IRQL_NOT_LESS_OR_EQUAL crap

[huesmann]

I'm getting the IRQL crap on one of my machines,

IRQL_NOT_LESS_OR_EQUAL

[bunch of crap]

STOP: 0x0000000A (0x00000016, 0x00000002, 0x00000000, 0x804F43B0)

[bunch of other crap]

Memtest86 seems to be happy, can anyone identify the issue from the codes?

 

[johnjkr1]

Usually it indicates a driver problem. Was anything changed? When does it happen?

 

[huesmann]

I dunno, it just happens spontaneously. Only thing that changed was I took out one of my 2 sticks of 512MB RAM.

I got another one:
STOP: 0x0000000A (0x00000016, 0x00000002, 0x00000000, 0x804FB013)

 

[johnjkr1]

maybe the new ram doesnt like your board, do you have the latest bios? Perhaps the settings are too high?

 

[Smilin]

What OS?

 

[huesmann]

XP, I didn't add any new RAM, it was working fine before and I took one stick out. Board has latest available BIOS, dated 9-16-03 for Epox EP-8K3A.

 

[Smilin]

Look in your windows\minidump folder. See if you have some minidumps available. If you do, zip up the last three or four of them and PM me.

You can't tell a lot from minidumps but I should be able to at least tell you if it's the same one each time (likely a driver issue) or it's random (likely hardware).


Actually, check your PM. I'll send you my email address so you can send them to me.

 

[nanaki333]

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/ddtools/hh/ddtools/bccodes_6vxj.asp

SUMMARY
=======

This article describes how to use a sample debugging session to determine
which specific driver is causing the following error message:

Stop Error IRQL_NOT_LESS_OR_EQUAL (0xA)

SYMPTOMS
========

After you install a driver, the system stops responding and you receive the
following error message in nt!KiActivateWaiterQueue+0x27:

Stop Error IRQL_NOT_LESS_OR_EQUAL (0xA)

The initial stack trace indicates that the problem is in the Fast Fat
driver.

CAUSE
=====

This problem typically occurs because drivers call either the
IoQueueWorkItem() function or the ExQueueWorkItem() function two times on
the same work item before the work item has been executed.

Device drivers that statically allocate either the IO_WORKITEM structure or
the WORK_QUEUE_ITEM structure are particularly prone to this problem. Device
drivers that perform such a static allocation must guarantee that they do
not try to use the statically allocated item while it is already queued.

 

[huesmann]

Yeah but what's it mean? Translate from Microsoftese to plain English.

 

[huesmann]

Got another error:

PAGE_FAULT_IN_NONPAGED_AREA

STOP: 0x00000050 (0xFFFFFFF7, 0x00000000, 0x80596C38, 0x00000000)

 

[KB]

I would swap the memory currently in there with the one you removed.. I have gotten these errors from bad memory. Did you run the most exhaustive memtest level?

 

[huesmann]

I set it to run all tests, I ran one pass of 8 tests, no errors. Should I make it do more passes?

 

[wraith3k]

Check your Event Viewer (under Administrative tools), usually this will indicate what is causing the problem. Or if it's still cryptic, try googling parts from the error to see if you find anything. Also check your device manager for any conflicts or stuff that isn't working.

 

[huesmann]

There were no conflicts in the Device Manager. Will check the Event Viewer when I get home.

 

[Mark R]

Random errors like this do seem suspicious for a hardware fault - it's not clear what component is the problem but RAM is certainly a possibility.

It may be worth trying a different RAM testing program - some are better than others. Memtest86 is certainly among the better ones, but I've found WMD to pick up errors that Memtest86 misses (in this case, I eventually traced the problem to an incompatability between RAM and mobo). Also remember, the more passes you run, the more likely you are to find a strange fault. Memtest86 did eventually find the faults - but it would take 3 or 4 passes before it would appear. WMD found them reliably every pass.

For a CPU test you could try running Prime95 torture test (but run in safe mode to stop extraneous drivers from loading). There are several modes, some which test CPU primarily - and other that test RAM.

Also check the PSU voltages in the BIOS to check that they are satisfactory.

 

[huesmann]

Ran a coupla passes of WMD, didn't seem to have any difficulties.

The computer seems to be stable running in safe mode.

I did check the Event Viewer, but can't really understand any of the gobbledegook there.