Solved Random BSOD after system maintenance

[badkarma11]

Well, the annoying saga continues. Was OK for about a week and then started blowing up like a Libyan village. I upped the DRAM voltage to 1.65 to no avail. New dump attached and I'm going to pull the Crucial memory when I get home tonight.

 

[Dave76]

Let me know how it's running with the Crucial RAM removed.

Last crashes:

STOP 0x0000001A: MEMORY_MANAGEMENT
Usual causes:  Device driver, memory, kernel
PROCESS_NAME:  bf3.exe
FAILURE_BUCKET_ID:  X64_0x1a_5001_nt!_??_::FNODOBFM::_string_+46620
-----------------------------------------------------------------------------------------------------------------
STOP 0x00000019: BAD_POOL_HEADER
Usual causes:  Device driver
PROCESS_NAME:  svchost.exe
IMAGE_NAME:  Pool_Corruption
FAILURE_BUCKET_ID:  X64_0x19_3_nt!ExDeferredFreePool+a53
-----------------------------------------------------------------------------------------------------------------
STOP 0x0000001A: MEMORY_MANAGEMENT
Usual causes:  Device driver, memory, kernel
PROCESS_NAME:  System
FAILURE_BUCKET_ID:  X64_0x1a_41284_nt!_??_::FNODOBFM::_string_+4ac3
-----------------------------------------------------------------------------------------------------------------
STOP 0x0000004E: PFN_LIST_CORRUPT
Usual causes:  Device driver, ?memory
Probably caused by : memory_corruption
PROCESS_NAME:  System
IMAGE_NAME:  memory_corruption
FAILURE_BUCKET_ID:  X64_0x4E_2_nt!MiUnlinkPageFromLockedList+8d
-----------------------------------------------------------------------------------------------------------------
STOP 0x0000003B: SYSTEM_SERVICE_EXCEPTION
Usual causes:  System service, Device driver, graphics driver, ?memory
Probably caused by : dxgmms1.sys 
EXCEPTION_CODE: (NTSTATUS) 0xc0000005 - The instruction at 0x%08lx referenced memory at 0x%08lx. The memory could not be %s.
PROCESS_NAME:  bf3.exe
IMAGE_NAME:  dxgmms1.sys
FAILURE_BUCKET_ID:  X64_0x3B_dxgmms1!VIDMM_DEVICE::RemoveCommitment+8f
-----------------------------------------------------------------------------------------------------------------
STOP 0x0000010E: VIDEO_MEMORY_MANAGEMENT_INTERNAL
Usual causes:  Video driver, ?Video memory
This indicates that the video memory manager has encountered a condition that it is unable to recover from.
Probably caused by : dxgmms1.sys
PROCESS_NAME:  System
IMAGE_NAME:  dxgmms1.sys
FAILURE_BUCKET_ID:  X64_0x10e_1f_dxgmms1!VIDMM_GLOBAL::ReferenceAllocationForSubmission+148
------------------------------------------------------------------------------------------------------------------
STOP 0x000000C2: BAD_POOL_CALLER
Usual causes:  Device driver, ?Memory
FAULTING_IP: dxgkrnl!DXGDEVICE::DestroyAllocations+72d
PROCESS_NAME:  iexplore.exe
IMAGE_NAME:  dxgkrnl.sys
FAILURE_BUCKET_ID:  X64_0xc2_d_dxgkrnl!DXGDEVICE::DestroyAllocations+72d
-------------------------------------------------------------------------------------------------------------------
STOP 0x0000001A: MEMORY_MANAGEMENT
Usual causes:  Device driver, memory, kernel
PROCESS_NAME:  MsMpEng.exe
IMAGE_NAME:  ntkrnlmp.exe
FAILURE_BUCKET_ID:  X64_0x1a_3452_nt!_??_::FNODOBFM::_string_+356c3

They are mostly pointing to memory issues.
Stop code 4E: This bugcheck usually occurs for one of two reasons, the first reason being memory corruption, the second is found with Driver Verifier.

A couple are blaming Video driver or memory.
Run OCCT GPU test.
OCCT stability checking tool
Run the GPU:3D test for at least 30 minutes, be sure to monitor your CPU and GPU temps.
Set the time before you start.
When its done, pass or fail, it will make some graphs. Post these here as they are useful for analyses.

From your Sytem Event Logs:

Event[23605]:
  Log Name: System
  Source: Disk
  Date: 2012-09-24T18:37:49.629
  Event ID: 11
  Task: N/A
  Level: Error
  Opcode: N/A
  Keyword: Classic
  User: N/A
  User Name: N/A
  Computer: GhengisSSD
  Description: 
The driver detected a controller error on \Device\Harddisk4\DR4.

If you start getting a lot of these you will need to run the hard drive tests again.

Event[26594]:
  Log Name: System
  Source: Schannel
  Date: 2012-09-28T14:42:13.333
  Event ID: 36874
  Task: N/A
  Level: Error
  Opcode: Info
  Keyword: N/A
  User: S-1-5-18
  User Name: NT AUTHORITY\SYSTEM
  Computer: GhengisSSD
  Description: 
An SSL 3.0 connection request was received from a remote client application, but none of the cipher suites supported by the client application are supported by the server. The SSL connection request has failed.

Event[26595]:
  Log Name: System
  Source: Schannel
  Date: 2012-09-28T14:42:13.333
  Event ID: 36888
  Task: N/A
  Level: Error
  Opcode: Info
  Keyword: N/A
  User: S-1-5-18
  User Name: NT AUTHORITY\SYSTEM
  Computer: GhengisSSD
  Description: 
The following fatal alert was generated: 40. The internal error state is 107.

Do you know what this is from?
Download and run Malwarebytes, be sure to update it before running.
Malwarebytes : Download free malware, virus and spyware tools

More HDD errors
2012-09-30 - The driver detected a controller error on \Device\Harddisk4\DR4
2012-09-30 - The driver detected a controller error on \Device\Harddisk4\DR5

 

[badkarma11]

Thanks for the reply. I will do as you ask. the SSL message was probably from my attempt to use a remote tool from work to install the 64 bit Cisco VPN client so you can ignore it. I want to do this in steps so I'm going to do the following in the following order:
1. set the DRAM voltage back to 1.60
2. remove the Crucial RAM and run it to see if more BSODs
3. I have downloaded the OCCT tool but not installed that yet. I will run it if I get a BSOD after the RAM removal

Question: In the first part of your post, you mentioned driver verifier. Should I spin that up after I remove the RAM, or wait for a kaboom?
HDD controller errors: How do I tell which drive is Harddisk4? I've got 2 HDDs and an SSD plus 2 DVD burners and a stick that is always plugged in. I'm going to pull the USB stick except when I actually use it (backing up a couple of files) Edit: It appears that DR is 'Drive, Removable' and these errors could be attributed to USB sticks. I've been using some (as many do) to move files around. If I don't see any of those errors when I don't use the sticks, I could probably ignore them when they reappear. Probably not causing the crashes I think.

Will report back with results or lack thereof.

Additional Edit:
I came home, booted up 'puter, hit 'delete' to enter setup and pulled the USB stick at the same time. BIOS screen locks up... I power off, restart, go into BIOS, set DRAM voltage to 1.60. Boot computer. Played some BF3 (usually crashes, often with BSOD). No problems. Installed OCCT, ran GPU 3D test for 30 min fullscreen, error check. I ended it prematurely (by accident). CPU core temps were core 0, 75c, the others in the mid 60s. I ran it again, windowed, for 30 minutes. What graphs would you like to see? I'm going to fire up driver verifier and see if it dies instantly like it did last time. Will report. Still haven't removed the Crucial, one variable at a time.

 

[Dave76]

Good steps.

You can try Driver Verifier again later if needed, for now it's not high on the list.
Pull the Crucial RAM, this will be a very useful test.
If it's found to be stable afterwards, then we can focus on RAM settings w/Crucial RAM installed, see if we can get it stable.

The hard drive numbering is usually, not always, the same as the motherboard manual numbering.
Best way to check is to disconnect the 2 DVD burners and see if that helps, if not disconnect the HDDs one at a time and see if that helps.
Your SSD should be connected to SATA port 0, make sure it is on the Intel port.

Remove all USB devices if you aren't using them, only plug in a USB memory stick when needed, then unplug when done.
USB devices can cause crashes, have seen many be the cause of BSoDs.

Post all of the OCCT graphs.

 

[badkarma11]

Keeping the USB stick out. System was stable last night and I did everything that usually blows it up. OCCT graphs attached. If it turns out to be the USB stick, wow. I had no idea that they could cause problems. Will pull RAM on next BSOD. Will check which port SSD is on and move if necessary.

Edit: Right after I posted this, I got a BSOD. I pulled the Crucial RAM and moved the SSD to drive slot 0. Somehow I hosed Windows completely so I've reinstalled and am in the process of recovering my system. I will report if any problems. So far, so good, at least it didn't blow up installing Windows

 

[Dave76]

Is the 'USB Stick' a memory stick?

Did you do a Clean Install?

Which port was the SSD previously plugged into?

Remaining Ram should be in the Black slots.

The OCCT voltage graphs don't look very good, too much fluctuation.
This puts the PSU in question.

Run the OCCT Power Supply test.
Run the Power Supply test for at least 30 minutes, be sure to monitor your CPU and GPU temps.
Set the time before you start.
Check the box to 'Use All Logical Cores'.
When its done, pass or fail, it will make some graphs. Post these here as they are useful for analyses.

 

[badkarma11]

Yes, a memory stick. It's been plugged in for the life of the machine (not plugged in now).
Port 2 (they are numbered 1-6 on the mobo, drives are enumerated 0-5 in the BIOS, SSD is drive 0)
OK, I will post the graphs in 1/2 hour. Also, when I first booted this AM, the 'puter blew up all over the place. Eventually windows repair resurrected it.

 

[badkarma11]

I ran OCCT power supply test and Core 0 reached 85 degrees and OCCT bailed on the test after around 1 minute. I went to the BIOS and selected 'Optimized settings' because I don't remember if I changed anything that perhaps I shouldn't have. The only setting I changed back was ACHI. Graphs attached. I think I might pull the CPU and re heat paste it.
EDIT: OCCT wouldn't let me select 'Use all Logical Cores'
EDIT2: Repasted the heat sink. OCCT when not 'testing' shows core temps hovering around 40 degrees. Started the PSU test and same results, 85+ temps after about a minute

 

[Dave76]

What CPU cooler are you using?
Those temps are high, pull the CPU cooler and check the thermal paste, too much or too little is bad.
Did you clean with Isopropyl alcohol?
Should not have paste pushed out the side of the CPU.
Is the CPU cooler fan working like it's suppose to?

OCCT graphs time period is too short to be useful.
When you get the temps under control, run the OCCT Power Supply test again.

 

[badkarma11]

Stock heatsink/fan. The fan appears to be running properly. At idle, the core temps are now hovering in the mid 30s (when I repasted the heat sink, I hadn't locked it back down properly). They only spike on the Power supply test. Am I running the test incorrrectly? I tried again with the PSU test and it quit after < 1 minute.

EDIT: I have pulled and repasted the CPU cooler again (properly,not too much paste, cleaned with iso alky). I set the BIOS to spin the CPU fan to max if it gets above 65, minimum speed is 75 % of max. Regardless, if I run the OCCT PSU test, the temps spike above 85 in 45 seconds. Absent that test, any thoughts? Since it blew up all over the place AFTER I pulled the Crucial RAM, I put it back in tonight (it's been out since Saturday and nothing really got better).
Ugh....

 

[Dave76]

The core temps should be between 60-70°C max.
Idle temps are much better .
This could be a CPU cooler fan problem, did you clean the cooling fins?


Take the PNY RAM out and run the CPU:OCCT test for 30 minutes, stop the test if the CPU goes over 80°C, be sure to monitor your CPU and GPU temps closely.
Set the time before you start.
When its done, pass or fail, it will make some graphs. Post these here again.

You might want to consider a new CPU cooler.
I have used this one, it works very well and the price is good.
COOLER MASTER Hyper 212 Plus

Check the height of your RAM cards, a big heatsink on the RAM cards may contact the Hyper 212.
I had very little clearance with standard Corsair XMS cards.

 

[badkarma11]

The fan seems OK, I cleaned the fins a while back, still clean. THe PNY cards have heat sinks, the Crucial do not so I'll probably have to swap them if I get that cooler. I'll run the 30 minute test with what I've got and post results.

EDIT: Ran the CPU test, temps spiked after about a minute and a half and killed the test. I think I'll buy that cooler you recommended. Mine is pretty weak, comparitively speaking. When I get and install it, I'll report back. Thanks for hanging in there.

 

[Dave76]

You're welcome.
We'll get to the end of it soon, I hope.

Swapping the RAM cards should work, the Hyper 212 Plus cooler fan will only be over the first RAM slot from the CPU.

Let me know how the new cooler is performing.

 

[badkarma11]

ordered the new cooler finally. All last week (since I repasted my existing CPU cooler) the machine behaved flawlessly. No app crashes, no BSODs no matter what I was doing. This morning it went back to hell. I had returned the NB and DRAM voltages to stock settings after I repasted and have not touched them since. dump attached, for your amusement.

 

[Dave76]

Change the DRAM voltage to 1.600v and the NB voltage to 1.220v.

Which RAM cards do you have installed now?
Can you let me know which RAM is installed on each reply?
You need to check if the Crucial or PNY is faulty, so run them separately, if you get a crash swap them.
One of the RAM cards may be faulty, the two sets may not run together stable, you may need to adjust the settings to get them to run together.
So test with either Crucial or PNY for now.

These crashes are from RAM issues, your last description confirms that is the most likely cause.
Last three crashes:

STOP 0x00000050: PAGE_FAULT_IN_NONPAGED_AREA 
Usual causes:  Defective hardware (particularly memory - but not just RAM), Faulty system service, Antivirus,  Device driver, NTFS corruption, BIOS 
GetUlongFromAddress: unable to read from fffff800030fa1c0 
FAILURE_BUCKET_ID:  X64_0x50_nt!CmpKcbCacheLookup+1de
-----------------------------------------------------------------------------------------------------------------------------------------------------
STOP 0x00000019: BAD_POOL_HEADER 
Usual causes:  Device driver
Probably caused by : Pool_Corruption ( nt!ExDeferredFreePool+118b )
PROCESS_NAME:  WerFault.exe
IMAGE_NAME:  Pool_Corruption
FAILURE_BUCKET_ID:  X64_0x19_3_nt!ExDeferredFreePool+118b
-----------------------------------------------------------------------------------------------------------------------------------------------------
STOP 0x0000004E: PFN_LIST_CORRUPT
Usual causes:  Device driver, ?memory
Probably caused by : memory_corruption ( nt!MiBadShareCount+4c )
IMAGE_NAME:  memory_corruption
FAILURE_BUCKET_ID:  X64_0x4E_99_nt!MiBadShareCount+4c

 

[badkarma11]

I currently have the PNY in slots 1 and 3, and the crucial in the other 2. Before I repasted the cpu heat sink last week, I had run on only the PNY in slots one and three for the previous week. I had gotten crashes and you told me to run the OCCT tests. The system 'heat spikes' on either the CPU or power supply tests after 30 -90 seconds and OCCT ends the tests. Yesterday, after the system steadied down, I ran OCCT and then played some games and watched the CPU temps. They were topping out at 75-80 degrees under load (not good). There is obviously a heat issue, besides whatever else might be wrong with the RAM. I ordered the new cooler and when I crack the case to install that I will pull the PNY and run on the crucial. That way I can hopefully run the OCCT power supply and CPU tests and see if it's stable with the crucial alone. I will reset the DRAM and NB voltages to 1.60 and 1.20 respectively. I will report back when I have the new cooler installed and the PNY RAM removed. Thanks for your help and I apologize for not being clear in my last post.

 

[Dave76]

Thanks for the update.

It helps a lot if you can keep me advised to any changes you make on your system.

 

[badkarma11]

Yay cooler came! Boo, have to utterly disassemble the machine to install it. more to follow..

 

[Dave76]

Good news, let us know how the install goes.

 

[badkarma11]

I will. Probably won't get to it until the weekend but I'll update when it's in. I'll run the OCCT CPU and PSU tests and post the graphs after it's in and the PNY memory is out.