BSODs - Pagefaultinnonpagedarea - Memory Management, while browsing

[Troopy]

Hi,

I'm having problems with pretty much any game randomly quiting to desktop (mainly XCOM as thats what I'm playing atm) recently seems to have gotten worse then shortly afterwards a couple of times bluscreens as mentioned above, (Page fault in nonpaged area or Memory Management), the BSODs occured while browsing using firefox, iirc.

After reading this forum for other peoples similar faults I have done the tests below:-

Run SFC /SCANNOW Command - System File Checker
Run RAM - Test with Memtest86+

Both come back clear, I ran memtest for about 28 hours it completed 8 or 9 passes, no problems found....
I know, I know, you still think its memory!!!
Any suggestions where to go from here - put each memstick in individually and rerun memtest?
(it originally had 4x4GB and i removed 1 stick for suspect bad ram, but it was some time back and i cant remember the exact circumstances now)

I've also ran Furmark to overheat the GFX, I dont think theres a problem there, (especially as it runs XCOM without breaking a sweat,) while Furmark was running i can hear the fan kick up a gear and I let the temp go up to about 100degC, no problems.

While the games run they run without problems, until they suddenly quit to desktop.

Any help would be appreciated, I have attached SF dump

Thanks

 

[essenbe]

Troopy, the problem I am having is your files list several dump files (which is mostly what we use to debug) but, they are all from 12/2013-1/2014 none of them are recent. Yes, all of the error messages you say you have do sound like memory. But, if you ram memtest86+ for 8 Passes and passed, I see no reason to run it again, at this point.

If you are overclocking anything, set everything back do default level, at least until you get this resolved. If you don't already have it, please download and install CPUz (CPU-Z CPUID - System & hardware benchmark, monitoring, reporting ) Do you have a CPU and a GPU temp monitoring program? Also, do you have a GPU utility that will allow you to see an on screen display while gaming and also allow you to set a custom fan profile? Never let your GPU get to 100C again. Also, please download GPUz (GPU-Z Video card GPU Information Utility) and post screenshots of the GPUz main page and the sensors tab also post screenshots of CPUz CPU, Mainboard, Memory and SPD tabs. Also are all of your ram modules the same, did they come as a kit?

Please make sure your computer is configured for small memory dumps. This will tell you how http://www.sevenforums.com/tutorials/174459-dump-files-configure-windows-create-bsod.html

Please get back to me with the answers to the questions I asked. Also, when downloading and installing those programs, be sure to watch out for the 'add on' programs they like to install with the program. You can uncheck the boxes and not install them. Also, be aware that you mentioned running 12 GB of ram. In most boards that will make you running in single channel mode. In single channel mode you may see better performance in single threaded applications, but dual channel mode is generally better. That requires you to run the exact same ram modules in all 4 slots or in 2 slots of the same color. Your motherboard manual will tell you which two is better.

 

[Troopy]

Hi Essenbe,

I appreciate you taking the time to try to help me.

I have included a zip with GPUZ and CPUZ files as requested - CPUZ is the HTML report but that should have all the information in there.

I actually found a GPU-Z sensor log of when i was using Furmark to push the temp up and it was getting to 95 before i was bottling it not 100. (On GPU Temp. #2 which was about 5°C hotter than the other two).
VRM Temperature 1 seems to be approx double of VRM Temperature 2 [°C]. But i dont think its temperature based as it ran Furmark without issue.

Xcom will sometimes crash while running in background as I'm on desktop.

I have altered setting to create small dump as requested

I'm pretty sure the Ram is all the same, looking at the CPU-Z info.

I'm going to try and play now see what happens.

Again thanks for help much appreciated.

 

[essenbe]

Troopy, let us know how it goes. Maybe you will get us some current dump files at least. I would like you to post screenshots like this, please http://www.sevenforums.com/tutorials/9733-screenshots-files-upload-post-seven-forums.html Do you have a CPU and GPU monitoring program? If you do, please use the GPU program to set a custom fan profile to keep your graphics card cooler. Please monitor your CPU and GPU temps.

 

[Troopy]

Hi,

Well had a couple of really long sessions with no problems (2-3 hours even on Saturday night), then tonight it did a hard crash (straight to reboot - may have been a BSOD, but i only had centre screen powered on and not sure if it will display on there).

This was literally 5 minutes at most after booting PC, Xcom hadnt got past loading screen.

attached is the dump, and the CPU-Z screens, all SPDs are the same, the memory is all 1 kit looking at it both physically and the serial numbers etc.

Any help appreciated.
Thanks

 

[essenbe]

Troopy, I know you said you ran memtest, but this is what I get from your dump file and little else

FOLLOWUP_IP: 
nt!MiMapProcessMetaPage+cc
fffff800`031074ac ea              ???
SYMBOL_STACK_INDEX:  0
SYMBOL_NAME:  nt!MiMapProcessMetaPage+cc
FOLLOWUP_NAME:  MachineOwner
MODULE_NAME: nt
DEBUG_FLR_IMAGE_TIMESTAMP:  531590fb
IMAGE_VERSION:  6.1.7601.18409
STACK_COMMAND:  .cxr 0xfffff88003175b20 ; kb
IMAGE_NAME: [COLOR=red]memory_corruption[/COLOR]
FAILURE_BUCKET_ID:  X64_0x24_nt!MiMapProcessMetaPage+cc
BUCKET_ID:  X64_0x24_nt!MiMapProcessMetaPage+cc
ANALYSIS_SOURCE:  KM
FAILURE_ID_HASH_STRING:  km:x64_0x24_nt!mimapprocessmetapage+cc

Did you reduce the page file on your computer to save space on the SSD? Does your Kingston have a utility that can check it for errors? Have you run that manufacturer's diagnostic test on your Mechanical drive? Where are your games installed? I assume on the Mechanical drive. I'm asking these questions, and probably more, because the error code usually has to do with an NTFS file system error or it can also be memory corruption or the fact that you ran out of memory. As you can see the debugger is calling it memory corruption.

Here's what to try, run the hard drive manufacturer's diagnostic test for your mechanical drive. Run the short drive self test and the long drive test. I would also like you to run checkdisk with the /f/r switches http://www.sevenforums.com/tutorials/433-disk-check.html for example, if your mechanical dive is E drive in option 2 run chkdsk E:/F /R It will most likely tell you it can't be run while it is in use, then will ask if you would like to schedule one on your next reboot, select yes and reboot. DO NOT STOP IT ONCE IT HAS STARTED. It will take usually quite a while. I am reluctant to run check disk on the SSD as I have been told it can damage it. However, I own 8 SSDs and have run it on most of mine with no problem. If you will tell me the manufacturer of your hard drive, I will try to get you a link to the diagnostic tool, although you can usually use Seagate's Seatools. It is always better to run the test for DOS as you boot into the media you use and the drive is not in use. Try that and see if we get any results.

 

[Troopy]

thanks for speedy reply,

No i havent altered page file, the SSD has about 30+GB free, only runs OS and a few other bit and pieces all games stored on 1TB mechanical disc.

I will run those tests tomorrow its late here and i work early shifts.

PS. since that 1st crash I rebooted, made earlier post then played XCOM again for an hour or more no problems.

 

[essenbe]

Ok, no problem Sometimes, these things are hard to figure out, and take a lot of time and testing.

 

[Troopy]

Not had much time to do anything with this but left chkdsk running on the D: drive (1tb mech), looks ok?


Checking file system on D: The type of the file system is NTFS. Volume label is Games2. A disk check has been scheduled. Windows will now check the disk.

CHKDSK is verifying files (stage 1 of 5)... 584192 file records processed. File verification completed. 219 large file records processed. 0 bad file records processed. 0 EA records processed. 1 reparse records processed.

CHKDSK is verifying indexes (stage 2 of 5)... 648306 index entries processed. Index verification completed. 0 unindexed files scanned. 0 unindexed files recovered.

CHKDSK is verifying security descriptors (stage 3 of 5)... 584192 file SDs/SIDs processed. Cleaning up 121 unused index entries from index $SII of file 0x9. Cleaning up 121 unused index entries from index $SDH of file 0x9. Cleaning up 121 unused security descriptors. Security descriptor verification completed. 32058 data files processed. CHKDSK is verifying Usn Journal... 1149120 USN bytes processed. Usn Journal verification completed.

CHKDSK is verifying file data (stage 4 of 5)... 584176 files processed. File data verification completed.

CHKDSK is verifying free space (stage 5 of 5)... 42674977 free clusters processed. Free space verification is complete.

Windows has checked the file system and found no problems.

976759807 KB total disk space. 805146120 KB in 549926 files. 232336 KB in 32059 indexes. 0 KB in bad sectors. 681439 KB in use by the system. 65536 KB occupied by the log file. 170699912 KB available on disk. 4096 bytes in each allocation unit. 244189951 total allocation units on disk. 42674978 allocation units available on disk.

Internal Info: 00 ea 08 00 6d e1 08 00 a3 35 10 00 00 00 00 00 ....m....5...... 87 00 00 00 01 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................

 

[essenbe]

Are you monitoring the CPU and GPU temps? The temp you reported during Furmark were was too high. Of course, that's what all stress tests do, it push hardware to the limits and drastically raise temps. But, if you have a good GPU monitoring program you can keep watch in the temps and set a custom fan profile to keep temps down. High temps can cause BSODs. The computer will shut down to protect itself from damage.

 

[Troopy]

Yeah I've monitored gpu temp and I think it seems fine, the GPU load was listed as 40%, the temps were 78[°C] 82[°C] and 77[°C], the fan was running at 2200rpm ( I'm pretty sure the fan ramps up as it needs to as i can hear it under stress testing, also the log file says it does)

Also I CHKDSK C: /f /r
nothing too exciting

I'll set something up to monitor CPU temp

 

[Troopy]

oops reposted yesterdays comment

 

[essenbe]

The GPU temps you mentioned are too high. Were you running Furmark when You reached those temps? If your GPU monitoring program has the option, do a custom fan profile and increase your fan speed .

 

[Troopy]

Ok thanks for the info,

No i wasnt running furmark - just Xcom

So another quit to desktop today ( i thought i had turned the fan up, using AMD Cat control centre, but hadnt enabled overdrive.) - No BSOD

temps not insanely high - around same as posted previously
Turned fan to 100%, you could probably hear it from where you are
played for another 30 mins or so no problems, temps sat at about 60[°C]
Unfortunately ran out of time so will have to test again

 

[essenbe]

Troopy, see if you can clean any dust out of the fans on the card with canned compressed air and I don't recall, but I don't think CCC will let you set a custom fan profile. I would try MSI Afterburner. It is primarily an overclocking utility, but it will also let you set a custom fan profile and set an on screen display where you can see your temps, as well as many other things while in game. I would set an aggressive fan profile with it to keep the temps down. I don't know what case you have, but if you can maybe add another fan or 2 to the case to bring in cooker air, and if possible I would have one blowing directly on the card. The temps should not be that high, although I don't think it will damage the card or cause the crashes, but the card should be cooler.