[SOLVED] Random restarts are not producing any minidump files

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
Hi all, I am currently experiencing a problem I can't seem to wrap my head around and could really use the help.

About a month ago I did a couple upgrades on my build. I swapped the GPU, CPU, and motherboard out and replaced them with knew components (prev. build and current build listed below). Since then everything has been just dandy for the most part. The only problem I have encountered is that my computer will randomly restart itself and/or BSOD for no apparent reason and 99% of the time it's when my pc is idle. Upon checking the minidump folder I cannot find any from today's date. The thing is, is I know my computer creates minidump files as when I was attempting to OC in the past I had plenty from all the BSODs. I tried messing with the settings and checking and unchecking "overwrite existing minidump file", choosing different file sizes, etc... nothing works. After all of this was said and done I figured the random restarts had to be because I didn't do a fresh install of windows when I swapped the mobo and CPU out.

I ended up completely reformatting my OS SSD and reinstalling windows and drivers. I kept my PC on for 5 days with no problems. I played Cyberpunk 2077 for 7 hours last night with no problems. Today, randomly, my computer decided to restart and still provided no minidump file??? So confusing! Event viewer shows that it shutdown unexpectedly too so I am lost. From a month ago up until this point I have swapped literally every component out except for the PSU and the ssd that the OS sits on. I've run /chkdsk and /sfc as well as other programs that tell me my SSD is relatively healthy (~6 yrs old). I thought maybe it could be the PSU as it is approaching 7 to 8 years, however, I feel like I would encounter issues when gaming and under load...not idle. If I could get any insight or help on this that would be greatly appreciated! Thank-you!

Prev. Build ( ~1 month ago)
  • AsRockz77 mobo
  • i5 3570k
  • msi gtx 770
  • Corsair HX750
New(er) Build
  • Asus B450-f mobo
  • Ryzen 5 3600
  • Evga gtx 1660ti
  • Corsair HX750
 
Solution
if it worked it would set it to a locked speed. it would be a temporary fix until amd puts out their updates in january of 2021
hopefully it will allow you to use your system without bugchecking all of the time.
I have been telling people to turn off power management functions but I still see a lot of these bugcheck going out of amdppm.sys
I hope that disabling the driver will block the power management functions that are calling bad memory addresses.
i have mostly seen this with people using the newer amd chips out of china. These were on the grey market and were meant for OEM machines that would have proper BIOS settings. here is the version string:
REG_SZ AMD64 Family 23 Model 113 Stepping 0
I can find the list of known...

Colif

Win 11 Master
Moderator
restarts with no minidumps can mean windows isn't the cause, and its hardware.

PC crashing at idle... idle is when it runs background tasks. its funny how many high powered PC crash while they doing nothing but can run full power all the time.

what storage drives do you have?? what brand is the ssd?
 

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
restarts with no minidumps can mean windows isn't the cause, and its hardware.

PC crashing at idle... idle is when it runs background tasks. its funny how many high powered PC crash while they doing nothing but can run full power all the time.

what storage drives do you have?? what brand is the ssd?

Yup, funny indeed lol. And storage setup is a bit weird atm. Currently have 3 ssds. Kingston Hyperx (120gb, I put the os on here), another Kingston hyperx(250gb), and I recently bought a 500gb Samsung ssd for extra room. Thinking about swapping the os to the newer ssd and trying that out.
 

Colif

Win 11 Master
Moderator
its not really funny, its just a common problem. i need to figure it out so its less confusing.

120gb is too small for windows 10 these days, we suggest a 250 at very least as it gives you room to grow into instead of having to worry about running out and moving files around. I have 1tb boot drive, its perhaps too much but I won't be running out any time soon. Its so big it makes my other hdd kind of pointless.
 

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
its not really funny, its just a common problem. i need to figure it out so its less confusing.

120gb is too small for windows 10 these days, we suggest a 250 at very least as it gives you room to grow into instead of having to worry about running out and moving files around. I have 1tb boot drive, its perhaps too much but I won't be running out any time soon. Its so big it makes my other hdd kind of pointless.
I know what you meant lol, I've been tearing my hair out the past few weeks trying to figure it out. Will definitely switch the os to the 500gb ssd. I honestly think it could be a problem with ryzen cpus? In a lot of threads I've been seeing people with ryzen cpus experience the same idle reboots but maybe I am overthinking it?
 

Colif

Win 11 Master
Moderator
my pc
Gigabyte X570 motherboard
Ryzen 5 3600xt
Nvidia 2070S
Seasonic Focus 750watt Plat PSU.

If Ryzen CPU crashed from sitting at idle, I would know about it as that is all mine does most of the day. Idle shouldn't be any more stressing on PC than full load.
 
  • Like
Reactions: xRatspike

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
another possible reasons for no minidumps is, if the drive its trying to record them on is the cause, it can't write crashes to it

you could run this on your Kingston drives, it only really looks at SMART scores but its a start - https://www.kingston.com/en/support/technical/ssdmanager

samsung ssd would use Samsung Magician - https://www.samsung.com/semiconductor/minisite/ssd/download/tools/
Thanks for the links ill definitely be running them on the drives after work. For what its worth, I was messing around with an OC a few days ago (returned everything to normal so stock atm) so I was running into bsod as a result. Minidumps were recorded for one of the bsod. So it seems its capable of writing them...sometimes? Will be running those scans for further info.
 

Colif

Win 11 Master
Moderator
oh. well, if it can make them and doesn't when it crashes, its not windows. We need to look at hardware causes.

starting with storage as good a place as any

I assume it didn't crash on last PC. since its almost all new parts.
Is PC clean inside? crashes can be caused by heat build up.

any other parts in PC you haven't mentioned?

what power plan are you running? AMD Balanced?
 

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
oh. well, if it can make them and doesn't when it crashes, its not windows. We need to look at hardware causes.

starting with storage as good a place as any

I assume it didn't crash on last PC. since its almost all new parts.
Is PC clean inside? crashes can be caused by heat build up.

any other parts in PC you haven't mentioned?

what power plan are you running? AMD Balanced?
My thoughts as well. Pc is kept clean and air dusted often, temps are fine with cpu ~40 idle and never goes above 70/75 when gaming. Gpu is around the same never going above 70. You are correct this has not happened before I swapped hardware. I swapped the ram out as well and have run memtest multiple times with no issues. I was running balanced but recently switched it to high performance to see if itd fix anything.

I ran sfc and it fixed some corrupted files which I thought was odd considering I only reinstalled windows about a week ago.
 

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
Update:
Clean installed windows onto the new Samsung drive. Did it with the other drives unplugged which I did not do before. I am going to run the build off of this single ssd just to completely eliminate error from the other two.

I wanted to add as well, aside from swapping hardware I also upgraded from windows 7 to windows 10. The Kingston drive that I always had the os on always ran windows 7 until upgrading to windows 10. Not sure if this could imply anything but figured I'd throw it out there.

Now we play the waiting game!
 

Colif

Win 11 Master
Moderator
I wanted to add as well, aside from swapping hardware I also upgraded from windows 7 to windows 10. The Kingston drive that I always had the os on always ran windows 7 until upgrading to windows 10.
I don't think its likely to be reason for lack of dumps, the location for those hasn't changed in several versions and the instructions are almost same for moving them in win 7 as they are in 10.

Upgrades can have weird problems though. Many of us clean installed win 10 after we upgraded to make sure it was working correctly. I was lucky and lasted until Xmas 2016 before I needed to reinstall
 

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
I don't think its likely to be reason for lack of dumps, the location for those hasn't changed in several versions and the instructions are almost same for moving them in win 7 as they are in 10.

Upgrades can have weird problems though. Many of us clean installed win 10 after we upgraded to make sure it was working correctly. I was lucky and lasted until Xmas 2016 before I needed to reinstall

Gotcha, that makes sense. PSU comes in tomorrow. Hopefully after swapping this out I won't have to make another update on this thread, however, if it does happen again after this swap I am inclined to believe that there may be something wrong with the motherboard or cpu but time will tell. Thank-you for you're time so far as well!
 

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
I don't think its likely to be reason for lack of dumps, the location for those hasn't changed in several versions and the instructions are almost same for moving them in win 7 as they are in 10.

Upgrades can have weird problems though. Many of us clean installed win 10 after we upgraded to make sure it was working correctly. I was lucky and lasted until Xmas 2016 before I needed to reinstall
UPDATE: Unfortunately, after changing the psu, surge protector, ssd, gpu, and running memtest for 10 hours on the ram, my computer restarted today again. Good news is that I actually have a dump file for the crash. If you, or anybody could analyze this for me I would greatly appreciate it so I can be one step closer to sorting this out! Thank-you!

https://drive.google.com/file/d/1cbnSe1vYUWce3mO2AQTbh4uUSfjhmJU9/view?usp=sharing (minidump)

edit: I also am unsure why this produced a minidump while the others did not. I was away at work and came back to the login screen.
 
Last edited:
you might want to remove this file:
C:\Windows\system32\drivers\AsIO2.sys Tue Apr 9 02:59:16 2019
(hackers released a code sample showing how to hack a system via this driver)


the system died after going into this file
SystemRoot\System32\drivers\amdppm.sys
this is the microsoft version of the amd power management module.

something is just messed up with this version of the file.


-it might be related to amd indicating that they will have fixes for 400 series chipsets and bios settings coming in January of 2020.

I would see if your microsoft update has and driver fix they are trying to push out to your machine.
(windows update, check the optional updates and then the check for updates option)

I have seen this dozens of times and no one really has a real fix.



6: kd> !sysinfo machineid
Machine ID Information [From Smbios 3.2, DMIVersion 0, Size=2530]
BiosMajorRelease = 5
BiosMinorRelease = 13
BiosVendor = American Megatrends Inc.
BiosVersion = 3103
BiosReleaseDate = 06/17/2020
SystemManufacturer = System manufacturer
SystemProductName = System Product Name
SystemFamily = To be filled by O.E.M.
SystemVersion = System Version
SystemSKU = SKU
BaseBoardManufacturer = ASUSTeK COMPUTER INC.
BaseBoardProduct = ROG STRIX B450-F GAMING
BaseBoardVersion = Rev 1.xx
6: kd> !sysinfo cpuinfo
[CPU Information]
~MHz = REG_DWORD 3600
Component Information = REG_BINARY 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
Configuration Data = REG_FULL_RESOURCE_DESCRIPTOR ff,ff,ff,ff,ff,ff,ff,ff,0,0,0,0,0,0,0,0
Identifier = REG_SZ AMD64 Family 23 Model 113 Stepping 0
ProcessorNameString = REG_SZ AMD Ryzen 5 3600 6-Core Processor
Update Status = REG_DWORD 1
VendorIdentifier = REG_SZ AuthenticAMD
 

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
you might want to remove this file:
C:\Windows\system32\drivers\AsIO2.sys Tue Apr 9 02:59:16 2019
(hackers released a code sample showing how to hack a system via this driver)


the system died after going into this file
SystemRoot\System32\drivers\amdppm.sys
this is the microsoft version of the amd power management module.

something is just messed up with this version of the file.


-it might be related to amd indicating that they will have fixes for 400 series chipsets and bios settings coming in January of 2020.

I would see if your microsoft update has and driver fix they are trying to push out to your machine.
(windows update, check the optional updates and then the check for updates option)

I have seen this dozens of times and no one really has a real fix.



6: kd> !sysinfo machineid
Machine ID Information [From Smbios 3.2, DMIVersion 0, Size=2530]
BiosMajorRelease = 5
BiosMinorRelease = 13
BiosVendor = American Megatrends Inc.
BiosVersion = 3103
BiosReleaseDate = 06/17/2020
SystemManufacturer = System manufacturer
SystemProductName = System Product Name
SystemFamily = To be filled by O.E.M.
SystemVersion = System Version
SystemSKU = SKU
BaseBoardManufacturer = ASUSTeK COMPUTER INC.
BaseBoardProduct = ROG STRIX B450-F GAMING
BaseBoardVersion = Rev 1.xx
6: kd> !sysinfo cpuinfo
[CPU Information]
~MHz = REG_DWORD 3600
Component Information = REG_BINARY 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
Configuration Data = REG_FULL_RESOURCE_DESCRIPTOR ff,ff,ff,ff,ff,ff,ff,ff,0,0,0,0,0,0,0,0
Identifier = REG_SZ AMD64 Family 23 Model 113 Stepping 0
ProcessorNameString = REG_SZ AMD Ryzen 5 3600 6-Core Processor
Update Status = REG_DWORD 1
VendorIdentifier = REG_SZ AuthenticAMD

Just checked, there aren't any updates unfortunately. Do you think rolling back chipset drivers could work? Or are those different drivers? Really unsure on where to go from here besides going intel lol.

Edit: Just checked the reliability history and included screenshots below. Is this an unrelated issue?! It is telling me hardware failure, however, I don't know what the error codes mean.

https://drive.google.com/file/d/1FB0HR6cQ_VgJcIeyUIQOmg2sDp6aUz0-/view?usp=sharing

https://drive.google.com/file/d/151Q2eSIG2zpOHvS_cnq7gjeylrCrf74z/view?usp=sharing
 
Last edited:
the error 117 might be
bugcheck 0x117 VIDEO_TDR_TIMEOUT_DETECTED
and
bugcheck 0x141 VIDEO_ENGINE_TIMEOUT_DETECTED

it would be consistent with your error at the same time that said
radeon software host stopped working

basically your graphics card stopped responding for some reason. if it is a older graphics card it can get dust in the fan and over heat. I even had one that the fan popped off the little motor and motor would spin and the fan did not.
when I went to blow the dust out, the fan blade came out and I was able to put it back on.

sorry, i don't know if it could cause these other problems or if the other problems cause the GPU driver to mess up.

I was thinking you might disable the amdppm.sys driver. It will make the system run with defaults and you will not be able to make changes to various items until it is re enabled.

I think the registry key would be located here:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\amdppm
(look at the start value)
see what it is currently set to (maybe 2 for automatic startup?)
you can set it to 4 for disabled then reboot and see if the system is more stable.
if it does not help, you would just set it back to its original value.
here is the info on what the values might be:
ServiceStartMode Enum (System.ServiceProcess) | Microsoft Docs

 

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
the error 117 might be
bugcheck 0x117 VIDEO_TDR_TIMEOUT_DETECTED
and
bugcheck 0x141 VIDEO_ENGINE_TIMEOUT_DETECTED

it would be consistent with your error at the same time that said
radeon software host stopped working

basically your graphics card stopped responding for some reason. if it is a older graphics card it can get dust in the fan and over heat. I even had one that the fan popped off the little motor and motor would spin and the fan did not.
when I went to blow the dust out, the fan blade came out and I was able to put it back on.

sorry, i don't know if it could cause these other problems or if the other problems cause the GPU driver to mess up.

I was thinking you might disable the amdppm.sys driver. It will make the system run with defaults and you will not be able to make changes to various items until it is re enabled.

I think the registry key would be located here:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\amdppm
(look at the start value)
see what it is currently set to (maybe 2 for automatic startup?)
you can set it to 4 for disabled then reboot and see if the system is more stable.
if it does not help, you would just set it back to its original value.
here is the info on what the values might be:
ServiceStartMode Enum (System.ServiceProcess) | Microsoft Docs
Thank you for the reply! I just got this card and it runs great but amds adrenaline drivers are weird so it's no surprise. I actually don't think this is that big of an issue and am pretty certain it might just be radeon bugging out. Anyway, as far as disabling the driver goes, would this prove to be a long term solution considering it works? I read somewhere that this set some guys cpu to a locked speed. Would I be able to use the pc as normally? Really considering just buying an Intel board and cpu seeing as this is sort of out of my control to a certain extent yknow? Really unfortunate that this happens with a new amd board and cpu but what can you do I guess.
 
Thank you for the reply! I just got this card and it runs great but amds adrenaline drivers are weird so it's no surprise. I actually don't think this is that big of an issue and am pretty certain it might just be radeon bugging out. Anyway, as far as disabling the driver goes, would this prove to be a long term solution considering it works? I read somewhere that this set some guys cpu to a locked speed. Would I be able to use the pc as normally? Really considering just buying an Intel board and cpu seeing as this is sort of out of my control to a certain extent yknow? Really unfortunate that this happens with a new amd board and cpu but what can you do I guess.
if it worked it would set it to a locked speed. it would be a temporary fix until amd puts out their updates in january of 2021
hopefully it will allow you to use your system without bugchecking all of the time.
I have been telling people to turn off power management functions but I still see a lot of these bugcheck going out of amdppm.sys
I hope that disabling the driver will block the power management functions that are calling bad memory addresses.
 
if it worked it would set it to a locked speed. it would be a temporary fix until amd puts out their updates in january of 2021
hopefully it will allow you to use your system without bugchecking all of the time.
I have been telling people to turn off power management functions but I still see a lot of these bugcheck going out of amdppm.sys
I hope that disabling the driver will block the power management functions that are calling bad memory addresses.
i have mostly seen this with people using the newer amd chips out of china. These were on the grey market and were meant for OEM machines that would have proper BIOS settings. here is the version string:
REG_SZ AMD64 Family 23 Model 113 Stepping 0
I can find the list of known bugs but only for the CHIPS made directly for AMD not for these new chips out of china
they have new functions, new power requirements and they show a different version than the other amd chips.
I think microsoft might not apply the correct amd cpu microcode patch for these CPUs this would mean certain bugs in the cpu would not be patched.
one of the known bugs in the earlier versions of the chip has to do with DMA access, and the fix was to reserve the address in bios so no one attempts to use it.
if it does not get reserved then it could be used by your GPU DMA controller and cause crashes.

that is why I was hoping people have windows error reporting turned on, microsoft can live debug your machine and send out fixes in the windows update, advance driver update.

there has been a lot of changes going on in the low level code for the cpu, bios, and cpu patches
since some people have published code on how to use motherboard utilities to put code in the bios so they can infect machines that get wiped.
vendors are trying to plug the holes and restrict all of the little extras that allow motherboard vendors tools to work. mostly overclocking tools and monitoring tools. why find a windows flaw when you can just the interface to a overclocking tool to infect a machine.
i think in the future you will not be able to install these tools and use secure boot. Microsoft will block access or and will bugcheck just like other attacks on the system.
 
Last edited:
Solution

xRatspike

Distinguished
Jan 9, 2015
31
0
18,540
i have mostly seen this with people using the newer amd chips out of china. These were on the grey market and were meant for OEM machines that would have proper BIOS settings. here is the version string:
REG_SZ AMD64 Family 23 Model 113 Stepping 0
I can find the list of known bugs but only for the CHIPS made directly for AMD not for these new chips out of china
they have new functions, new power requirements and they show a different version than the other amd chips.
I think microsoft might not apply the correct amd cpu microcode patch for these CPUs this would mean certain bugs in the cpu would not be patched.
one of the known bugs in the earlier versions of the chip has to do with DMA access, and the fix was to reserve the address in bios so no one attempts to use it.
if it does not get reserved then it could be used by your GPU DMA controller and cause crashes.

that is why I was hoping people have windows error reporting turned on, microsoft can live debug your machine and send out fixes in the windows update, advance driver update.

there has been a lot of changes going on in the low level code for the cpu, bios, and cpu patches
since some people have published code on how to use motherboard utilities to put code in the bios so they can infect machines that get wiped.
vendors are trying to plug the holes and restrict all of the little extras that allow motherboard vendors tools to work. mostly overclocking tools and monitoring tools. why find a windows flaw when you can just the interface to a overclocking tool to infect a machine.
i think in the future you will not be able to install these tools and use secure boot. Microsoft will block access or and will bugcheck just like other attacks on the system.
If anything, I will check it out tomorrow and see if changing the registry value works. I'm not sure I completely understand what you are saying but basically the type of chip I have is the one from China correct? Is there a way to find out? I understand that this issue is more or less dependent on Windows and/or AMD for releasing drivers and such like you mentioned.

Would it be unreasonable to assume that all of the other restarts I was having in the past were also due to this amdpps.sys driver? And for whatever reason I was lucky enough to get a dump file this time? If this is the case I might just be switching to Intel. Primarily because I think sacrificing my CPUs performance for a temporary fix is kind of ridiculous. However from my understanding there really isn't much else I can do.

Thank you for all the information you've provided, especially with all the indepth info on why this might be happening, I really do appreciate it. Honestly I am just trying to have my system stable even if it means buying more new parts. I would like to wait until January 2021 for a possible fix from amd or even windows but I just don't know if I want to sacrifice my performance until that happens (if it happens).