Really need help with a BSOD problem (0x4e pfn_list_corrupt) ntoskrnl.exe

Hello guys and gals of Major Geeks. I'm really in a desperate place right now as I have been unable to figure this out on my own now since I built the PC a few weeks ago and could really use some help.

First here are the specs:

Mobo - ASUS M5A99FX Pro R2.0
Cpu - AMD FX-8350 8 Core AM3+ 4.0GHZ
Cpu Cooler - Corsair Cooling Hydro Series H100I
Gpu - Gigabyte GeForce GTX 760
RAM - Patriot Viper 3 8GB 2X4GB PC3-12800 DDR3-1600 9-9-9-24
HD - WD Caviar Black 1TB SATA3 6GB/S
PSU - Thermaltake TR2 RX 650W
Tower - Cooler Master Haf 932
OS - Windows 7 Ultimate 64-bit

I've attached the dump files as a zip that I saved through BlueScreenView and zipped the dmp file found in the c:\\Windows\Minidump folder. I've also attached my readings I get with Hardware Monitor to check my voltages coming from my PSU and a few other things. Disregard the TMPIN2 reading 128C, I've looked in to that and apparently it's common with software like this to get a messed up readings sometimes. Speedfan displays the same temperature but at -128C. I've felt around the board and other hardware and there is nothing anywhere near those temps.

The PC boots up fine and I can use it and play games, watch movies or browse the net for a while and after a couple hours I get the blue screen with the stop code: 0x4e pfn_list_corrupt (I have showing it here http://i.imgur.com/Ox1aBLu.jpg)

I've ran memtest86 and my ram came out clean. I also ran the Western Digital LifeGuard Diagnostic tool on my HD and it is clean.

I don't overclock my systems so nothing in this new build has been overclocked. My temps are normal with the GPU running between 29-31C under no load, and the CPU at 31C under no load. The HD is at 34C.

I've tried reseating my ram and checked the voltages to make sure they were clocked properly (with CPU-Z) and they are where they should be @ 9-9-9-24 1.5v.

Any ideas and help are greatly appreciated, if you require any more information feel free to ask.

 

Here's what I could get out of the Windows Debugger Tool.

 

Doing a scan with MemTest (not memtest86) came up with an error on one of the RAM sticks, although it could be one of the mobo's DIMM slots as well so I'll do a bit more testing and look at getting the RAM replaced.

 

As soon as memtest hit 75% coverage it's been error city on the first stick of ram. 101.2% coverage and it has 1958 errors. Going to see about a replacement set of ram for tomorrow to test with.

 

Good plan :major

Post back here if you still get crashes after RAM replacement, include the zipped folder output from carefully following the instructions here.

And Welcome to Majorgeeks!

 

Hey, thanks for the welcome.

Just finished testing the 2nd stick of ram in the same dimm slot as the 1st.

The 2nd stick is clear of errors so far at 150% coverage and seems to be fine.

The 1st stick started getting errors around 75% and by time 100%+ came around it was already in the thousands.

Being tested in the same dimm slot rules the mobo's dimm slots being faulty (I hope), and new ram should solve things. Otherwise I'll definitely be following that link.

 

Left memtest running for about 5 hours while I was out for dinner tonight and came home to it at 352% coverage with 0 errors, so looking good there.

Now I'm going to try and see if I blue screen before I go to sleep.

 

So far so good. I was able to play some Battlefield 4 Beta and LoL (was the one most commonly causing me to BSOD) for a couple hours last night uninterrupted so it looks like it was just faulty ram.

I wonder why memtest86 failed to find any errors but memtest didn't? :confused Oh well, problem solved. :-D

 

I'd give it another week or so of normal usage before considering it to be fixed.

Memory compatibility and testing is pretty complex. The generally accepted 'foolproof' method is using Memtest86+ for 7+* runs on each stick/slot - even then with all sticks/slots testing as good, a fully populated motherboard might still have memory issues.

From the compatibility point of view, memory is best selected and bought after using the online configurators of Crucial or Kingston.




* (though very rarely, errors don't show until 11+ runs; usually, longer runs than this, 20-30+, tend to increase the chances that any errors seen are false positives.)