MEMORY_MANAGEMENT BSOD, probably hard drive related.

Creator13

Reputable
Jun 14, 2014
22
0
4,510
Hello,

I'm on Windows 10 and I've rrecently been getting MEMORY_MANAGEMENT blue screens. All the signs are pointing towards the hard drive as the culprit, but I want to be sure. Here are the facts:

I've been getting the MEMORY_MANAGEMENT blue screen for about two weeks now, maybe longer. I see no correlation between PC load and crashes, except that they don't occur when the computer is idle. I've run a memtest (1 pass, 9+ hours) with no errors. I've run sfc /scannow just today and it said it found corrupted files and repaired them. I'm not sure if this solved the issue. I have a backup of my drive. My drive is a 1TB Seagate Barracuda disk of about 2.5 years old. There's currently 25GB of free space. I've had problems with my Windows installation for as long as I have this PC, which is 2.5 years. I'm planning on installing a fresh Windows 10 installation when I upgrade, but I might do that now as well (and get a transferable retail key for when I need to upgrade). I'd also buy an SSD for those extra MBs per second (off topic). Sometimes after these blue screens, my PC runs a chkdsk /f after restarting. I ran it myself and it returned without problems. I'm currently running chkdsk /r, I'll update you with the results as soon as I can. Now for the most interesting part: the latest crash two days ago happened like this: I was watching YouTube videos and my pc gets slow. Soon after, I hear a sound come from my hard disk. It sounded like the disk blocked and braked real fast and then started spinning again. After that, my pc was still working, but everything that required disk access stopped working (websites no longer connected, software didn't switch views, etc). It got slower quickly and then crashed. This all happened within a minute.

Now, this does point very clearly in the directions of the hard drive. But can this problem be caused by some other component failing? I'm thinking PSU or motherboard sata controller, or is RAM still an option? Basically, will this problem be fixed if I get a new SSD and clean Windows 10? What is your opinion?

System specs:

  • ■ OS: Windows 10 (upgraded from 7)
    ■ Motherboard: Gigabyte 970A-UD3P
    ■ CPU: AMD FX-8320
    ■ GPU: MSi Radeon R9 270 GAMING
    ■ RAM: Crucial Ballistix 8GB 1600MHz 2x4 (BLS2CP4G3D1609DS1S00CEU)
    ■ HDD: Seagate Barracuda 1TB (SKU ST1000DM003)
 
Solution
If you want to totally eliminate the drive you could reinstall windows on it first and see if the problem returns. if so then yes it's 100% the hard drive otherwise you could still get a new SSD install windows on it and use the other drive as secondary. This would give you additional storage room and if it decides to continue having problems you'll most likely lose data but not necessarily encounter a blue screen.
Sounds like a dying hard drive to me as well, get close to the drive can you hear any clicking? Particularly when the drive is doing something? Personally once any drive I own starts to return errors I make backups and look into replacing it, especially drives I rely on.

I'm doubtful the sata controller would cause an issue like that, I doubt its ram if metest didn't show anything, and im pretty certain the psu wouldn't cause a memory management blue screen.
 
I personally think it's too soon to make the determination that it's a failing hard drive at this time. You've mentioned you have been receiving BSOD pointing to memory management. Have you tried viewing the crash dump and seeing what it says? It will mention the memory management but it might associated the culprit that actually caused the memory management error.
 


Yes, I do hear clicking, but that just sounds like the normal clicking hard drives do anyway.



Could you explain how I'd view the crash dump?
 


Alright, I ran whocrashed over my minidumps and they all say the same thing: ntoskrnl.exe is the problem. Further info:

Bug check description: This indicates that a severe memory management error occurred.
This might be a case of memory corruption. More often memory corruption happens because of software errors in buggy drivers, not because of faulty RAM modules. This problem might also be caused because of overheating (thermal issue).
The crash took place in the Windows kernel. Possibly this problem is caused by another driver that cannot be identified at this time.

My temperatures are nominal. Any ideas? I'm not very familiar with problems like this.
 
Ok so just like it says it's most likely that something caused the windows kernel to crash and it's most likely a hardware driver. Could you upload a few of the mini dumps so I can look at them? Also is your windows up to date on it's updates etc?
 


I know for a fact that my Windows 10 isn't updating to the anniversary update, although I'm still getting other updates. I have tried everything to start the update but without result. I have given that one up.

Link to minidump zip.

What do you think about the blocking/braking sound my harddisk made as I described in the top post? That was certainly related to the blue screen that followed directly after (this was the latest crash, so the most recent minidump file).
 
Ok I thought I'd post a little update here. I mentioned I was running chkdsk /r, right? Well, it got stuck at 37% for hours. (like, I believe it was around 35% when writing the original post, so it's been stuck for at least 4.5 hours). That's just now turned into another BSOD. This time: NTFS_FILE_SYSTEM. More HDD problems, I guess? Currently, it's rebooting. It went from Automatic system restore is being started to "Disk errors are being restored. This can take over an hour" (I'm translating this freely from Dutch, actual message may be different on an English OS). I'll update whenever I know more.
 
Did you do a normal check disk and find errors? Were you running the check disk while in windows? Run the check disk but this time have it perform the check after you restart your computer. This will allow the check disk to run without causing a blue screen.


I'll take a look at your dump files now.
 
Looking at a couple of your dumps I'm starting to believe just as it was previously mentioned the problem seems to be with your hard drive. You're going to want to run that chkdsk /f on boot up to prevent any blue screen errors. It looks like the problem is coming from wherever your pagefile is stored on your physical drive.
 
I'll give one answer for all of your posts. I did run a chkdsk /f before I ran /r, and it returned without any errors. I ran both /f and /r scheduled at restart, so I was surprised to see a blue screen as well. Sadly, the automatic repair seemed to have initiated chkdsk /r again, and it seems to be stuck again (at 37%). I did not manually shut down any process, I simply waited very long until it suddenly stopped itself.

EDIT: I also can't access the minidump from the latest blue screen, seen as we're stuck in what's probably going to be an infinite boot loop.
 
So you are quite certain my hard drive is failing? As I've made a backup of all my important data right before I started running tests and checks, I'm not worried about my data. Would it be possible to buy a new SSD and install a clean Windows on it, then completely reformat my current hard disk to use it as a secondary drive? Or is the disk probably physically damaged/unstable and no longer usable?
 
If you want to totally eliminate the drive you could reinstall windows on it first and see if the problem returns. if so then yes it's 100% the hard drive otherwise you could still get a new SSD install windows on it and use the other drive as secondary. This would give you additional storage room and if it decides to continue having problems you'll most likely lose data but not necessarily encounter a blue screen.
 
Solution
So another update: after a series of reboots, I managed to successfully boot into the computer again. So, disk not entirely broken. Would you still recommend to completely reinstall windows to repair the pagefile?
 
This file timntr.sys is 9 yrs old. It looks like it belongs to True Image. Either update it or uninstall it

Old drivers can crash windows. I think Itunes is also out of date

I would update this file xhcdrv.sys your USB 3 drivers. If you installed any. Theyre 2 yrs old. Disable AMD Overdrive, this is a known cause of crashes

Wouldnt be surprised if Avira is causing a crash





 
You can't repair the pagefile and it's not the pagefile itself that's the issue but instead the section of the hard drive that is being used for the pagefile could be at fault. As for as re-installing windows... that's up to you and how far you want to go to insure that the problem has been resolved. Here are your options:

1. Leave everything as is and hope it doesn't blue screen again.
2. Re-install windows to see if the problem happens again if it does proceed to step 3.
3. Purchase another hard drive and install windows on it so you can eliminate the older hard drive being used in step 1.
 


So yeah, drivers... I can't really disable OverDrive, as this is an integrated feature of the GPU driver. My USB 3 drivers have always been a cause of problems, as they sometimes fail to work. I reinstall them every so many months, but the problems always return. Other programs might indeed be the cause of the crash, but True Image never runs, Avira is just annoying but I never bothered to switch, iTunes also never runs.



I really have no problems with reinstalling Windows; as I've mentioned in my original post it's terribly buggy and laggy and old and won't update anymore. I also don't mind having an SSD over a hard disk (in fact, I just ordered one). I was mostly concerned about possible other components being at fault, and if that wasn't the case about whether I could still use the drive as secondary storage. Since both of those questions have now been answered, I think I'll choose option 3 and take over the wheel from here again.
 
Just letting you know that I'm now running Windows 10 on my new SSD (I've never seen a computer so fast in my life!) and it works perfectly. I've formatted my old disk and I'm running a chkdsk /r on it to see how badly it's damaged, or if it is at all. Thanks for the help everyone!