Weird Crash And After Moment System Restarted On 2080 Ti

Monitor:AORUS AD27QD
Cpu: i7-9900K stock 4.7 with Noctua NH-D15
Gpu: Rtx 2080 Ti Aorus Xtreme Waterforce ( stock no oc,temps max 63C on load )
Ram: 32gb 2x16 GB DDR4 GSKILL 3000MHZ XMP
Psu: Seasonic Ultra Prime Titanium 850W
mobo:Asus Prime Z390-A
SSD: Crucial 1TB
HDD: WB BLUE 1 TB
Case: Cooler Master Cosmos C700P

Hello.I left Metro Exodus running by character standing on the same place for 12 hours. When i back to home i press ESC and while exiting from game mission to MAIN MENU ( moment where GPU LOAD drops ) i get BLACK SCREEN and no monitor signal for 1 minute, and crash to desktop with TDR driver crash in event log. After that i log off from system and log back in again and system rebooted unexpected.



I dont have any other issues with stability in games. It happened once. Thx also not able to reproduce. Very weird.

C:/Windows/minidump ,not exist. C:/Windows/memory.dmp not exist.

Here is screens from dump file —> C:/Windows/Livekernelreports/pow32Kwatchdog.dmp <–only that file appear. I think it is related to that pc reboot but not sure.

“A callout to Win32k did not return promptly”


https://i.postimg.cc/xCv9gqjd/w1.jpg


https://i.imgur.com/01kGvUb.jpg



It was psu issue or not?

Also my gpu drivers is from november 2019. Windows 10 up to date. I dont know what is that pow32kwatchdog.dmp.

 

60c is a tad bit warm that is 140 F or so.

 

It's fine, looks like a rare combination of other drivers/Services/background Processes/3rd-party software combining to create a logjam and prevent a timely response to the gfx subsystem. Unlucky timing and nothing more.

Reporting you to Greta Thunberg for flagrant abuse of the planet's finite resources too:

Oh, the Intel forum's waiting for a reply from you, don't waste everyone's time, 4 weeks asking everywhere (you're not already banned from) and still there's no letup from you. Reply there and move on. Your rig's fine.

 

Ok thx Satrow. Really appreciate it for help. So no worries satrow about psu?

 

I'll swap you my 360W Seasonic Gold for your Titanium 850W when I'm gifted more recent/powerful gear that would benefit from it It's fine, no problem.

Your utilities costs though, way above what I'd want to use, esp. for no purpose.

 

giggles i only pay about 60 dollars in electric bill.And most of my stuff is electric and i am running a 850 watt power supply.

 

If you don't have MSI afterburner i would download it and use the fan adjuster on it to about 40 speed.Other wise your GPU is going to get hotter if it get's dusty.

https://www.majorgeeks.com/files/details/msi_afterburner.html

 

Game was running maybe 15 hours. Game not crashed before i click EXIT TO MAIN MENU. It crashed when i press ESC and then clicked EXIT TO MAIN MENU--->moment when GPU LOAD drops it crashed ,instead of EXIT TO MAIN MENU ofc.

I get BLACK SCREEN and no monitor signal for 1 minute, and crash to desktop with TDR driver crash in event log. After that i log off from system and log back in again and system rebooted unexpected.

Happened once,not able to reproduce

Just more precised. Anyway its not reason to worry? Thx Satrow and all you guys for help.

 

In 4+ weeks. unattended tests of up to 20 hours, no-one on ~10 fora able to pinpoint the problem = a complete one-off/unique situation.

Just play your games and have fun, switch off PC when you finish. Get an outdoor hobby and don't worry, your PC is fine.

Did you contact the Intel forum again?

 

I think you explained me enough ,Satrow. So no reason to answering on INTEL forum.

 

To me it looks like there's a suspicion of something Intel may be able to prevent or reduce the frequency of, maybe they suspect a specific 3rd party driver/software of occasionally triggering odd behaviour? They're interested, give them whatever details/logs they ask for please.

 

Ok i will try to answer to INTEL. Also what is pow32Kwatchdog.dmp? Why it generated during unexpected reboot after that crash? ( ctrl alt del when i try to log off and log on again it rebooted )

Just asking i am not technician heh

 

pow = Power, 32K = Win32k.sys (graphics subsystem of the Windows kernel), Watchdog = alert system, flags up potential problems and writes the data, the .dmp file, to the page file on the System drive so it can be logged after a reboot.

So, you pulled the plug on the game, *something somewhere* was blocking the 'reduce power to the gfx card now' call from triggering and sending the reply via the CPU. Might have been that some CPU cores were slow in unparking, too many other things being dealt with at the same time, some of which may well have been triggered by the sudden killing of the game: calls by the nVidia bloatware/telemetry etc, maybe the game or the game launcher platform (Steam, etc.).

Very likely to be a combination of factors, including basic Windows tasks/logging/prefetching/freeing of memory/other 3rd party apps and background processes all hitting the CPU at the same time iy was trying to get all cores up to speed.

Very difficult to diagnose, esp. with someone who has a long history of not giving logs that are asked for by his helpers and who keeps repeating the same questions ad nauseum.

 

So it was caused by psu or not necesaary?

 

NO, not caused by PSU, caused by software in a logjam.

The CPU couldn't tell the GPU the actions were completed in time, *something else* was blocking the queue.

 

Thank you for reply dear Satrow. Last question. You said:" pow = Power So, you pulled the plug on the game, *something somewhere* was blocking the 'reduce power to the gfx card now' ".


So its not caused by psu,because it said "power"?( last question and we can close topic )

 

Same repetitive questions that we always end up with:

NO, it was NOT caused by the PSU, it was caused by *some software/game bug/driver/background Process/Service combination* preventing the CPU from completing the order to reduce the power to the GPU.



They're still waiting for your details over on the Intel forum:

LeoC_Intel (Intel)
2 days ago

Hello SLitw


We would like to look into the issue that you are having, please provide me the following details and data:

• what is the system utilization (CPU, IGPU, DGPU) st the time of the behavior?
• Did the system work properly at some point?
• when did this issue starting happening? Were there any hardware or software changes on the system?
• Please provide me win the information and reports requested in the following link: https://forums.intel.com/s/question...-for-submitting-a-graphics-bug?language=en_US

Leonardo C.

Intel Customer Support Technician

A Contingent Worker at Intel

 

Ok thank you Satrow for help and answer. We can close topic,really appreciate it. Ah yes i answered to INTEL. My language is not very good also. Cheers.

 

Ah last question to be sure. So that unexpected reboot ( ctrl alt del after i log out and log back in again ) was related to that pow32kwatchdog.dmp?
I dont saw a bluescreen. Anyway no worries?

 

The pow32kwatchdog.dmp was created as a result of a combination of factors culminating in a logjam of data which triggered the CPU Watchdog logging the data for the .dmp.

Now that you've made (another) last question here, you can reply on the Intel forum. It'll make a pleasant change for you to answer a question.

 

So not worry about that 1 unexpected reboot after crash? It not caused by ram or psu etc?

 

SOmebody from intel forum said that this reboot was caused by ,power issue ,or ram? Is that true?


https://forums.intel.com/s/question...5-memory-access-violation-crash-after-20-hour


"
Again, if you are also seeing spontaneous reboots (which you have not mentioned before), then yes, you likely have bad DRAM (or a failure in your processor, but very unlikely).

...S"

 

Satrow can you answer me please thanks. Somebody said that KERNEL 41 POWER is an hardware issue.

But also i had C:/Windows/livekernelreports/pow32kwatchdog.dmp

 

I see nothing to make me rethink my opinion, the CPU called an error because the data it was expecting to receive didn't arrive in time.

 

Okies. So before we close topic. It was not hardware fault issue?

You said:" the CPU called an error because the data it was expecting to receive didn't arrive in time."

 

Not a hardware fault imo. Buggy software/bad usage (unnattended game uptime not planned for or expected by game devs - it's not a game server software), possibly made worse by the sudden closure of the game whilst Windows was still recovering from a lower power usage or maintenance mode.

 

@satrow
I wish I had your patience.

 

Hey. Somebody said this to me:"1-It seems gpu/psu degraded. increase gpu core volts a notch and test the game that was causing problem. but before that just install old nvidia drivers and test....
2-RMA the card and try with new one
"


So it can be psu or gpu related or not listen to him?

https://forums.guru3d.com/threads/w...-restarted-on-9900k-and-2080-ti.430153/page-4

 

Not sure how a gpu/psu can become 'degraded'.
If your hardware is capable, then default settings should suffice!
If your overclocking, then a whole new can of worms is opened.

 

Anyone who's followed you through more than 3 of your 'problems' will know how much you love sniffing new hardware boxes, preferably weekly - isn't RMA wonderful?

I'd would be much more concerned about you leaving games running over lengthy periods of time that they're not tested/expected to be run over. Shut it down when you go AFK, it'll last for years and saving on power costs will cover some new tyres for your bicycle too Better still an electrician to check your home's wiring!

It's still not hardware, just like the vast majority of your 'problems', self-inflicted crashes aren't hardware.

 

Okies thank you. Really appreciate for help. Cheers. I go jogging now hehe

 

Shutdown the PC first!!!

 

Wait i dont ask about . Why after reboot i dont had any bugcheck in event log? Only power KERNEL 41.

Somebody said this to me:"
This event says the following:

The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

If there isn't a BugCheck event close by, then the hardware itself has malfunctioned in some way. You could have a faulty power supply, a faulty VRM, an intermittently-shorting system reset switch, or input power voltage drop that's too much for the PSU to handle.

You can rule out an input power problem with a UPS, but the other issues aren't something that you can pinpoint on consumer-grade hardware without external testing equipment. Given that this has only happened once, it's probably not worth pursuing.

 

The same person said this, about an hour ago:

So the jog wasn't a far enough distance to tire you, next time (tomorrow) you reach the Wisła, keep jogging for another three days and give us all a rest, eh.

 

Hehe ok to WISLA next time go jogging. My mother tells me the same lol Ok anyway no worry about hardware fault?

 

Hi Satrow. Can you just explain me that bevaviour.
When game crashed to desktop after black screen, there was an circle instead of arrow,and i was not unable to do anything. So i make CTRL ALT DELETE logged from system and logged back in again, and then reboot happened. Can you explain me why?
Just logging again to system something messed up.

 

The PC/CPU was busy, waitibg for some things, trying to do others - then you gave it something extra to do where you should have waited - relogging reloads the same System state as before (confused/busy) so a reboot happened.

Have patience Grasshopper :~)

 

Oki so nothing wrong with hardware then? last question and we can close topic definitelly

 

Nothing's wrong with the hardware, it'll last for 10 years if you buy a UPS to connect it to and stop abusing the planet by leaving it under load when you go out.

 

ok thank you

 

Hey Satrow. I just forget. I made a screen with that.

https://i.postimg.cc/Qxm87L4X/gf.jpg

1 arrow: 3 driver crashes from a game a CTD
2 arrow: i decide to ctrl alt delete , log out and log back again in
3 arrow: kernel 41 power and weird reboot

 

I am worried somebody said this:"Something is *almost* stable. Things like this are incredibly hard to pinpoint, but generally you'll need to run whatever stress tests you're playing with for 24 hours or more to catch them. My bet is on RAM voltage being slightly too low (which will eventually show up in Memtest after a few days). ANY errors in Memtest, no matter how long you run it (within reason) point to a problem."

What you think something can be not stable ,Satrow? Or just no trust him? I passed memtest,ramtest for many hours no errors. Last question definitelly.

 

Stop playing your helpers against one another, we each have had different experiences, which doesn't make any of them untrustworthy - different times, hardware, software versions = different faults/fixes.

I think your setup is fine, your hardware is fine but the usage (abuse) is off the sensible scale.

 

Thank you SAtrow. All is fine. Just want to ask ,technically. You have excellent knownledge.

That reboot it was BSOD ( even if i dont saw a blue screen on screen ) ? I dont have configured DUMPS in settings.
So Windows/MEMORY.DMP not exist. I dont had bugcheck in event log.

Only file appeared: C:/windows/livekernelreports/pow32kwatchdog.dmp <--- that file was related to restart system ?

Or that pow32kwatchdog.dmp doesnt matter and was not related to that restart.



ps:
This is the last thing i want to know

 

You've almost never supplied any info/logs that have been asked so I must assume there was no BSOD unless you can show me the logs.

The C:/windows/livekernelreports/pow32kwatchdog.dmp was part of a chain of events that led up to the unexpected restart, nothing more.

 

ok thank you. You are fantastic. Thanks again cheers.