Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

BSODs, usually ntoskrnl related


  • Please log in to reply
22 replies to this topic

#1 Underside

Underside

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 19 January 2018 - 09:41 PM

Hello,

 

I am having some recurrent issues with BSODs. They seem to be mostly related to ntoskrnl.exe, although the Bug Check String in BlueScreenView can be a variuety of things, mostly memory related.

 

Now, I have already went through some extensive (IMO...) trying to fix this, but noting seems to do it. I can get a week, maybe even 2 free of BSODs, and then they'd happen again. Last one (which is attached here) happened as I was trying to customize the OSD from MSI Afterburner. Ran GPU-Z and tried to start the rendering just so I have some 3D app to see the OSD, and it got stuck on Loading. I told Windows to close it, and while it tried to do that, when it was "collecting info" or something the PC BSODed... well, not even a full BSOD, as it just froze and seeing the HDD going off I already knew it was saving a dump. After the reboot, yup, bugcheck in Event Viewer. The bugcheck before this (which I guess got deleted) was from Assassin's Creed: Origins. The game is somewhat adept at crashing. In fact, the vast majority of BSODs were during gaming. The last one was quite exceptional occurring with this GPU-Z crash.

 

What I did trying to fix this:

 

- reinstall Windows 10

- avoid installing anything other than the barebones applications; Adobe PS/Illustator, Kaspersky Internet Security, Office are just a few apps I am sadly missing just in attempting to fix the BSODs

- ran memtest86 a number of times, must be like 4 times now, 3 times with the regular 4, and 2 days ago I left it for 8 loops. Nothing, all is fine...

- tried a number of torture tests on the PC, prime, AIDA64, some Kombustor for the GPU... nothing...

- ran Driver Verifier for everything but Microsoft drivers, I think I tried this for days - couldn't get ONE, single BSOD out of it. I even went and played Assassin's Creed: Origins with the horrific stuttering 10-70 fps jumping all over the place, and couldn't extract a BSOD in hours.

- reinstalled Windows 10 (again...)

- tried my best to keep things as default as they can be

- updated the hell out of anything, including some DDU Nvidia reinstalls from Safe Boot

- updated BIOS with the latest that is supposed to contribute to fixing Meltdown/Spectre

- cleared CMOS, ran optimized defaults for a while, then changed to UEFI, which was OK for many days now

- checked HDD with surface scan from both HD Tune and chkdsk

- sfc /scannow, dism, nothing is ever corrupted...

- reseated RAM, CPU, GPU, cables... even SATA and the HDD, everything

 

The power supply is a Zalman 700W GV, which while not ideal obviously, is what I have. Sacrifices had to be made for the 1070ti/8700K combo. It should provide more than enough power to the components, including the GPU, and it does seems to do it as stress testing is just fine. I think the rest of the specs should be addressed by Speccy:

http://speccy.piriform.com/results/jJkoNj1TcN4M8o4XXcHsChT

 

The PC is clean&cool. No throttling or thermal issues. 

 

In case I might have forgot anything, please let me know. Did my best to provide as many details as possible.

 

Thanks!

Attached Files



BC AdBot (Login to Remove)

 


#2 Underside

Underside
  • Topic Starter

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 19 January 2018 - 09:55 PM

Attached are the 4 memtests I was mentioning above. They cover the latest 2 BIOS versions and include the 8 passes test. Performed with XMP, all fine.

 

Edit: attached the Long Generic Test from Seagate Tools. It passed as well as I expected. Also left Prime95 with Small FFTs for 8 hours, no issues, again.

 

But, I bet that if I continue normal use and play some games, give it a week or so and another BSOD will occur for sure.

Attached Files


Edited by Underside, 20 January 2018 - 08:16 AM.


#3 usasma

usasma

    Still visually handicapped (avatar is memory developed by my Dad


  • BSOD Kernel Dump Expert
  • 25,091 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Southeastern CT, USA
  • Local time:04:40 AM

Posted 20 January 2018 - 09:28 AM

I'd suggest also checking over in the General Security forums for your status with the Spectre/Meltdown updates.
They can cause problems with booting and/or BSOD's - but I'm not familiar with the details on them.
Here's a link to the General Security forums:  https://www.bleepingcomputer.com/forums/f/45/general-security/

 

ntoskrnl.exe (also seen as ntkrnlpa.exe, ntkrnlmp.exe, or ntkrpamp.exe) is the kernel (core) of the Windows operating system.  It is protected by security features and the Windows System File Checker.  As such, if ntoskrnl.exe was to blame, you'd be experiencing many more problems other than the occasional BSOD.

In most cases ntoskrnl.exe was blamed because a driver (typically a 3rd party driver) has corrupted the memory space that ntoskrnl.exe considers as it's own.  When this happens, ntoskrnl.exe typically finds unknown data (from the 3rd party driver) in it's memory space.  At this point the OS panics and throws a BSOD to prevent damage to the system.

If the culprit (the offending 3rd party driver) hasn't exited yet, then a BSOD analyst may be able to find traces of it in the reports/dumps.  If the culprit has exited, then the chase is on and further tests/reports will be needed to help identify what actually caused it.

More info here:  https://en.wikipedia.org/wiki/Ntoskrnl.exe

 

From the above info, you can also see that most often the things that cause BSOD's occur in memory - so it's not unusual to see memory related BSOD's.  This doesn't mean that there's a problem with the memory.

 

 

MSI Afterburner and Riva Tuner (along with EVGA Precision) are known to cause BSOD's in some Windows systems (it's driver is usually RTCore64.sys).  Please un-install them immediately!

If you're overclocking, please stop the overclock while we're troubleshooting.  Feel free to resume the overclock once the system has been stabilized.



Your UEFI/BIOS (version 1.20) dates from Dec of 2017.  Please check at the manufacturer's website to see if there are any UEFI/BIOS updates available for your system.  This is just in case there has been a more recent update.
FYI - W8 and W10 communicate more with the UEFI/BIOS than previous versions of Windows, so it's important to ensure that the UEFI/BIOS is kept up to date (and that outdated UEFI/BIOS' may be the cause of some compatibility issues).

Only 5 Windows Update hotfixes installed.  Most build 16299 (1709/Fall Creators Update) systems have more than this.  Please visit Windows Update and get ALL available Windows Updates.
The actual number is not important.  Rather it's important that you checked manually, installed any available updates, and didn't experience any errors when checking or updating.
Check about Spectre/Meltdown in the Security forums first.

 

Please uninstall GPU-Z.  It's driver dates from 2010 and may not be compatible w/W10.

Please uninstall HWINFO.  It's driver dates from before W10 was released and it may not be compatible w/W10

 

Beyond that, the lone memory dump points to video.  As you've clean installed the video drivers (using DDU), I'd have to suspect the card itself to be at fault.
The next step is to test with another card.  If you don't have one, I'd suggest purchasing one from a shop that will let you return it for your money back (if it's not needed).

 

Analysis:
The following is for information purposes only.
The following information contains the relevant information from the blue screen analysis:
**************************Fri Jan 19 20:06:57.081 2018 (UTC - 5:00)**************************
Loading Dump File [C:\Users\john\SysnativeBSODApps\012018-181109-01.dmp]
Windows 10 Kernel Version 16299 MP (12 procs) Free x64
Built by: 16299.15.amd64fre.rs3_release.170928-1534
System Uptime:0 days 2:32:45.763
*** WARNING: Unable to verify timestamp for nvlddmkm.sys
*** ERROR: Module load completed but symbols could not be loaded for nvlddmkm.sys
Probably caused by :memory_corruption
BugCheck 7F, {8, fffff801705b8e50, fffff801705a8f80, fffff8016d8bebff}
BugCheck Info: UNEXPECTED_KERNEL_MODE_TRAP (7f)
Arguments:
Arg1: 0000000000000008, EXCEPTION_DOUBLE_FAULT
Arg2: fffff801705b8e50
Arg3: fffff801705a8f80
Arg4: fffff8016d8bebff
BUGCHECK_STR:  0x7f_8
PROCESS_NAME:  GPU-Z.2.6.0.exe
FAILURE_BUCKET_ID: MEMORY_CORRUPTION_LARGE
CPUID:        "Intel® Core™ i7-8700K CPU @ 3.70GHz"
MaxSpeed:     3700
CurrentSpeed: 3696
  BIOS Version                  1.20
  BIOS Release Date             12/20/2017
  Manufacturer                  Micro-Star International Co., Ltd.
  Product Name                  MS-7B61
  Baseboard Product             Z370 GAMING PLUS (MS-7B61)
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``


3rd Party Drivers:
The following is for information purposes only.
My recommendations were given above. The drivers that follow belong to software or devices that were not developed by Microsoft.  You can find links to the driver information and where to update the drivers in the section after the code box:

**************************Fri Jan 19 20:06:57.081 2018 (UTC - 5:00)**************************
MpKsl878e1588.sys           Mon Feb 27 20:54:41 1989 (2409FBE1)
GPU-Z.sys                   Wed Oct  6 12:14:37 2010 (4CACA06D)
intelppm.sys                Thu Sep 22 06:43:11 2011 (4E7B113F)
HWiNFO64A.SYS               Tue Mar 31 05:51:32 2015 (551A6E24)
iqvw64e.sys                 Mon Apr  3 11:14:26 2017 (58E266D2)
PROCEXP152.SYS              Sun Apr 30 19:23:14 2017 (590671E2)
TeeDriverW8x64.sys          Tue Jul 18 10:15:12 2017 (596E17F0)
iaStorA.sys                 Mon Aug  7 10:23:28 2017 (598877E0)
e1d65x64.sys                Mon Sep 25 08:45:31 2017 (59C8FA6B)
NvStreamKms.sys             Mon Nov  6 22:30:57 2017 (5A0128F1)
RTKVHD64.sys                Tue Nov 14 06:05:07 2017 (5A0ACDE3)
nvvad64v.sys                Thu Dec 14 00:40:29 2017 (5A320ECD)
nvhda64v.sys                Fri Dec 15 03:17:43 2017 (5A338527)
nvlddmkm.sys                Wed Jan  3 18:40:06 2018 (5A4D69D6)


MpKsl878e1588.sys - this driver hasn't been added to the DRT as of this run. Please search Google/Bing for the driver if additional information is needed.
http://www.carrona.org/drivers/driver.php?id=GPU-Z.sys
http://www.carrona.org/drivers/driver.php?id=intelppm.sys
http://www.carrona.org/drivers/driver.php?id=HWiNFO64A.SYS
http://www.carrona.org/drivers/driver.php?id=iqvw64e.sys
http://www.carrona.org/drivers/driver.php?id=PROCEXP152.SYS
http://www.carrona.org/drivers/driver.php?id=TeeDriverW8x64.sys
http://www.carrona.org/drivers/driver.php?id=iaStorA.sys
http://www.carrona.org/drivers/driver.php?id=e1d65x64.sys
http://www.carrona.org/drivers/driver.php?id=NvStreamKms.sys
http://www.carrona.org/drivers/driver.php?id=RTKVHD64.sys
http://www.carrona.org/drivers/driver.php?id=nvvad64v.sys
http://www.carrona.org/drivers/driver.php?id=nvhda64v.sys
http://www.carrona.org/drivers/driver.php?id=nvlddmkm.sys


 
My browser caused a flood of traffic, sio my IP address was banned. Hope to fix it soon. Will get back to posting as soon as Im able.

- John  (my website: http://www.carrona.org/ )**If you need a more detailed explanation, please ask for it. I have the Knack. **  If I haven't replied in 48 hours, please send me a message. My eye problems have recently increased and I'm having difficult reading posts. (23 Nov 2017)FYI - I am completely blind in the right eye and ~30% blind in the left eye.<p>If the eye problems get worse suddenly, I may not be able to respond.If that's the case and help is needed, please PM a staff member for assistance.

#4 Underside

Underside
  • Topic Starter

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 20 January 2018 - 10:45 AM

Alright let's go one by one.

 

- I'll get MSI AB off, although it saddens me to no end, as it leaves me without any tool to analyze what happens to the GPU, especially when coupled with GPU-Z removal

- HWinfo64/GPU-Z are standalone, not installed so I guess nothing loads unless I actually run them, yet it is true, the BSOD happened during GPU-Z

- there's a lot of Intel stuff in the drivers too; unsure I can do anything about it, it gets installed with Windows 10/patches. In a few cases I just updated from the manufacturer, MSI or Intel, who had the latest.

- there's also the antimalware from Microsoft which I hate and really want KIS back bur hey...

- and lastly there's the Realtek audio driver. 

- there's no overclock at all on anything, well, you can consider XMP as OC, but as you saw the RAM seems to go through the motions just fine with memtest86

- the low number of updates is because the OS install is quite fresh

- the BIOS is the latest available, with the Spectre/Meltdown "fix"; but there were BSODs with the previous BIOS too. The videocard has no other Vbios available yet.

 

Now for the GTX 1070Ti, the card. I wish there was a way to properly test it for errors, like there is with RAM, I actually looked into it, but there's nothing I could find that is able to test all the VRAM properly, and stress testing didn't result in any BSODs. I can't see how I will ever be able to RMA the card if they can't replicate the BSODs, and I can't either, they just happen way too inconsistently and sometimes it can take 1-2 weeks for one to actually happen. If I RMA it like this there's a big chance they will just send it back as it is if they cannot prove the card is unstable. 

 

Any advice on how to test the card properly and maybe reliably cause BSOD if it is the culprit? Because even if I'd buy another GPU, there's a good chance I can achieve no BSODs/stability for weeks even with the current one, and there's no way I can buy a new GPU and keep it for a month here, the most is 10 days.

 

For now, I'll remove MSI AB (I just reinstalled it yday) and stop using GPU-Z/HWinfo64. I'll give the video drivers another DDU reinstall, play some games see if it crashes. Still, would be amazing to find a way to reliably BSOD.

 

Thanks a lot for the detailed response. I will return with more dumps if they happen and updates if there are any. The help is appreciated!



#5 HyperHenry

HyperHenry

  • Members
  • 775 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Iowa (but travel)
  • Local time:03:40 AM

Posted 20 January 2018 - 12:01 PM

The voltages reported in the Speccy appear to be low but Speccy isn't amazingly accurate. Could you test your voltages with CPUZ? Have you tried to update the graphics drivers? They also spawn a lot of BSODs.



#6 Underside

Underside
  • Topic Starter

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 20 January 2018 - 01:03 PM

The voltages reported in the Speccy appear to be low but Speccy isn't amazingly accurate. Could you test your voltages with CPUZ? Have you tried to update the graphics drivers? They also spawn a lot of BSODs.

The graphics drivers are the latest available. There were BSODs on past drivers too however. I just reinstalled them again, see what happens.

As for voltages, CPU-Z has pretty much nothing but the core voltage, which is on Auto on BIOS. If anything, the MSI board seems to set voltages quite high. The lower values are just the result of power saving features. There is one voltage that worries me a bit, the 12V line is 11.712 in idle and drops to 11.520 when I would do something like stress the CPU for example in CPU-Z and run the GPGPU bench in AIDA64 in parallel. I really don't remember those numbers being that low in my old FX 8350/GTX 970 PC.

Then again, people keep saying the sensor readings are not accurate, and I don't have the tools to read the exact voltages. I will change the PSU later this year anyway, sadly now it's not an option as I barely managed to squeeze the 8700K/RAM/motherboard/1070Ti/cooler in my budget. A new, better PSU, larger case, a decent SSD are my first priorities a bit later this year, but they will have to wait a while.



#7 HyperHenry

HyperHenry

  • Members
  • 775 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Iowa (but travel)
  • Local time:03:40 AM

Posted 20 January 2018 - 01:46 PM

11.52 is in a healthy range. Can you provide the full specs of your computer please?



#8 Underside

Underside
  • Topic Starter

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 20 January 2018 - 02:17 PM

11.52 is in a healthy range. Can you provide the full specs of your computer please?

Everything should be in the Speccy link in the first post. Only things that are not there are the case (Thermaltake Versa 2), PSU (ZALMAN GV 700W), CPU cooler (Zalman CNPS10x Optima). Full names for the other hardware:

 

- Intel i7 8700K

- ADATA XPG Dazzle 16GB DDR4 3000MHz CL16 Dual Channel kit

- MSI Z370 GAMING PLUS 

- GIGABYTE GeForce GTX 1070 Ti GAMING 8G

- some sort of  WD Blue 1TB Desktop Hard Disk Drive - 7200 RPM SATA 6Gb/s 64MB Cache from a few years ago

 

11.52 indeed seems within specs, checked the Zalman page for the PSU - 11.4-12.6 range. The 12v rail should offer 648W/56A, and with the Bronze 88% efficiency it's still enough for the 8700k/1070ti without OC. The CPU sucks 95W and the GPU around 170-180, topping at 220W or so if you allow it the max. of 120% Power Target in MSI Afterburner. Even when I did that and ran torture tests no BSODs happened. They just happen randomly and mostly during some game. Torture testing just seem to be pretty fine for whatever reason, which is somewhat annoying as I can't RMA stuff that withstands torture, the store will mail it back and charge me for transport/their diagnostic procedures.



#9 HyperHenry

HyperHenry

  • Members
  • 775 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Iowa (but travel)
  • Local time:03:40 AM

Posted 20 January 2018 - 02:37 PM

Have you tried reseating your GPU and blowing out the slot? Also can you use it for some gaming or other intense use and then run another Speccy report?



#10 Underside

Underside
  • Topic Starter

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 20 January 2018 - 03:07 PM

Have you tried reseating your GPU and blowing out the slot? Also can you use it for some gaming or other intense use and then run another Speccy report?

Yeah, reseated everything like twice already. Everything is clicked into place and clean. I'm gonna try some Deus Ex: Mankind Divided later on, it's pretty demanding, actually it might just be the 2nd most demanding title I've ever played (after Assassin'S Creed: Origins, but I finished that). Chances are there won't be any BSODs,as they are quite rare :( But, if we're "lucky", we gonna get one, and I'll post it here.

 

From Reliability History:

 

- fresh OS 10.01.2018

- 1 bugcheck: 13.01, i was actually able to find some info about it, guess they get written over when a new one occurs: 

The bugcheck was: 0x0000000a (0x0000000000000020, 0x0000000000000002, 0x0000000000000000, 0xfffff801d05d4b92). 

Pretty sure this one was IRQL_NOT_LESS_OR_EQUAL something-ish

- 1 bugcheck 18.01

The bugcheck was: 0x000000fc (0xffffe38d000041c0, 0x810000013c140863, 0xffff88813acf7010, 0x0000000000000002)

ATTEMPTED EXECUTE ON NOEXECUTE MEMORY

- 1 bugcheck 20.01

The bugcheck was: 0x0000007f (0x0000000000000008, 0xfffff801705b8e50, 0xfffff801705a8f80, 0xfffff8016d8bebff).

UNEXPECTED_KERNEL_MODE_TRAP

 

Last one it's the one from the original post.

 

Almost 100% sure all of them involved ntoskrnl. I should have posted each of them but I was hoping I could fix it myself, but after getting 2 of them in 3 days I got annoyed and posted, since it's kind of obvious I can't solve this through normal troubleshooting.

 

We'll see if it crashes with Deus Ex. Gonna try to push it as much as possible while keeping a playable framerate.



#11 HyperHenry

HyperHenry

  • Members
  • 775 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Iowa (but travel)
  • Local time:03:40 AM

Posted 20 January 2018 - 03:14 PM

I'm more interested in seeing the temps after a lot of use.



#12 Underside

Underside
  • Topic Starter

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 20 January 2018 - 03:20 PM

I'm more interested in seeing the temps after a lot of use.

I can create a log from HWinfo64 or even AIDA64 for all the components. There won't be anything interesting in it, I can already tell you that the GPU will reach a max 68C or so with the default fan settings which are quite conservative/quiet, the 8700K will spike for 75C or so and stabilize at around 65C for stress testing, and 55C for games that are able to use the 12 threads. I'll log a gaming session and post. I'm not sure what's "long" though, there's a good chance I won't be able to play for more than 1-2 hrs, kinda have to work as well on this PC.



#13 Underside

Underside
  • Topic Starter

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 20 January 2018 - 07:51 PM

Attaching the log from about 1 hour of Dragon Age Inquisition, maxed/4MSAA/no Vsync/1080p. Told you it's gonna be boring, max 68C/GPU, max 60C/CPU core. Averages 66C for GPU and about 50C for CPU cores. 

 

You gonna need something like Office/Excel to read it. Google Sheets would work online. Select desired column, insert a chart, and you gonna see how things evolved over timed. But then again, it's still gonna be quite boring and nothing seems to overheat at first glance, as there's no way to guess how other components without sensors do.

Attached Files

  • Attached File  logz.zip   223.55KB   1 downloads


#14 HyperHenry

HyperHenry

  • Members
  • 775 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Iowa (but travel)
  • Local time:03:40 AM

Posted 20 January 2018 - 08:59 PM

Have you checked if this might be due to the Spectre/Meltdown thing? I am out of ideas.



#15 Underside

Underside
  • Topic Starter

  • Members
  • 15 posts
  • OFFLINE
  •  

Posted 20 January 2018 - 09:09 PM

Have you checked if this might be due to the Spectre/Meltdown thing? I am out of ideas.

No worries, I am out of ideas as well :) tried pretty hard to fix it. I'll have to wait for another BSOD, uninstalled some more software like MSI Afterburner, will stop using HWinfo/GPU-Z, reinstalled the driver AGAIN with DDU, and see if it happens again.

 

As for the Spectre/meltdown situation, well, I cannot be 100% it's not that. It might. I had some BSODs before the fixes, but thing is, I was using an old Windows install that I had with my old PC so I didn't pay attention to them as I expected some bugs to happen after I pretty much used a 2 yrs old OS upgrade 2-3 times that was running on a AMD FX CPU. Only started to pay attention after I installed everything clean and tested the RAM/CPU.

 

Another issue is that I don't really have time to play games for 5 hours on end, I can run tests while asleep, maybe I'll try leaving a game open overnight too. Or maybe I will try to loop some benchmark for a few hours, as Kombustor/Furmark tests didn't manage to BSOD even under Driver Verifier.

 

But yeah, in the end I have to somehow find 100% what is causing this, so I can either keep or send back the card. Not in a rush, as there are 3 years of warranty, but... the sooner the better. 






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users