Frequent Memory Related BSODs, but Memcheck claims my RAM is fine

Toggle sidebar Toggle sidebar
S

Shaikoten

Honorable
Jan 8, 2014
14
0
10,510
Having left my computer on over my Christmas vacation, I came back to realize random browser plugins had begun crashing. No matter, I thought, these things happen. And then as I started using it, the BSODs started happening, something my previously healthy, well maintained PC hadn't had much of. I'm mostly getting MEMORY_MANAGEMENT, PAGED FILE IN NON PAGED AREA, win32ksys, and other seemingly memory related posts.

So I did the logical thing and checked my memory for problems. None detected by memcheck. I checked my hard drives for problems too, just in case. No worries there either. So I began reinstalling drivers. I did GPU drivers and chipset, to no avail. Disabled everything overclocked, but crashes still persisted. Checked heat, nothing out of the ordinary, and crashes didn't even happen while under load, or even while the computer was in active use.

Finally, after another random crash, my windows install refused to boot, so I tried methods to repair it: system restore, system repair, nothing seemed to help. So, having all my things backed up onto other drives already, I reinstalled windows. It BSOD'd during the WINDOWS INSTALL. And now, the next day after a fresh install, it's still BSODing.

I've tried the RAM in different configurations too. Using individual sticks, in different slots, but none of it seems to be improving stability. Annoyingly I can't find one thing that induces this crash to recreate it, so I'm at the mercy of whether it crashes or not to tell if it's fixed every time I try something. The last thing I have yet to try is changing voltages for CPU or RAM, as I have heard this could fix it.

My Specs:
Windows 7 Pro
i5 3570k
Biostar TZ77XE3
Geil Evo Veloce 2x8GB DDR3 14900
XFX HD6950 2GB (Using triple monitors... maybe an issue?)
USB Devices: Logitech MX518, Filco Majestouch 2, Topping TP30 Sound Card/Amplifier

Hopefully we can solve things together! I can PM my most recent minidump if someone wants to take a gander.
 
Sort by date Sort by votes
It sounds like a hard drive/windows system/registry issue to be honest. The original error, "MEMORY_MANAGEMENT, PAGED FILE IN NON PAGED AREA" indicates the RAM's paging file on the hard drive was the source of the error. A pagefile is a designated area on your hard drive that RAM segregates for it's use and is generally used when you reach RAM saturation.

Your symptoms sound nearly identical to what happened to me last month. Would no longer boot and I couldn't repair the drive using the usual tools. Having everything already backed up on another drive, I just deleted both partition tables on it and ran the quick format during the re-installation of Win 7. No problems since then.
 
Upvote 0 Downvote


I would think any corruption would have righted itself after reinstalling windows however. And in my mind this still doesn't explain the BSODs during the windows install either. The paging file is definitely a possibility though; I was under the impression it was a bad idea to keep a paging file on one's SSD, so instead I switched it over to a 2TB Toshiba drive, simply using the default managed profile.

I've just begun a scan using Video Memory stress test as some searching indicated GPU memory could be the culprit. I almost hope this is the case because I've been looking for an excuse to update my card 😉
 
Upvote 0 Downvote
When I could no longer boot Windows... I feared my overclock had finally done some damage somewhere(I suspected hardware as you do now) especially when I could not recover the image using Recovery Console/System Restore/Safe Mode. Pretty sure at this point I had either a bad registry or unrecoverable boot sector.

GPU memory is mapped to your system memory via MMIO. I think it usually gets physically mapped to a DIMM but maybe MMIO can address the pagefile or they are inherently linked because of MMIO. Either way, I suppose it is worth a shot to change the GPU and see if you can reproduce the symptom.
 
Upvote 0 Downvote
Do you have the latest BIOS? and mobo drivers? Check and make sure those are up to date, then boot to BIOS and enable XMP...if problems persist chack and make sure that the base timings are to the DRAM spec, andif not correct manually, can also raise DRAM voltage + 0.06 from spec
 
Upvote 0 Downvote


The timings are a good point, and I don't think they've ever been to spec. I just never messed with them because for the 10~ months I've had this it hasn't been a problem. One redditor suggested it might be my power draw. 3 monitors, 3 hard drives, on a 600w PSU... and everything plugged into the same power source. Probably not a smart move, he figured there might be a ground loop issue. That's definitely something I'm gonna try.

As for the GPU, the stress test ran through just fine while I was at work. No problems there that I can see.

EDIT: Also, given that I have no idea what's presently causing the BSOD, and it happens at seemingly random times, is there a way to induce the same error I had before through BlueScreenView or any other utility to check to see if a fix worked?
 
Upvote 0 Downvote


Bios and mobo drivers are up to date, I did a sweep of all new ones when I did the windows reinstall.

My attempted fixes last night included setting the memory to factory settings, and then running chkdisk. While the voltage was correct, I had to switch it from 16000 to 18666 MHz and fix the timings. The disk scan also turned up a corrupt sector on my SSD (boot drive) although I'm unsure if this would hang over from a format or was a post windows install thing. No BSODs overnight but I'm going to give it at least 24 hours before I call it a success; I've still had a couple of issues like certain chrome tabs occasionally crashing so I'm not convinced it's totally gone.

johnbl: As for the dump file, I posted the raw on pastebin if you'd like a gander.
 
Upvote 0 Downvote
I would need the actual memory.dmp file so I can connect a debugger to it.
it would be about 250mb in size. you can put it on a cloud share like skydrive and make it public
(looked like you posted the first 1200 lines of the binary as text)



 
Upvote 0 Downvote


I just posted the minidump's output from Blue Screen Viewer. The full dump file is about 500mb. Uploading presently.
 
Upvote 0 Downvote
Here's the updated file, hopefully the sharing works now, if not PM me your email and I can invite you or some such.

The ram and hard drive fixes did not stop the BSOD, and you can see by the file size differences that this most recent dump happened not too long ago. So, still need a fix for sure.
 
Upvote 0 Downvote
Machine: Manufacturer
BIOSTAR Group
Product TZ77XE3
Version 5.0
BIOS Version 4.6.5
BIOS Starting Address Segment f000
BIOS Release Date 04/19/2012

Intel(R) Core(TM) i5-3570K CPU @ 3.40GHz



Windows 7 Kernel Version 7600 MP (4 procs) Free x64
Built by: 7600.16385.amd64fre.win7_rtm.090713-1255
System Uptime: 0 days 2:13:27.378
BugCheck 50, {ffffd8a00de5d130, 0, fffff880101490e1, 7}

PAGE_FAULT_IN_NONPAGED_AREA (50)

the bugcheck was in the ATI video driver being called from directx

You are using a current version of the ATI video driver from 2013 on a very old version of the OS that
has a directx version from 2009. You will want to update your OS with the service packs and updates and retest.
you may be able to install the new directx from microsoft
http://www.microsoft.com/en-us/download/details.aspx?id=35

I would recommend that you get all the OS updates just to eliminate other problems.

Note: debugger showed 16 corruptions in nt kernel (but it looked like a string in a resource was zeroed out)

 
Upvote 0 Downvote


Certainly possible, though when I originally got these crashes I was running a 100% updated version of windows. I'll give it a try and post any new versions of the dump file. The GPU specific ones aren't actually terribly common as opposed to the memory ones.

Also: In an attempt to rule out high load, I did unplug some of my hard drives to reduce draw, most notably my oldest one which I suspected might be the culprit. This didn't seem to help anything.

I'll keep you updated with any crashes in the next 24 hours.
 
Upvote 0 Downvote
Crashed again before I can update, though it was a different crash. This was upon opening a game. Still running windows update but hopefully this dump can be helpful.

https://drive.google.com/file/d/0B9TpJwkGkKFBMm02cFQ0M2RXVDA/edit?usp=sharing

EDIT: A very similar crash less than an hour apart from this. It was upon opening my IRC client, which is located on a different hard drive, though some of the user data might be on the other drive.
 
Upvote 0 Downvote
Well, I have been tirelessly trying to fix this, but still no end to the crashes. I swapped out my GPU and that didn't end up helping matters. Ran with all but one drive plugged in, that didn't help either. I have decided to get a new PSU, hoping this was the issue. I've run all the windows updates I could, though it froze and crashed while attempting to do most of them. Having continuously reinstalled the drivers, and windows, I really think I can safely say it's not a software issue. But to say it's the PSU is just a stab in the dark. I have no idea if it is or isn't. Worst case scenario, I have two PSUs, which means I can set up my alternate computer while I continue troubleshooting this one.
 
Upvote 0 Downvote
-crash dump is not made public access
you will want to boot and run memtest86 to confirm your memory timings are correct and eliminate windows drivers as the cause of the error.



 
Upvote 0 Downvote


Should be public now. And I've done memtest86 as well, all is coming up clean on that front. As well as readjust the timings to the factory recommended, they were originally different but had been running a bit lower for nearly a year with no problems.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom