Another random BSOD thread

[Lower]

I've been having sporadic, random blue screen crashes and thought you all might be able to help. This is with my ASUS p4B266 running a 1.6A @ 2.4 (I have since returned the clocking to normal, and loaded MB setup defaults). I have a Geforce TI4600 in the AGP slot, an SB Audigy in Slot 2, and a NIC card in slot 4. I have a 120 GB WD as primary master, a DVD-Rom as secondary master and a CDRW in secondary slave.

It started about a month ago - Several times a day IE would crash on me, with either with an error related to flash.ocx, mshtml.dll or iexplore.exe. Varied and random, but no blue screens -just IE crashing. I read through all of MS technet and MS knowledgebase, but nothing I did helped the trouble -I was still crashing 2-3 times a night.

Screw it, I thought -maybe XPSP2 will fix what ails me, and I held out for as long as I could. I didn't make it to XPSP2 release, but decided I'd attempt to "repair" the XP install (NOT via the recovery console). After attempting a repair of the install, I couldn't even boot into windows (either regularly or safe mode). The reinstall borked something. I reimaged the drive back to my year+ old image (known good, stable, and ready to rock).

The next day I was back to the same old same old. IE crashing.

About a week later the BSOD's started happening.
0x0000000A - IRQ not less or equal , 0x000000C2 - bad pool caller , 0x00000050 - fault in non paged area , 0x00000046 deref unknown logon session , various 0x0000008E win32k.sys errors. You name it, I've probably seen it.

I installed XPSP2, hoping it might sure the troubles -no change.

Again, I read through a ton of stuff -the majority suggesting system RAM or hardware RAM (motherboard cache, video RAM, hard drive cache). In an attempt to narrow the field, I ran 17 loops of Memtest86, and 8+ passes of DocMemory - NO ERRORS were found.

I tested out the WD 120GB drive through their utils using both the quick and the thorough tests to see if the drive was bad. Again, no errors found. I read more and more information about whichever stop error my system had most recently spit out, and came to the conclusion that my ASUS P4B266 motherboard was fading fast. I found a replacement P4B266-c and threw it in (it was a direct replacement as all the onboard extras of the P4B266 had never been enabled). No change. Money poorly spent.


I've updated everything I can, and generated a CAB. I'm still working up a few more minidump files.

 

[dclive]

Originally posted by: Lower
I've been having sporadic, random blue screen crashes and thought you all might be able to help. This is with my ASUS p4B266 running a 1.6A @ 2.4 (I have since returned the clocking to normal, and loaded MB setup defaults). I have a Geforce TI4600 in the AGP slot, an SB Audigy in Slot 2, and a NIC card in slot 4. I have a 120 GB WD as primary master, a DVD-Rom as secondary master and a CDRW in secondary slave.

It started about a month ago - Several times a day IE would crash on me, with either with an error related to flash.ocx, mshtml.dll or iexplore.exe. Varied and random, but no blue screens -just IE crashing. I read through all of MS technet and MS knowledgebase, but nothing I did helped the trouble -I was still crashing 2-3 times a night.

Screw it, I thought -maybe XPSP2 will fix what ails me, and I held out for as long as I could. I didn't make it to XPSP2 release, but decided I'd attempt to "repair" the XP install (NOT via the recovery console). After attempting a repair of the install, I couldn't even boot into windows (either regularly or safe mode). The reinstall borked something. I reimaged the drive back to my year+ old image (known good, stable, and ready to rock).

The next day I was back to the same old same old. IE crashing.

About a week later the BSOD's started happening.
0x0000000A - IRQ not less or equal , 0x000000C2 - bad pool caller , 0x00000050 - fault in non paged area , 0x00000046 deref unknown logon session , various 0x0000008E win32k.sys errors. You name it, I've probably seen it.

I installed XPSP2, hoping it might sure the troubles -no change.

Again, I read through a ton of stuff -the majority suggesting system RAM or hardware RAM (motherboard cache, video RAM, hard drive cache). In an attempt to narrow the field, I ran 17 loops of Memtest86, and 8+ passes of DocMemory - NO ERRORS were found.

I tested out the WD 120GB drive through their utils using both the quick and the thorough tests to see if the drive was bad. Again, no errors found. I read more and more information about whichever stop error my system had most recently spit out, and came to the conclusion that my ASUS P4B266 motherboard was fading fast. I found a replacement P4B266-c and threw it in (it was a direct replacement as all the onboard extras of the P4B266 had never been enabled). No change. Money poorly spent.


I've updated everything I can, and generated a CAB. I'm still working up a few more minidump files.

I can only comment on your most recent crash with the minidump.

Get rid of this (ie uninstall it - *not* just disabling it!) and try again:

Module[144] [C:\WINDOWS\SYSTEM32\DRIVERS\KLIF.SYS]
Company Name: Kaspersky Labs
File Description: spuper-ptor
Product Version: (6.12:10.38)
File Version: (6.12:10.38)
File Size (bytes): 114960
File Date: Tue Oct 14 12:46:18 2003

In other words, *completely* uninstall this product.

That may solve one crash you got, but it sounds like you've got a hardware problem given the variety and the fact it happened after ghosting a known-good image. I can't help with that.... about all I can suggest is that you contact your hardware vendor for a full pc diagnostic test.

 

[Lower]

Since I can't do anything major on this system, I decided to keep on running the Memtest86 on the system.
Now I actually get 1 error on each stick. (one stick at 486MB test 4 pass 3, and one stick at 265MB test 4 pass 9)
What I'm seeing is indicative of RAM errors to start with, isn't it?
Should I start there? I mean I would assume RAM should be error free with NO exceptions... right?

Since I'm running 2 stick of 512MB, what are the odds that both sticks went bad at the same time?

Could this be an undervolt issue? Will upping the RAM voltage solve error trouble?
Will upping the FSB increase errors?

Should I rerun this test without the cache enabled to rule out CPU cache as the failing part?
(Memtest states:
"Please be aware that not all errors reported by Memtest86 are due to bad memory. The test implicitly tests the CPU, L1 and L2 caches as well as the motherboard. It is impossible for the test to determine what causes the failure to occur. However, most failures will be due to a problem with memory module. When it is not, the only option is to replace parts until the failure is corrected.")

My other options might be failing power supply or a trashed CPU. Seems like maybe a failing PS might thwart RAM tests, and I know a bad CPU woudl certainly cause trouble. How can I rule these in or out?

 

[dclive]

You gave me a dump, and on the first I asked you to uninstall the Kapersky program. Did you have a chance to do that, and did that decrease the frequency of the crashing? If not, please send me a new dump.

 

[Lower]

I did uninstall it. The problems worsened until I reinstalled from image.
I took our last exchange to mean there was nothing you could do for me, and didn't save any of the numerous dumps I built up.

Strange thing is no matter what A/V I run, it seems to eventually get corrupted in the state I'm running. I'm back to Norton now, but I am currently running on a non-updated system, as I continue to reformat (completely) or reinstall (from image).

 

[dclive]

Originally posted by: Lower
I did uninstall it. The problems worsened until I reinstalled from image.
I took our last exchange to mean there was nothing you could do for me, and didn't save any of the numerous dumps I built up.

Strange thing is no matter what A/V I run, it seems to eventually get corrupted in the state I'm running. I'm back to Norton now, but I am currently running on a non-updated system, as I continue to reformat (completely) or reinstall (from image).

I don't understand. I got a dump, suggested that you uninstall Kapersky - and what is the latest status? I understand that problems got worse, and that did not help, so what new dumps were then generated?

If you install a clean OS with no additional drivers or anything non-Microsoft, does the machine work properly?

 

[Lower]

I had uninstalled Kaspersky, installed Norton A/V in it's place and continued to crash.
I formatted and installed fresh, and continued to see the similar errors, so I went back to the imaged installation. (So basically No -the errors are there even on a fresh MS only install -another sign of hardware trouble, I'm sure).
I had not updated my install -but I am currently updating it.

I've been running additional tests. The WD disgnostic test STILL comes back clean for both quick and thorough tests, but Memtest86 has now shown erros in BOTH sticks of RAM (I had them in seperately). Both errored out during test 4. One errored out 3 passes in, out of 25 run, and one 9 passes in out of 52 passes.


The dumps I sent were from today (I just got another I'll send too).

My concern/question now is: Is the RAM truly bad, or is a flaky Powersupply and/or fault CPU cache (again, this is a 1.6a that's been run at 2.4 for 2+ years), the true issue and it's just coming to light via the Memtest86 tests. (This particular test uses cache, so that raises my wonder).

 

[dclive]

Originally posted by: Lower
I had uninstalled Kaspersky, installed Norton A/V in it's place and continued to crash.
I formatted and installed fresh, and continued to see the similar errors, so I went back to the imaged installation. (So basically No -the errors are there even on a fresh MS only install -another sign of hardware trouble, I'm sure).
I had not updated my install -but I am currently updating it.

I've been running additional tests. The WD disgnostic test STILL comes back clean for both quick and thorough tests, but Memtest86 has now shown erros in BOTH sticks of RAM (I had them in seperately). Both errored out during test 4. One errored out 3 passes in, out of 25 run, and one 9 passes in out of 52 passes.


The dumps I sent were from today (I just got another I'll send too).

My concern/question now is: Is the RAM truly bad, or is a flaky Powersupply and/or fault CPU cache (again, this is a 1.6a that's been run at 2.4 for 2+ years), the true issue and it's just coming to light via the Memtest86 tests. (This particular test uses cache, so that raises my wonder).

OIC; as long as the information about the crashes when no non-MS software is installed is correct, it's hardware, and there's not much I can do... sorry.

 

[Lower]

Ok, that's what I thought -and what you had said in the first reply, so my bumping and posting of latter messages were more for someone, anyone to deny or confirm:

My concern/question now is: Is the RAM truly bad, or is a flaky Powersupply and/or fault CPU cache (again, this is a 1.6a that's been run at 2.4 for 2+ years), the true issue and it's just coming to light via the Memtest86 tests. (This particular test uses cache, so that raises my wonder).

dclive, thanks a bunch for your help!
Anyone want to give input on the above?