Solved Random BSOD after system maintenance

[badkarma11]

I have installed the new cooler. I have run the OCCT PSU and CPU tests for 30 minutes each. The CPU temps never reached 60 degrees (C) during the tests. When I installed the new CPU cooler, I removed the mobo from the case. While doing so, I noticed that 3 of the 7 screws that attached the mobo to the case were loose (no doubt the result of my tearing things apart and reseating them). There was also a loose power connection to my sound card. My current situation is this: All RAM is installed, all questionable connections have been tightened and I wish to start from a new 'scratch' from a diagnostic point of view. The OCCT output is attached. I will post again if BSODs occur and we can go from there. I prefer to change only one variable at a time, in this case I have changed two (heat issue SOLVED and potentially error causing grounding issues addressed to the best of my ability. We'll see what happens, I welcome whatever you divine from the OCCT results.
Cheers
Alex

 

[Dave76]

CPU core temps are nice, much better.

Loose power connection, grounding, can be the cause. Good idea to check the motherboard.
All the OCCT graphs look good, still seeing the voltage fluctuation they should be nearly flat line, this can be caused by PSU or motherboard.
It's not bad enough to start replacing components, just keep it in mind.

Let's wait and see if you get another crash.

 

[badkarma11]

*crosses fingers*

 

[badkarma11]

More boom. Ugh

 

[Dave76]

That's not good news.

Are the DRAM voltage 1.600v and the NB voltage 1.220v?

Run memtest86+ again, 7 full passes, all RAM cards installed.
You can stop the test if you get any errors.

When you changed the CPU cooler, did you inspect the CPU for bent pins?

Last four crashes:

STOP 0x0000001A: MEMORY_MANAGEMENT
Usual causes:  Device driver, memory, kernel
PROCESS_NAME:  MsMpEng.exe
FAILURE_BUCKET_ID:  X64_0x1a_403_nt!_??_::FNODOBFM::_string_+320ec
----------------------------------------------------------------------------------------------------------------------------------------------
STOP 0x000000C2: BAD_POOL_CALLER
Usual causes:  Device driver, ?Memory
Probably caused by : dxgkrnl.sys ( dxgkrnl!DXGDEVICE::DestroyAllocations+72d )
PROCESS_NAME:  bf3.exe
MAGE_NAME:  dxgkrnl.sys
FAILURE_BUCKET_ID:  X64_0xc2_d_dxgkrnl!DXGDEVICE::DestroyAllocations+72d
-----------------------------------------------------------------------------------------------------------------------------------------------
STOP 0x0000004E: PFN_LIST_CORRUPT
Usual causes:  Device driver, ?memory
Probably caused by : memory_corruption ( nt!MiBadShareCount+4c )
PROCESS_NAME:  explorer.exe
IMAGE_NAME:  memory_corruption
-----------------------------------------------------------------------------------------------------------------------------------------------
STOP 0x0000000A: IRQL_NOT_LESS_OR_EQUAL
Usual causes:  Kernel mode driver, System Service, BIOS, Windows, Virus scanner, Backup tool, compatibility
FAILURE_BUCKET_ID:  X64_0xA_nt!KeAcquireInStackQueuedSpinLock+5f

First one doesn't give much info.
Second is blaming DirectX.
Diagnosing basic problems with DirectX
Third is memory corruption.
Fourth is not much help.

 

[badkarma11]

Yeah, it's not. The voltages are set as described. The 1st one was when I was simultaneously closing Microsoft Money 2004 and removing a USB stick after backing Money up to it. Bad timing. the 2nd was exiting Battlefield 3 which is a known bug producing nation. The third was on shutdown after playing BF3 and I forget what was going on with the 4th. I'll run memtest overnight, but it hasn't thrown an error, ever in the 10 or so times I ran it. Last time was 11 passes (earlier in this thread, as I recall). Might be the mobo got cooked somewhere at some point, dunno. Thanks for checking in, though.

didn't look carefully at the CPU, but it's the contact pin as opposed to the socket pin type connector.

 

[Dave76]

Have to agree, you've testing everything we can run tests on.
Which leaves PSU and motherboard.
The OCCT graphs are showing a lot of Vcore (CPU voltage) fluctuation which may be caused buy either of these.
You can test the PSU by installing a different or new one. Borrow one, if you can, or buy one at a place with a good return policy, which is hard to find.

I have seen memtest86+ pass several times and then throw an error, new RAM fixing the issue.

Wish we could have narrowed it down further.

 

[badkarma11]

I've beat memtest to death. I could buy a new PSU and apply the current one to a future build if it turns out not to be the problem. Socket 775 mobos are all but unavailable, except for the one I have. If it looks like mobo, I'll live with it until my next build. At least the BSODs are intermittant. Thank you so very much for all the time you took on this. I'll PM you if/when I swap out the PSU and let you know what happened, but I guess this thread must now go to it's reward.

 

[Dave76]

Good plan, you can post the results of the new PSU here.
That way it might help someone else with a similar issue.

 

[badkarma11]

I have finally replaced the PSU. I ran OCCT briefly to watch the real time graphing. low voltage stuff was flat line, not jaggy like last time. If it blows up, I'll run and post the PSU and CPU tests. If it doesn't blue screen for a month, I win!!!

 

[Dave76]

Good news.
What PSU did you get?

Thanks for the update and let us know if it's stable.
This thread will help others with similar issues.

 

[badkarma11]

Cooler Master RS550 eXtreme power plus (550w) Stable so far...

 

[Dave76]

Nice PSU.

Let us know if you have any issues.

 

[badkarma11]

No problems so far, I ran the PSU test in OCCT just for giggles. Dave, perhaps you can tell me if it looks any better.

 

[Dave76]

The voltage graphs are only slightly better that the previous ones, it's still possible the new PSU will help stability.
The bad news is the cause of the voltage fluctuations may be the motherboard.

Let us know the system status.

 

[badkarma11]

I'll keep you posted. So far, so good, no BSOD or odd behavior since install. I'll post if/when I get BSODs. thanks for looking at the graphs.

 

[Dave76]

You're welcome.

Let us know your progress.

 

[badkarma11]

All was well from last Wednesday through this morning, 2x bsod. Dump attached. I guess it's the motherboard. Both were BAD_POOL_HEADER while using IE, not doing anything taxing, quite soon <5 minutes after start up

 

[Dave76]

STOP 0x00000019: BAD_POOL_HEADER 
Usual causes:  Device driver, memory, external device drivers
Probably caused by : Pool_Corruption ( nt!ExDeferredFreePool+cbb )
BUGCHECK_STR:  0x19_3
PROCESS_NAME:  svchost.exe
PROCESS_NAME(dump #2): MsMpEng.exe
LOCK_ADDRESS:  fffff800032e1b80 -- (!locks fffff800032e1b80)
WARNING: SystemResourcesList->Flink chain invalid. Resource may be corrupted, or already deleted.
WARNING: SystemResourcesList->Blink chain invalid. Resource may be corrupted, or already deleted.
PNP_TRIAGE: 
	Lock address  : 0xfffff800032e1b80
	Thread Count  : 0
	Thread address: 0x0000000000000000
	Thread wait   : 0x0
FAILURE_BUCKET_ID:  X64_0x19_3_nt!ExDeferredFreePool+cbb

Examples for Svchost.exe services are: "Automatic Updates", "Windows Firewall", "Plug and Play", "Fax Service", "Windows Themes" and many more.
MsMpEng.exe is a core process of Windows Defender.
Memory corruption is a likely cause.
This can also be caused by a third party driver.
As you mentioned, this type of crash can be caused by the motherboard.

Have you installed any new software or hardware recently?

Run memtest86+ again, all RAM cards installed and for 7 full passes, best to run overnight.

Let us know the results.

 

[badkarma11]

no new anything, I've memtested in every conceivable configuration. Must be mobo, probably fried something in the high heat environment it used to live in or it's just been marginal since day one. If I ever replace the mobo (won't be any time soon) I'll post back. Until then, the saga continues. I'll mark it solved and say the ultimate cause was mobo since we've done everything else. Thanks for looking at it, Dave.