BSOD couple times a month, usually (but not always) 0x0000101, 0000019

[arbeleg]

I ran the system for a couple of days with Driver Verifier on, and is BSOD's this evening, again with a 0x101 code. From the sounds of it, I've got bad hardware in there somewhere.

 

[x BlueRobot]

Could please upload the Kernel Memory dump like before?

 

[arbeleg]

Thanks again. Uploaded to http://sdrv.ms/12U8Ph6

 

[x BlueRobot]

Since, you've had a Stop 0x101 with Driver Verifier, then almost suggests that there is a hardware issue here.

Boot into the BIOS, and then post the voltages for the 3.3V, 5V and 12V rails if possible.

• BIOS Access Keys for Popular Motherboards (ASUS, MSI, GIGABYTE, Etc.)

Run the Small FFT's, Large FFT's and Blend Test separately for a few hours each, a report should be created within the Prime95 installation folder. Do not let the temperature exceed 80*C.

• http://www.sevenforums.com/tutorials/100352-hardware-stress-test-prime95.html

Enable "Round off checking" before the test (see first post below tutorial).

 

[arbeleg]

BIOS reports voltages:

12V: 12.302
5V: 5.184
3.3V: 3.376

I've started running the Prime95 tests. Should I disable Driver Verifier going forward, or leave it enabled during the stress testing?

 

[x BlueRobot]

Your voltages look okay, and disable Driver Verifier during the testing.

 

[arbeleg]

Thanks. I ran small FFT, large FFT and blend tests for multiple hours each with no errors or warnings reported. Also no BSODs. However, Driver Verifier was still enabled, so I will disable and repeat the tests.

 

[x BlueRobot]

There will be no need to

Have you tried this generic troubleshooting guide for Stop 0x101 - http://www.sevenforums.com/crash-lo...0x101-clock_watchdog_timeout-troubleshtg.html

I'll reply this evening (7pm onwards - local time) because I've got work.

 

[arbeleg]

I have tried Prime95 and Furmark burn in without reproducing the BSOD. Temperature readings have been <60C during all the tests, so I don't think it's an overheating problem. BIOS is fully updated. SFC/scannow came back clean.

Are we looking at either CPU or motherboard hardware issue at this point? Is there any way to tell which component is problematic, given that I don't have another CPU or mobo to swap in?

 

[x BlueRobot]

Did you get any errors from the Prime95 test; go to the installation folder of Prime95 and there should be a log file saved. Attach the log file in your next post.

 

[arbeleg]

No errors that I noticed. Attached is my results.txt from Prime95.

 

[x BlueRobot]

• http://www.sevenforums.com/tutorials/105647-ram-test-memtest86.html

Run Memtest86+ for least 7-8 passes, and preferably overnight as it can take a while to fully complete.

Test each RAM stick individually, if an error is found then move the same RAM stick into the next DIMM slot and test again, if errors are found for the same RAM stick in every available slot then you have a faulty RAM module. On the other hand, if no errors are found in the next slot or the other slots for the same RAM module, then you have a faulty DIMM slot.

Test each RAM stick and every motherboard DIMM slot available.

test|Slot1|Slot2
RAM1| Error | Error
RAM2|Good|Good

It is a RAM, a bad RAM.

But if you have got a result like that:

test|Slot1|Slot2
RAM1| Error |Good
RAM2| Error |Good

It is a motherboard issue. The particular slot is bad.

Sorry for the really slow reply.

 

[arbeleg]

Well it took me a long time to carve out some time to run Memtest86+, but I finally did so this weekend.

I let it run for 30 hours, 7+ passes worth, exercising all 16 GB across all 4 slots, and all CPU cores. No errors or warnings of any kind. CPU temps never got above 50 C. Since I couldn't reproduce a crash with all the memory installed, it didn't seem like testing each DIMM/slot combination separately would be productive.

Yet I'm still getting BSOD's sporadically over the past few months. Almost always with the "A clock interrupt was not received on a secondary processor" error. Still no pattern that I can see that relates to when BSODs occur. Sometimes the system is idle, sometimes actively typing/browsing but not heavily loaded. Sometimes will go a couple of weeks without a crash, other times it happens on back to back days. The system is attached to a good UPS (line-interactive, sine wave, voltage regulating) so I don't think it's a problem with dirty power.

Still at a loss for what the problem might be, and how to detect it. Any other advice or things to try?