BSOD BCCode 116 Intermittent issue

I built my system about a year ago. Early this year I started experiencing BSOD's. Most were BCCode 116, so I followed the recommendations in the "STOP 0x116: VIDEO_TDR_ERROR troubleshooting" thread but couldn't find anything wrong. After the BSOD's started to get more frequent, I decided to reinstall Windows and see if that fixed the problem. At the same time as the reinstall I updated the firmware for my SSD and Motherboard as well as flipped the crossfire configuration of my graphics cards.

Things ran ok for a while but unfortunately the BSOD's are back. I've now had 4 BSOD's, all with BCCode 116 since the reinstall. After the 3rd BSOD, I've done the following:

Ran Memtest86+ with 7 passes - no errors
Ran OCCT GPU:3D with each graphics card individually and in crossfire configuration - no errors
Ran OCCT Power Supply with each graphics card individually and in crossfire configuration - no errors
Ran IntelBurnInTest (I ran prime95 prior to reinstall) - 10 run, 64 bit for both Standard and Maximum - no errors
Ran Furmark - Max temp on cards reached 90degrees and no errors reported
Ran System File Checker - no errors

My setup (OS and hardware are roughly a year old):
Windows 7 Professional x64
Intel Core i7 26000k
Gigabyte P67A-UD5-B3
2 x Corsair CMZ8GX3M2A 1600C8 4GB
HiS AMD 6950 Crossfire
Corsair AX850 PSU

Attachment 208267

Additional info:
- The only game I've been playing is Total War Shogun 2 (all settings at highest)
- None of the BSOD's have happened while I've been gaming.
- The BSOD's happen randomly and have occured after watching Youtube videos, while browsing, while idle (I run the system 24/7 and had a BSOD while I was sleeping).
- The only BSOD that is directly gaming related is the 3rd one in the minidump files. I was trying to run the Direct X benchmark included with Total War S2 when it happened.
- The problem appears intermittent. Looking at the last 4 BSOD's, they seem to occur at roughly a week apart.

Any assistance in helping me to figure out what the problem is will be much appreciated.

D

Just in passing These are all Bcc116 as you surmised, one thing of note is they all happen after long periods of uptime. The average being 6+ days,. This in itself can cause the problem Try re-booting daily to test.

zigzag3143 said:

Just in passing These are all Bcc116 as you surmised, one thing of note is they all happen after long periods of uptime. The average being 6+ days,. This in itself can cause the problem Try re-booting daily to test.

Thanks for the reply. Just so I'm clear, you want me to reboot every day to see if I get a BSOD after 6 days or so, correct? Will do so and provide an update. Just out of curiosity, why would the longer uptime be a problem?

Dramius said:

zigzag3143 said:

Just in passing These are all Bcc116 as you surmised, one thing of note is they all happen after long periods of uptime. The average being 6+ days,. This in itself can cause the problem Try re-booting daily to test.

Thanks for the reply. Just so I'm clear, you want me to reboot every day to see if I get a BSOD after 6 days or so, correct? Will do so and provide an update. Just out of curiosity, why would the longer uptime be a problem?

It doesnt necessarily have to be every day but just while testing. If there is the slightest of memory leaks in any driver software it will get worse over time.

zigzag3143 said:

Dramius said:

zigzag3143 said:

Just in passing These are all Bcc116 as you surmised, one thing of note is they all happen after long periods of uptime. The average being 6+ days,. This in itself can cause the problem Try re-booting daily to test.

Thanks for the reply. Just so I'm clear, you want me to reboot every day to see if I get a BSOD after 6 days or so, correct? Will do so and provide an update. Just out of curiosity, why would the longer uptime be a problem?

It doesnt necessarily have to be every day but just while testing. If there is the slightest of memory leaks in any driver software it will get worse over time.

Thanks. Hadn't thought about that. Am rebooting as often as I can and will let you know if I get a BSOD after a week or so. One thing to note, I got one of the Action Center messages that indicated I had outstanding troubleshooting and when I looked at them, I could see the 4 instances of BSOD's as well as other line items related to them (ie showing up with same date and time stamp) but they had BCCode 117. For example, the last BSOD I had has 4 line items with the BCCode 117. Others had less line items. Not sure if these showed up in the minidump. Unfortunately, I clicked on something and I now don't have the error messages in the Action Center and can't get them back again.

BCC116 and BCC117 are similar and caused by the same issues.

Looks like that wasn't the problem. Was rebooting the machine every few days and didn't get any BSOD's. Last night I shut the machine down and this morning after rebooting, I started having problems with IE. It was crashing a lot. Checked Event viewer and I could see a lot of errors. I had microtorrent running at that time and it died as well. Ran sfc /scannow twice but the scans wouldn't complete. Checked the SR logs and I could see three distinct areas that couldn't be repaired, one of them had to do with mdmlasat.inf, can't remember the other two. Tried to reboot in safe mode and that basically caused a complete crash. Tried to get into startup repair but that failed as well. chckdsk was giving me a system vol corrupt error. If I tried to boot normally, I was ending up in a reboot, crash/BSOD cycle. Took out my original Windows 7 disc and tried to use that to do a repair but also got BSOD with a PAGE_FAULT_IN_NONPAGED_AREA message with the BSOD. Not sure if this means the hard drive had a problem but I'm going to do a fresh windows install on a new SSD and see how that goes. Hopefully I can recover some data from the HD that's crapped out.

Now it looks like the HD wasn't the problem. Got a new SSD and tried to install Windows on it but am getting an error. I get past the language select screen and when I click on begin install, I get an install windows error code 0xc0000005. It could be my optical drive, so I'm going to try to find an extra optical drive and see if I can get windows to install from that but I've got a sneaky suspicion that it's the motherboard that's got a problem. Problem is how do you check a motherboard for problems? Could a faulty motherboard cause the HD to get corrupted?

Turns out a stick of memory is dead. Ran memtest just for the sake of it and one of the memory sticks was giving lots of errors. Which is wierd as I didn't have any errors previously. And after removing the problem memory, I managed to resotre my HD. So, considering the system was stressed tested like crazy with no memory problems and now suddenly I have one stick that's faililng, would it be the memory or the mother board that's the issue?

Memory errors do not always show up in Memtest86+ on a first run. I have seen instances where it had to be run for 21 passes just to get one error. Windows stresses the memory in different/more ways than Memtest86+ does and can sometimes be a better test for Memory modules just by swapping the modules out and testing individual modules and slots by running Windows normally. It may be your problem was the memory module all along and it was not detected initially.

How is the system currently responding with the bad stick removed?