 (!) times with NO errors.

Download the mini and memory dumps here.

Did the dumps provide any info? I'm leaning towards a logic board or slot problem...

Your larger dump was corrupted and thus unreadable.

Your second minidump is 0xD1 and these are usually caused by faulty drivers. In this case the Windows driver ndis.sys was cited as the cause of your issues. usually OS drivers are too general to be of much help but in your case it might be helpful.

The NDIS library (Network Driver Interface Specification) is contained in ndis.sys. It is a collection of routines that applications can invoke to perform network-related operations. Ndis.sys is an integral part of the kernel and it must match other kernel components to work properly.

Further research shows that one way to rectify ndis.sys BSODs is to either update NIC card drivers or in some cases replace a defective NIC card.

Update your NIC card drivers.

I had done that recently as well. It's a built in:
Broadcom NetExtreme Gigabit for hp
driver date 7/30/2008
version 10.86.0.0

For now I replaced it with an older 100MB NIC to see what happens.

Good move. I'll be very interested to see the results.