Question Silent random BSOD occurring in the background

[JohnnyGui]

Hello,

Out of nowhere, I noticed two hardware errors by now at different times in the Reliability History without noticing anything weird around that time when using my PC; there were no freezes, black screens, BSOD's or reboots. The hardware errors seem to happen during a PC boot. Nothing is overclocked, it's a fresh install of Windows 10 Pro x64 October Update. PC is 10 months old and it's working fine.

Here's the log
https://www.dropbox.com/s/ea24etzuz9scvtb/Aantekening 2019-05-21 134138.png?raw=1

Here's a part of the dump file:
BugCheck AB, {3, 200, 0, 5}
Probably caused by : win32kfull.sys ( win32kfull!CheckSessionPoolAllocations+8caae )
0: kd> !analyze -v

SESSION_HAS_VALID_POOL_ON_EXIT (ab)
Caused by a session driver not freeing its pool allocations prior to a
session unload. This indicates a bug in win32k.sys, atmfd.dll,
rdpdd.dll or a video driver.
Arguments:
Arg1: 0000000000000003, session ID
Arg2: 0000000000000200, number of paged pool bytes that are leaking
Arg3: 0000000000000000, number of nonpaged pool bytes that are leaking
Arg4: 0000000000000005, total number of paged and nonpaged allocations that are leaking.
nonpaged allocations are in the upper half of this word,
paged allocations are in the lower half of this word.

Here's the full dump file: https://1drv.ms/u/s!AsuDIHiSqDG5hop_zgXSRyiFfOQZog

Specs:
Gigabyte GTX 1070 Windforce OC
Z370 HD3 Motherboard
CPU i5-8600K
16GB RAM

CHKDSK and sfc /scannow don't show any problems. Someone else at Tenforums.com has a similar bug report. I'm very curious if anyone else here has these hardware errors in the Reliability History.

 

[Colif]

that isn't a normal BSOD to get, I would try running memtest86 on each of your ram sticks, one stick at a time, up to 4 passes. Only error count you want is 0, any higher could be cause of the BSOD. Remove/replace ram sticks with errors.

its my 1st reaction to an unusual error, I get someone to look at dump files.

where did you get NVIDIA drivers from? I would run DDU and depending where you got GPU drivers from, I would get opposite. So if you got them from NVIDIA website, try running windows update after ddu - https://forums.tomshardware.com/faq...n-install-of-your-video-card-drivers.2402269/

 

[JohnnyGui]

that isn't a normal BSOD to get, I would try running memtest86 on each of your ram sticks, one stick at a time, up to 4 passes. Only error count you want is 0, any higher could be cause of the BSOD. Remove/replace ram sticks with errors.

its my 1st reaction to an unusual error, I get someone to look at dump files.

where did you get NVIDIA drivers from? I would run DDU and depending where you got GPU drivers from, I would get opposite. So if you got them from NVIDIA website, try running windows update after ddu - https://forums.tomshardware.com/faq...n-install-of-your-video-card-drivers.2402269/

Thanks for the reply. I got my Nvidia drivers from Nvidia's website. If it's a RAM corruption, doesn't it seem too coincidental that someone else at tenforums.com with the same Windows build number 17763-437 has the exact same error according to the dump file as mine?

 

[Colif]

The new NVIDIA drivers aren't what I would call stable, I would run ddu and then run windows update afterwards and let windows find drivers for card. They still NVIDIA drivers, just a little older and more reliable.

I went to NVIDIA because of this:

This indicates a bug in win32k.sys, atmfd.dll, rdpdd.dll or a video driver.

atmfd.dll = adobe type manager font driver
rdpdd.dll = RDP Display Driver (RDP = Remote Desktop)

While it could be windows, its more likely NVIDIA.

I can't read the dumps but have asked someone who has to convert them for me, he likely asleep now... or should be.

 

[JohnnyGui]

The new NVIDIA drivers aren't what I would call stable, I would run ddu and then run windows update afterwards and let windows find drivers for card. They still NVIDIA drivers, just a little older and more reliable.

I went to NVIDIA because of this:


atmfd.dll = adobe type manager font driver
rdpdd.dll = RDP Display Driver (RDP = Remote Desktop)

While it could be windows, its more likely NVIDIA.

I can't read the dumps but have asked someone who has to convert them for me, he likely asleep now... or should be.

The weird thing is that the user with the exact same error is using integrated Intel graphics, so I'm not sure if this is really at Nvidia's end. Furthermore, the latest Nvidia drivers are apparently not recommended and also causes glitches in games that i'm playing right now.

Thanks for asking someone to further clarify this, I'll wait his answer. Please let me know.

 

[JohnnyGui]

UPDATE:

Found 2 users so far having the same problem:
https://answers.microsoft.com/en-us...n/2f1d52b0-6205-4b51-af20-bc1f9876bf8c?page=4
https://www.tenforums.com/bsod-cras...ilent-random-bsod-occurring-background-2.html

 

[onespeedbiker]

LOL, one of the other users is you. Anyway, your MB has onboard graphics, try removing your GPU card and use the onboard graphics to see if is the NVIDIA drivers or not. Further, check GPU bios is up to date, the current is F2, 4/16/2019 https://www.gigabyte.com/Graphics-Card/GV-N1070WF2OC-8GD-rev-10#support-dl-bios

 

[gardenman]

Hi, I ran the dump file through the debugger and got the following information: https://pste.eu/p/Bsva.html

File information:	win32k.sys-20190512-2130.dmp (May 12 2019 - 15:30:08)
Bugcheck:	SESSION_HAS_VALID_POOL_ON_EXIT (AB)
Probably caused by:	memory_corruption (Process: csrss.exe)
Uptime:	2 Day(s), 19 Hour(s), 54 Min(s), and 07 Sec(s)

Comment: The dump file seemed to have some corruption in it.

Possible Motherboard page: https://www.gigabyte.com/us/Motherboard/Z370-HD3-rev-10
There is a BIOS update available for your system (version F12). You are using version F11. Wait for additional information before deciding to update or not. Important: Verify that I have linked to the correct motherboard. Updating your BIOS can be risky. Never try it when you might lose power (lightning storms, recent power outages, etc).

This information can be used by others to help you. I can't help you with this. Someone else will post with more information. Please wait for additional answers. Good luck.

 

[JohnnyGui]

LOL, one of the other users is you. Anyway, your MB has onboard graphics, try removing your GPU card and use the onboard graphics to see if is the NVIDIA drivers or not. Further, check GPU bios is up to date, the current is F2, 4/16/2019 https://www.gigabyte.com/Graphics-Card/GV-N1070WF2OC-8GD-rev-10#support-dl-bios

Please look more carefully, in one link I started a thread at tenforums to which someone replied having the same problem (I linked to his specific post). In the other thread someone else started a thread to which I replied having the same problem.

As for the GPU BIOS update, this is for the first revision of the GPU while I have the 2nd revision, its support page doesn't contain a section for BIOS updates for some reason.

Hi, I ran the dump file through the debugger and got the following information: https://pste.eu/p/Bsva.html

File information:	win32k.sys-20190512-2130.dmp (May 12 2019 - 15:30:08)
Bugcheck:	SESSION_HAS_VALID_POOL_ON_EXIT (AB)
Probably caused by:	memory_corruption (Process: csrss.exe)
Uptime:	2 Day(s), 19 Hour(s), 54 Min(s), and 07 Sec(s)

Comment: The dump file seemed to have some corruption in it.

Possible Motherboard page: https://www.gigabyte.com/us/Motherboard/Z370-HD3-rev-10
There is a BIOS update available for your system (version F12). You are using version F11. Wait for additional information before deciding to update or not. Important: Verify that I have linked to the correct motherboard. Updating your BIOS can be risky. Never try it when you might lose power (lightning storms, recent power outages, etc).

This information can be used by others to help you. I can't help you with this. Someone else will post with more information. Please wait for additional answers. Good luck.

Thanks a lot for the info. I have 2 questions:

1. The debugging states memory_corruption (csrss.exe). Does this mean my RAM could be corrupted?
2. The latest BIOS update (F12) only adds support for other products that I don't have. Can a newer BIOS fix this problem even though the update doesn't address this issue?

 

[onespeedbiker]

How about trying to use your onboard graphics?

 

[Colif]

Did you check ram?

so tenforums got you to install the newest nvidia drivers? wonder if older ones would be better.

csrss.exe = client services for windows. Its what your user runs in. probably not helpful info in regard to problem, but windows cannot run without it. Its a critical component

1. almost all BSOD dump files now mention memory corruption. I don't have an adequate explanation as to why but it doesn't specifically point at anything,
2. I wouldn't do the bios but you should update Intel Management Engine Interface (its under chipset drivers on motherboard website here) as the version you have is older than the actual bios you have installed and could be part of problem - as they meant to match each other.

So guy debugging your problem on tenforums discovers he has same thing. If anyone can figure it out, they should. Nothing wrong with asking in multiple places except when we offer conflicting advice. It might not be the ram, they might be right.

I try not to look in reliability history, it can get you worried about problems that aren't problems, Its like looking in event viewer, which always has errors but most of them can be ignored. I have had 2 red X in there in recent days but I have also had 3 windows updates in that time as they preparing way for 1903 to be released. I don't want to know what cause of the 2 X are as windows kept working, I only worry about it if my PC suddenly starts playing up. For 2 months my GPU would randomly turn off, I got sick of looking in reliability history and now feel the same way, I don't want to know.

 

[JohnnyGui]

Did you check ram?

so tenforums got you to install the newest nvidia drivers? wonder if older ones would be better.

csrss.exe = client services for windows. Its what your user runs in. probably not helpful info in regard to problem, but windows cannot run without it. Its a critical component

1. almost all BSOD dump files now mention memory corruption. I don't have an adequate explanation as to why but it doesn't specifically point at anything,
2. I wouldn't do the bios but you should update Intel Management Engine Interface (its under chipset drivers on motherboard website here) as the version you have is older than the actual bios you have installed and could be part of problem - as they meant to match each other.

So guy debugging your problem on tenforums discovers he has same thing. If anyone can figure it out, they should. Nothing wrong with asking in multiple places except when we offer conflicting advice. It might not be the ram, they might be right.

I try not to look in reliability history, it can get you worried about problems that aren't problems, Its like looking in event viewer, which always has errors but most of them can be ignored. I have had 2 red X in there in recent days but I have also had 3 windows updates in that time as they preparing way for 1903 to be released. I don't want to know what cause of the 2 X are as windows kept working, I only worry about it if my PC suddenly starts playing up. For 2 months my GPU would randomly turn off, I got sick of looking in reliability history and now feel the same way, I don't want to know.

Thanks a lot for your reply. I already have that latest version of Intel Management Engine Interface but for some reason the info file I attached as well as Device Manager shows an older version, while the Installed Apps & Software dialog shows I installed the latest one (perhaps the setup file consists of several versions from which it picked the one suitable for my PC)

So far, I have found 4 users reporting the exact same problem as mine quite recently

1. https://www.tenforums.com/bsod-cras...-bsod-occurring-background-2.html#post1633555
2. https://answers.microsoft.com/en-us...d=4a63dc91-2000-48fc-b8c8-bedb8fab5bd6&page=1
3. https://www.tenforums.com/bsod-cras...reen-message-after-shut-down.html#post1630048
4. https://www.dell.com/community/Inspiron-Desktops/Blue-Screen-after-shut-down/td-p/7300662

I completely understand what you mean regarding looking in the Reliability History. The thing is, I'm not worried when it comes to regular errors in there, but when I start noticing hardware errors that has happened more than once, then I can't help but think that something is wrong with my PC.

 

[Colif]

not all hardware errors in there point at a hardware problem. It will show Hardware error if PC restarts unexpectedly, which is what happens after a BSOD as well.

the GPU problem I had last year reported itself as a Hardware error

but it was pointing me at the wrong hardware. I believe it was my mouse as there was 2 changes I made the week it stopped, I removed Bitdefender and I bought a new mouse. Bitdefender is back on PC now and hasn't caused any hiccups so have to blame mouse. That doesn't help you

I can't check my reliability history anymore as I updated to Win 10 1903 this week and it wiped it. Its currently at 100%

it wasn't always so good

 

[JohnnyGui]

not all hardware errors in there point at a hardware problem. It will show Hardware error if PC restarts unexpectedly, which is what happens after a BSOD as well.

the GPU problem I had last year reported itself as a Hardware error

but it was pointing me at the wrong hardware. I believe it was my mouse as there was 2 changes I made the week it stopped, I removed Bitdefender and I bought a new mouse. Bitdefender is back on PC now and hasn't caused any hiccups so have to blame mouse. That doesn't help you

I can't check my reliability history anymore as I updated to Win 10 1903 this week and it wiped it. Its currently at 100%

it wasn't always so good

So after 3 weeks another silent BSOD occurred. I'm curious what the cause is this time. Here's a link to the dump file:
https://1drv.ms/u/s!AsuDIHiSqDG5hpBlSk7cIkm3cC3IjA

Could you please check for me?

 

[gardenman]

I ran the dump file through the debugger and got the following information: https://pste.eu/p/8ISd.html

File information:	win32k.sys-20190612-0059.dmp (Jun 11 2019 - 18:59:17)
Bugcheck:	SESSION_HAS_VALID_POOL_ON_EXIT (AB)
Probably caused by:	memory_corruption (Process: csrss.exe)
Uptime:	3 Day(s), 6 Hour(s), 59 Min(s), and 54 Sec(s)

This information can be used by others to help you. I can't help you with this. Someone else will post with more information. Please wait for additional answers. Good luck.

 

[Colif]

Seems the error, although rare, has similar causes to most, outdated drivers.

This issue occurs because some objects that the Win32k.sys driver allocates from the session pool are not freed before the session ends. Defective printer drivers may cause this memory leak condition. When this memory leak occurs, the memory manager issues a Stop error to prevent the system from continuing in an unstable state.

link - just for evidence error can be printer

Do you have a printer? if you do, what make/model? You need newer drivers as the ones below are way too old

Jul 13 2002

serial.sys

Serial Device driver (Microsoft) OR Brother Serial I/F driver http://www.brother.com/

I was not expecting motherboard to have a serial port header but it does