BSODs galore

[jmack]

All,

I recently risked building my own computer, which worked great for a few weeks. Unfortunately it has begun experiencing increasingly frequent BSODs of all sorts. I could sure use the help of the experts here on the forum for diagnosing the problem(s) behind the dreaded blue screens.

Computer Specs:
CPU - Intel Core i7 920 (2.67GHz)
RAM - 6 GB (3 DDR3 DIMM's @ 1333 MHz)
OS - Windows Vista Home Premium x64
Mobo - EVGA x58 SLI
GFX - Nvidia GTX 285
Antivirus - Norton 360

I've tested the hard drive and memory and didn't have any problems returned. I've also followed the instructions on jcgriff2's blog and run the latest minidump through the debugging tool (results pasted below). There are 12 minidumps showing in my windows/minidump folder dating back to September. BSODs have included the following:

Page fault in non paged area;
Pfn list corrupt;
IRQ less than or not equal (or something similar);
system service exception (the minidump below); and
a buffer overrun of some kind.

I appreciate any help you all can provide.

Sincere thanks,
Jason


Microsoft (R) Windows Debugger Version 6.11.0001.404 AMD64
Copyright (c) Microsoft Corporation. All rights reserved.

Loading Dump File [C:\Windows\Minidump\Mini112409-02.dmp]
Mini Kernel Dump File: Only registers and stack trace are available
Symbol search path is: SRV*c:\symbols*Symbol information
Executable search path is:
Windows Server 2008/Windows Vista Kernel Version 6002 (Service Pack 2) MP (8 procs) Free x64
Product: WinNt, suite: TerminalServer SingleUserTS Personal
Built by: 6002.18082.amd64fre.vistasp2_gdr.090803-2339
Machine Name:
Kernel base = 0xfffff800`01e57000 PsLoadedModuleList = 0xfffff800`0201bdd0
Debug session time: Tue Nov 24 18:24:01.733 2009 (GMT-7)
System Uptime: 0 days 0:50:53.548
Loading Kernel Symbols
...............................................................
................................................................
...........................
Loading User Symbols
Loading unloaded module list
.......
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
Use !analyze -v to get detailed debugging information.
BugCheck 3B, {c0000005, fffff80001e85120, fffffa600958e0f0, 0}
Probably caused by : ntkrnlmp.exe ( nt!ExQueryPoolUsage+e0 )
Followup: MachineOwner
---------
6: kd> !analyze -v
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
SYSTEM_SERVICE_EXCEPTION (3b)
An exception happened while executing a system service routine.
Arguments:
Arg1: 00000000c0000005, Exception code that caused the bugcheck
Arg2: fffff80001e85120, Address of the exception record for the exception that caused the bugcheck
Arg3: fffffa600958e0f0, Address of the context record for the exception that caused the bugcheck
Arg4: 0000000000000000, zero.
Debugging Details:
------------------

EXCEPTION_CODE: (NTSTATUS) 0xc0000005 - The instruction at 0x%08lx referenced memory at 0x%08lx. The memory could not be %s.
FAULTING_IP:
nt!ExQueryPoolUsage+e0
fffff800`01e85120 8379e400 cmp dword ptr [rcx-1Ch],0
CONTEXT: fffffa600958e0f0 -- (.cxr 0xfffffa600958e0f0)
rax=0000000000020c00 rbx=0000000000000000 rcx=63fb8336746efb83
rdx=fffffa600958ea7c rsi=0000000000000000 rdi=000000000b96ebe0
rip=fffff80001e85120 rsp=fffffa600958e950 rbp=fffffa600958eca0
r8=fffffa600958ea78 r9=fffff80001fe8a10 r10=fffffa6001a65380
r11=0000000000000005 r12=0000000000486ba7 r13=0000000000183c10
r14=0000000000000000 r15=fffff8000207e640
iopl=0 nv up ei pl nz na pe cy
cs=0010 ss=0018 ds=002b es=002b fs=0053 gs=002b efl=00010203
nt!ExQueryPoolUsage+0xe0:
fffff800`01e85120 8379e400 cmp dword ptr [rcx-1Ch],0 ds:002b:63fb8336`746efb67=????????
Resetting default scope
CUSTOMER_CRASH_COUNT: 2
DEFAULT_BUCKET_ID: VISTA_DRIVER_FAULT
BUGCHECK_STR: 0x3B
PROCESS_NAME: SearchIndexer.e
CURRENT_IRQL: 0
LAST_CONTROL_TRANSFER: from fffff800021207ba to fffff80001e85120
STACK_TEXT:
fffffa60`0958e950 fffff800`021207ba : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`0b96ebe0 : nt!ExQueryPoolUsage+0xe0
fffffa60`0958e960 fffff800`01eb0ef3 : fffffa80`0c048060 00000000`00fbe000 fffffa60`0958ebf8 00000000`00000000 : nt!NtQuerySystemInformation+0x66a
fffffa60`0958ec20 00000000`778e707a : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : nt!KiSystemServiceCopyEnd+0x13
00000000`0b96eb18 00000000`00000000 : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : 0x778e707a

FOLLOWUP_IP:
nt!ExQueryPoolUsage+e0
fffff800`01e85120 8379e400 cmp dword ptr [rcx-1Ch],0
SYMBOL_STACK_INDEX: 0
SYMBOL_NAME: nt!ExQueryPoolUsage+e0
FOLLOWUP_NAME: MachineOwner
MODULE_NAME: nt
IMAGE_NAME: ntkrnlmp.exe
DEBUG_FLR_IMAGE_TIMESTAMP: 4a7801eb
STACK_COMMAND: .cxr 0xfffffa600958e0f0 ; kb
FAILURE_BUCKET_ID: X64_0x3B_nt!ExQueryPoolUsage+e0
BUCKET_ID: X64_0x3B_nt!ExQueryPoolUsage+e0
Followup: MachineOwner
---------

 

[jmack]

Just a quick update. I've run scannnow and found no errors. I've also opened up the remining minidumps and the common theme seems to be corrumpt memory, so I'm going back and retesting my memory one stick at a time.

 

[usasma]

STOP 0x3B = SYSTEM_SERVICE_EXCEPTION
More info on it here: BSOD Index

This is most likely caused by an issue with a system service - although it can be caused by memory problems. In this case the first parameter (c0000005) means that it's an error in accessing memory - not necessarily a problem with the actual memory.

Please zip up the contents of the C:\Windows\Minidump folder and upload/attach the .zip file with your next post. Often it's the analysis of all the dump files (and a bit of massaging with different commands) that will reveal a pattern that's not otherwise obvious.

 

[jmack]

usamsa, thank you for your help. Unfortunately I'm running into a problem zipping my minidumps. I get a 'file not found or no read permission' error when using the send-to function or dragging and dropping the minidumps. As soon as I figure out how to get around this I will post them.

Continuing my saga, I've run memtest on each stick of ram and tested each stick of ram under a heavy load (Modern Warfare 2 at max settings on 1920 x 1080) without any problems. I've moved to testing two-stick combinations without problems so far either. I thought I read somewhere that Vista can get sticky if it was installed with more than 4 gb of ram. Could that be an avenue to pursue, since I installed Vista with 3 sticks of 2GB each?

 

[jmack]

Here are the minidumps! Thanks again.

 

[stoneys-nutz]

I thought I read somewhere that Vista can get sticky if it was installed with more than 4 gb of ram. Could that be an avenue to pursue, since I installed Vista with 3 sticks of 2GB each?

Hello jmack,
That error stops the installation, as you already have Vista installed, i doubt that is an issue.

Error message when you try to install Windows Vista on a computer that uses more than 3 GB of RAM: "STOP 0x0000000A"

Do you have the latest Bios ?.

 

[jmack]

Hi Stoneys,
Yes I have the latest BIOS and video drivers. I've also tested various legacy video drivers.

I've finished testing all my memory in various two-stick combinations without a BSOD. I'm wondering if the third DIMM slot on my motherboard is bad, or if there is a BIOS setting somewhere that causes problems when I have three sticks in (not sure if that is even possible ... just theorizing!).

 

[stoneys-nutz]

I had a problem where if i used three 2 Gig sticks, no problem, but if i used the other stick, four 2 Gig sticks, i would get a crash.
I manually set the memory to the manufactures settings, (timings and voltage), my recommended voltage was 2.1V, i set mine to 2.2V, then i increased the NB voltage by the same amount and it's been OK ever since.
I'm not saying this will fix yours, but must be worth a try.
Just keep an eye on the temps.

 

[jmack]

Thanks Stoneys,
I'm not ready to tinker with timings and voltages yet. I had a new development though - with just the 2 sticks in (4GB total) I had the pc randomly reboot and a black screen came up informing me that windows couldn't restart because ntoskrnl.exe was missing or corrupt and provided the following code: 0xc0000221.

Does this or the previous minidumps provide any further insight? I'm repairing vista now.

Cheers,
Jason

 

[jmack]

Vista restore got me back to the desktop, but I had a BSOD shortly therafter while downloading the latest Realtek HD Audio Driver. Attached the minidump. Interestingly, the physical dump to memory stopped at 70 (%?) and I had to shut the computer down.

I'm really not sure what to do here.

Latest BSOD = stop 0x0000003b; SYMEFA64.sys.

Best,
Jason

 

[usasma]

Latest BSOD points to Norton.

Uninstall it, then use the free Norton Removal Tool ( Download and run the Norton Removal Tool )
Then immediately install another antivirus (NOT an Internet Security) program to provide protection while testing.

 

[jmack]

Thanks, usasma. Do you have a suggestion for the replacement antivirus? Any idea what the conflict is with Norton?

Cheers,
Jason

 

[jmack]

I uninstalled Norton, used the removal tool, and replaced it with Avast. It promptly went into a BSOD loop that even safe mode and repairing wouldn't solve. I've done a clean install and performed all required updates.

I am just using Vista Firewall and Avast. Things seem stable with no BSODs yet, but I did come down to a black screen with a message about a checksum. I hit enter and it sent me to my BIOS. I've restored the defaults there and put everything to auto. I noticed when doing this that it reduces my memory speed to 1060 MHz instead of the 1333 MHz I had it set to previously (per manufacturers instructions).

I closed out of BIOS and the computer restarted.

 

[stoneys-nutz]

Can you confirm that you have the latest Video and Sound drivers, (not the ones from a DVD/CD), and that you have the latest Chipset drivers too.
You would be surprised how many people do that after a clean install lol.
It's a good idea to keep everything set to default, untill you are happy that this issue is fixed.
Thanks.

 

[jmack]

Hi Stoney,
Yes, I have the latest video drivers, sound drivers, mobo BIOS ... everything I can think of I have updated.

I've tested RAM modules with Memtest and at load, ad nauseum, including every posible individual and two-stick combination.

I have everything set at defaults and/or auto in the BIOS.

I continue to have BSODs of the following types:

Memory Management
Page Fualt IN Non-Paged Area
IRQ Not Less or Equal

The physical memory dumps with each BSOD will not finish resulting in corrupt dumps. Interestingly enough, my folders are only showing one main.dmp and one mini.dmp though I have had numerous BSODs since reinstalling.

I appreciate everyone's support from the forum. I'm sure it goes without saying how frustrating and disheartening this is. Any new ideas or old ideas I should revisit? I'm still suspcious of my hardware combination and I would really prefer not having to purchase replacement parts just to diagnose whether or not the part was bad.

Mobo - EVGA x58 SLI
Processor - Intel i7 920
GFX - EVGA Nvidia GTX 285
RAM - 3x Super Talent 2GB DDR3 PC 10600 (mobo automatically downclocks to 1066 Mhz 1.5V ... I have tested @ 1333 MHz and slightly higher voltages)

 

[jmack]

A quick addendum to my last post - I had a new BSOD (Bad Pool Caller) and the physical dump completed okay. Attached are the mindumps since my reinstall.

Cheers,
Jason

 

[stoneys-nutz]

The "Bad Pool Caller" C2 error is caused by, and i quote,
"A kernel-mode process or driver incorrectly attempted to perform memory operations. Typically, a faulty driver/hardware or buggy software causes this".

This would normally show itself after coming out of hibernation or sleep mode.
Just to eliminate this can you check your power settings, make sure they are set to High Performance, and that the "Turn Off Display", and "Put The Computer To Sleep" is set to never.
It would typically happen on a laptop, but i just want to make sure nothing is missed.

Unfortunately the mini dump file still has corruption in it. The first one was unreadable. The second pointed to "fileinfo.sys"

Can you try something for me, remove all USB devices, leave keyboard and mouse, and try the machine with just one stick of ram.

I know you will have done this, but just make sure the MB, Video Card, and HDD power leads are as they should be, and even re-seat the Video card.

If you get another BSOD, do you have another Video Card you can try, just to eliminate it.
Because this minidump was corrupt in places, i went back over the last good one and it was a 0X3B error.
This error has been linked to excessive paged pool usage and may occur due to user-mode graphics drivers crossing over and passing bad data to the kernel code

I never thought i would want a BSOD, but i would like a clean one anytime soon, lol.

Give this a try, and if we don't get the results we are looking for, then i would go in to the Bios and set the memory to the recommended timings and voltage, and the CPU to the recommended voltage. This is a tweakers MB, so we may need to tweak.
When this system is running properly, it will be a nice system.

Just so you know, i will only give up when you do.

Good luck and post your results, thanks.

 

[jmack]

Thanks for the confidence booster Stoneys, I needed it.

Remove all USB devices except mouse and keyboard ... check! (simple, since the only USB device is the mouse and keyboard

Use only one stick of RAM ... check!

Double check all power leads ... check!

Re-set video card ... check!

BSOD after 2 minutes of Modern Warfare 2 ... check!

Unfortunately the physical memory dump didn't even start this time. It just hung on the blue screen with the following information:

dxgkrnl.sys

FFFFFA60037D610A base @ FFFFFA6003718000 datestamp 4abc1C73

STOP 0x0000007E (0xFFFFFFFFC0000005 0xFFFFF60037D610A 0xFFFFFA6004381838 0xFFFFA6004381210)

I've switched out to an old ATI x1950. I will keep you posted!

Cheers,
Jason

 

[jmack]

Stoneys, just a quick question. If my memory specs require a minimum tRAS of 36 and my motherboard only offers settings between 1 and 31 (its auto detecting at 24) could this be the source of the problem?

On another note, I've tried my other video card and it seemed to work without a problem. Any way to test the card itself or do I just need to work with the vendor to get a replacement to test with?

 

[stoneys-nutz]

Is there another PC, yours or a friends, that you can try the Video card on, just to confirm it ?, the problem is, is this a conflicting hardware problem, or is the card itself faulty ?.

As for the memory, I'm not sure, looking at the figures, i would say its a possibility, is your memory on your MB makers QVL ?, (qualified vendors list).

If you are thinking of changing the memory, check the QVL, it will save a lot of time and heartache.

Its good to hear its stable, at least you are going in the right direction.