[SOLVED] Many BSODs on fresh build ?

[codelode]

Starting Friday, December 2nd, I started building my new PC with custom water-cooling. I thought the hardest part would be the loop and not murdering my PC with liquid, but the real trouble started late Friday night when I transferred my SSDs from my current build to the new build, and tried to reinstall windows on it.

Specs:

CPU	Ryzen 9 7950x
MOBO	Gigabyte x670e AORUS XTREME
GPU	Gigabyte RTX 4090 Gaming OC
RAM	2x16gb G.Skill Trizent Z5 NEO RGB
Boot Drive	Sabrent 2TB (SB-Rocket-2TB)
Second & Third Drive	Sabrent 2TB (SB-Rocket-2TB)
Fourth Drive	Sabrent ROCKET 4 Plus-G 4TB (SB-RKTG-4TB)
PSU	Corsair HX1200

After installing all the drives, I booted it up and was greeted with a low-res windows 11 install from my current build. This confirmed to me it was working well enough, so I shut it down then plugged in the win 11 boot media I had created in order to reformat the drive and start fresh.

When I was in the win 11 installation process, I chose to do a custom install, and deleted + formatted my previous C drive, then chose that drive as the drive to install to. The installer seemed happy with this for a moment, then I was hit with the first of many BSODs of the night. I don't recall exactly what the first BSOD was, but I'll list all of the BSODs I would encounter below.

Next, I tried to install again in various ways, all of them failing with BSODs, until eventually I also tried to boot into an Ubuntu live media drive, which got me some progress, save for some issues with nouveau that had to be dealt with. I got to the install process of ubuntu, then it hung for a bit, then hung for a long bit, and I gave up.

I would go to retry Windows 11 again, and this time after messing with BIOS settings around CMS and secure boot, I managed to get windows 11 fully installed, but I was stuck in a loop of doing the first 3 steps of the introduction (choose country, keyboard, language) where without fail in these three views I would BSOD again, or just crash to black screen, and be greeted with the first step again on next boot.

Thats where I ended this morning, at 4am. Next, I decided to force access to cmd by getting into recovery mode and tried to do some repair and scanning there with sfc, dism, and chkdsk and some other tool, only sfc found corrupt files and repaired them (after every bsod, it would find more corrupt files).

Finally, I caved to friend pressure and tried installing Windows 10. Windows 10 somehow went through the whole install process without fail, and I was able to get to desktop and install device drivers and programs like Steam etc. Then, when I was installing some windows updates, I got BSOD again. Also, I would get BSOD on the first launch of any game on steam, but not thereafter. I've encountered this crash with Skyrim SE and MW2/Warzone 2.0, and thats where I am now. As long as I don't install anything, or perform any updates, the machine seems stable. But when I start installing and updating things, crashes appear.

Temperatures, power levels, clock speeds etc all look perfectly within line for a waterloop pc (CPU, GPU on 2x 360mm rads), and I was able to get into a warzone match and run the game for ~30m without any issues, consistent framerate, etc.

Personally, I think the issue lies with RAM, and I've ordered some replacement test ram to try out for tomorrow, but I want to get feedback here as well. I have 3 minidump files here. The list of BSODs I remember I've experienced across this whole ordeal are as follows:

IRQL_NOT_LESS_OR_EQUAL
DPC_WATCHDOG_VIOLATION
ATTEMPTED_TO_WRITE_TO_READONLY_DIRECTORY
REFERENCE_BY_POINTER
KERNAL_SECURITY_CHECK_FAILURE
ntoskrnl.exe
SYSTEM_THREAD_EXCEPTION_NOT_HANDLED

Any ideas or feedback on what the issue might be are greatly appreciated! Let me know if you need any more info from me.

EDIT:

Updated Minidump here.

EDIT:

Updated Minidump here. 12/6/22

EDIT:

Updated Minidump here. 12/7/22

 

[Lutfij]

the real trouble started late Friday night when I transferred my SSDs from my current build to the new build, and tried to reinstall windows on it
You should have the drive you intend to install the OS on as the only drive in your platform, then reinstall the OS(without connectivity to the www) then manually install all drivers in an elevated command, i.e, Right click installer>Run as Administrator, then power down and then plug each additional drive one after the other(making sure to power up and boot to OS after each drive addition.

BIOS version for your motherboard this point of time?

PSU	Corsair HX1200

How old is the PSU in your build?

plugged in the win 11 boot media I had created
Where did you source the installer for your OS?

 

[codelode]

the real trouble started late Friday night when I transferred my SSDs from my current build to the new build, and tried to reinstall windows on it
You should have the drive you intend to install the OS on as the only drive in your platform, then reinstall the OS(without connectivity to the www) then manually install all drivers in an elevated command, i.e, Right click installer>Run as Administrator, then power down and then plug each additional drive one after the other(making sure to power up and boot to OS after each drive addition.

BIOS version for your motherboard this point of time?

PSU	Corsair HX1200

How old is the PSU in your build?

plugged in the win 11 boot media I had created
Where did you source the installer for your OS?

You should have the drive you intend to install the OS
Thats a fair point, I should have thought of that but by that time I had done some hours of painful water routing, installing waterblocks, and installing the PSU and cablemod cables. Then came filling it and all that fun...

BIOS version for your motherboard this point of time?
My current BIOS version is F5, I initially tried the latest F6c but this caused near constant BSODs the second I left the BIOS so I gave up on that pretty fast. I was going to try all the versions going down from F6c until something worked and the first downgrade worked the best.

How old is the PSU in your build?
I bought it for this build, in October. The cables are the proper matching cables from cablemod with the 12vphwr for the 4090.

Where did you source the installer for your OS?
I tried both directly from Microsoft with the installation media, with their iso, and lastly with rufus. I think the version that ended up working for me right now is the rufus install.

 

[Bob.B]

Starting Friday, December 2nd, I started building my new PC with custom water-cooling. I thought the hardest part would be the loop and not murdering my PC with liquid, but the real trouble started late Friday night when I transferred my SSDs from my current build to the new build, and tried to reinstall windows on it.

Specs:

CPU	Ryzen 9 7950x
MOBO	Gigabyte x670e AORUS XTREME
GPU	Gigabyte RTX 4090 Gaming OC
RAM	2x16gb G.Skill Trizent Z5 NEO RGB
Boot Drive	Sabrent 2TB (SB-Rocket-2TB)
Second & Third Drive	Sabrent 2TB (SB-Rocket-2TB)
Fourth Drive	Sabrent ROCKET 4 Plus-G 4TB (SB-RKTG-4TB)
PSU	Corsair HX1200

After installing all the drives, I booted it up and was greeted with a low-res windows 11 install from my current build. This confirmed to me it was working well enough, so I shut it down then plugged in the win 11 boot media I had created in order to reformat the drive and start fresh.

When I was in the win 11 installation process, I chose to do a custom install, and deleted + formatted my previous C drive, then chose that drive as the drive to install to. The installer seemed happy with this for a moment, then I was hit with the first of many BSODs of the night. I don't recall exactly what the first BSOD was, but I'll list all of the BSODs I would encounter below.

Next, I tried to install again in various ways, all of them failing with BSODs, until eventually I also tried to boot into an Ubuntu live media drive, which got me some progress, save for some issues with nouveau that had to be dealt with. I got to the install process of ubuntu, then it hung for a bit, then hung for a long bit, and I gave up.

I would go to retry Windows 11 again, and this time after messing with BIOS settings around CMS and secure boot, I managed to get windows 11 fully installed, but I was stuck in a loop of doing the first 3 steps of the introduction (choose country, keyboard, language) where without fail in these three views I would BSOD again, or just crash to black screen, and be greeted with the first step again on next boot.

Thats where I ended this morning, at 4am. Next, I decided to force access to cmd by getting into recovery mode and tried to do some repair and scanning there with sfc, dism, and chkdsk and some other tool, only sfc found corrupt files and repaired them (after every bsod, it would find more corrupt files).

Finally, I caved to friend pressure and tried installing Windows 10. Windows 10 somehow went through the whole install process without fail, and I was able to get to desktop and install device drivers and programs like Steam etc. Then, when I was installing some windows updates, I got BSOD again. Also, I would get BSOD on the first launch of any game on steam, but not thereafter. I've encountered this crash with Skyrim SE and MW2/Warzone 2.0, and thats where I am now. As long as I don't install anything, or perform any updates, the machine seems stable. But when I start installing and updating things, crashes appear.

Temperatures, power levels, clock speeds etc all look perfectly within line for a waterloop pc (CPU, GPU on 2x 360mm rads), and I was able to get into a warzone match and run the game for ~30m without any issues, consistent framerate, etc.

Personally, I think the issue lies with RAM, and I've ordered some replacement test ram to try out for tomorrow, but I want to get feedback here as well. I have 3 minidump files here. The list of BSODs I remember I've experienced across this whole ordeal are as follows:

IRQL_NOT_LESS_OR_EQUAL
DPC_WATCHDOG_VIOLATION
ATTEMPTED_TO_WRITE_TO_READONLY_DIRECTORY
REFERENCE_BY_POINTER
KERNAL_SECURITY_CHECK_FAILURE
ntoskrnl.exe
SYSTEM_THREAD_EXCEPTION_NOT_HANDLED

Any ideas or feedback on what the issue might be are greatly appreciated! Let me know if you need any more info from me.

Just to get it out of the mix.

Put a copy of memtest86 on a flash stick.
Boot the flash stick and let it run.
No errors allowed.

 

[johnbl]

second bugcheck was a stack overflow in
PROCESS_NAME: Razer Synapse 3.exe

ERROR_CODE: (NTSTATUS) 0xc0000409 - The system detected an overrun of a stack-based buffer in this application. This overrun could potentially allow a malicious user to gain control of this application.

looked like a corrupted timer
----------
first bugcheck was in some human interface device attached to usb
bogus memory address used. (looks like a driver bug)

maybe update firmware to your mouse?

note: if you remove the device, you will want to go into control panel device manager. find the option to show hidden devices and delete the greyed out entries. windows just hides the usb devices and leaves the drivers hidden an running when the physical usb device is removed. (causes lots of confusion if you have a bad device driver)

 

[codelode]

Just to get it out of the mix.

Put a copy of memtest86 on a flash stick.
Boot the flash stick and let it run.
No errors allowed.

Going to run memtest overnight starting in a few hours here.

second bugcheck was a stack overflow in
PROCESS_NAME: Razer Synapse 3.exe

ERROR_CODE: (NTSTATUS) 0xc0000409 - The system detected an overrun of a stack-based buffer in this application. This overrun could potentially allow a malicious user to gain control of this application.

looked like a corrupted timer
----------
first bugcheck was in some human interface device attached to usb
bogus memory address used. (looks like a driver bug)

maybe update firmware to your mouse?

note: if you remove the device, you will want to go into control panel device manager. find the option to show hidden devices and delete the greyed out entries. windows just hides the usb devices and leaves the drivers hidden an running when the physical usb device is removed. (causes lots of confusion if you have a bad device driver)

Thanks for the info! I have had a few more bsods since then and will update the link in the OP, let me know if there is new info in them if you would.

 

[johnbl]

Going to run memtest overnight starting in a few hours here.



Thanks for the info! I have had a few more bsods since then and will update the link in the OP, let me know if there is new info in them if you would.

looks like mostly access violations. You will want to make sure memtest86 works as expected.
I would also be checking the voltages from your power supply. One of the errors hints at
error in a page table reserved bits being set and indicates a hardware issue. I would make sure you install the motherboard driver updates.


very new motherboard, 4800Mhz ram, new released CPU. Check for updates from motherboard vendor and directly from cpu vendor.

 

[codelode]

looks like mostly access violations. You will want to make sure memtest86 works as expected.
I would also be checking the voltages from your power supply. One of the errors hints at
error in a page table reserved bits being set and indicates a hardware issue. I would make sure you install the motherboard driver updates.


very new motherboard, 4800Mhz ram, new released CPU. Check for updates from motherboard vendor and directly from cpu vendor.

Are you thinking RAM hardware issues or PSU? Mobo would be quite painful to replace.

One thing to note is that my psu has a switch for single or multi rail, i believe it is on single rail mode. Think that matters?

Memtest is running now, has been for about 4h but i won't be able to check it on it for about 24h total since i had to leave home.

Think it is worth it to try the latest bios again? Currently on F5 but latest is 1 month after, F6c.

I did all my driver updates before starting memtest and had no issues installing them.

I'll post back here after i check memtest, and the new ram after that. Thanks for the help.

 

[johnbl]

Are you thinking RAM hardware issues or PSU? Mobo would be quite painful to replace.

One thing to note is that my psu has a switch for single or multi rail, i believe it is on single rail mode. Think that matters?

Memtest is running now, has been for about 4h but i won't be able to check it on it for about 24h total since i had to leave home.

Think it is worth it to try the latest bios again? Currently on F5 but latest is 1 month after, F6c.

I did all my driver updates before starting memtest and had no issues installing them.

I'll post back here after i check memtest, and the new ram after that. Thanks for the help.

I think you have a newly released cpu and motherboard and your BIOS needs to have some patches applied to tweak voltages to the cpu. These patches might not come out for months so you will want to apply the CPU chipset driver directly from the CPU vendor (AMD) this will load a special microcode patch to fix issues that AMD knows about and have not been picked up and patched into your motherboard BIOS yet.

So I think the problem in not in the RAM sticks or the PSU but the default settings in the BIOS that set the clock frequencies and voltages to the CPU pins for your particular CPU version. (ie in the RAM inside the cpu, voltages and clock rates are used to synchronize the movement of data thru the cpu, if a voltage is slightly wrong then frequency will be wrong and the data can be corrupted as it moves thru the levels of the CACHE RAM inside the cpu. )

I normally do no recommend people run the ryzenmaster drivers but in this case they would likely help override/fine tune settings in the BIOS. I would then remove the drivers after your motherboard vendor had done a bios update. (in a few months)

you might look to see what this driver is for:
AMD Ryzen™ 9 7950X Drivers & Support | AMD
came out 12/1/22

 

[codelode]

I think you have a newly released cpu and motherboard and your BIOS needs to have some patches applied to tweak voltages to the cpu. These patches might not come out for months so you will want to apply the CPU chipset driver directly from the CPU vendor (AMD) this will load a special microcode patch to fix issues that AMD knows about and have not been picked up and patched into your motherboard BIOS yet.

So I think the problem in not in the RAM sticks or the PSU but the default settings in the BIOS that set the clock frequencies and voltages to the CPU pins for your particular CPU version. (ie in the RAM inside the cpu, voltages and clock rates are used to synchronize the movement of data thru the cpu, if a voltage is slightly wrong then frequency will be wrong and the data can be corrupted as it moves thru the levels of the CACHE RAM inside the cpu. )

I normally do no recommend people run the ryzenmaster drivers but in this case they would likely help override/fine tune settings in the BIOS. I would then remove the drivers after your motherboard vendor had done a bios update. (in a few months)

you might look to see what this driver is for:
AMD Ryzen™ 9 7950X Drivers & Support | AMD
came out 12/1/22

Ill take a look at ryzenmaster for sure now.

Report back on memtest, I passed 48/48 tests no errors running for about 8.5h. I left it running when I left and got back after about 28h, I thought it would take a lot longer to test. So I guess thats good news and I dont need this replacement ram I received today.

Ill update after ryzen master.

 

[codelode]

I just played 3h of Hell Let Loose on max settings with no issues. I did BSOD once today, but after installing ryzen master and applying the default settings it's been (knock on wood) stable?

Last BSOD I got was 'KERNEL_SECURITY_CHECK_FAILURE'.

 

[codelode]

Leaving updated minidump in OP again.

 

[johnbl]

most current bugcheck: looked like it was inside the file system called from some file filter.
Looked like the call generated a error and the error was used as a valid address which caused the bugcheck. (address of -1)

you might consider removing this driver until your system is stable:
gdrv3.sys
I think it is a gigabyte overclock tool.


could be getting a stack overflow from audio drivers
ElgatoUsbAudio.sys Wed Dec 1 05:35:42 2021
ElgatoUsbAudio_mixer.sys Wed Dec 1 05:38:20 2021
ElgatoUsbAudioks.sys Wed Dec 1 05:37:00 2021
ElgatoVirtUsbAudioEmu.sys Tue Sep 13 04:29:53 2022

 

[codelode]

I have conducted a full system reinstall again to try to debug the issue further. Here are my methods and findings:

1. While PC is off and switched off at PSU, uninstall all M.2 drives
2. Install single M.2 drive in slot 1
3. Boot into windows installation media
   1. Use SHIFT+F10 to access privileged CMD
   2. Enter diskpart and clean drive, convert to GPT, create primary partition
4. Install windows on clean drive
5. Once in windows, install the following:
   1. All AMD chipset drivers from AMDs site for X670E
   2. Nvidia GeForce Experience & Latest Driver
   3. Run windows update and install latest updates
   4. Install all C++ redistributable packs from 2005 to latest, both x86 and x64
   5. Install Java runtimes 8, 11, 17 both x86 and x64
   6. Install .NET 5, .NET 6, .NET 7 runtimes
   7. Install 7zip

At this point, things felt pretty stable, and I was ready to move on. There were no BSODs during all of this if you can believe it or not.

Next, I installed the 2nd M.2 drive, and things started getting interesting. The first drive I tried showed valid in windows, and I tried to install a game to it. Got a fairly instant BSOD from this, and on next boot the drive would not allow anything written to it. I looked in diskpart and the size was reported as 2mb. I tried to clean it, and diskpart froze, then the drive disappeared from diskpart and windows entirely.

I booted into BIOS and tried to see if the device was there, and sure enough it was. This was a brand new drive and it had the functionality of running self-test in BIOS, so I ran the short self test and this came back as PASS. I then ran the extended self-test and this came back as an instant FAIL. Then the short test would also FAIL.

I uninstalled this drive, and tried another one. Booted into windows again, installed a game to it, and everything was fine. So I tried a third drive, and the key thing to note here is that on this mobo, the 3rd and 4th M.2 slots share bandwidth with the PCIE x16 slot, causing it to run in x8 instead.

When I booted next with the 3rd stick in, there was no video output, but I did boot to windows, and could hear various windows sounds going on as normal, including discord notifications and device connect/disconnect sounds.

I went rebooted to bios and found that the mobo no longer recognized my GPU as being installed.

I tried powering down and removing the 3rd drive, and now the GPU was back. I confirmed this one more time with a drive in the 4th slot, to the same result. I looked around the bios and made sure the split for the x16 and m.2 drives was in the correct spot using the integrated graphics of the CPU on the mobo socket, and tried the setting for 8x 8x, and 8x 4x 4x, neither of which would bring my GPU back into existence.

When I shutdown and removed the 3rd drive, the GPU was back and I could enter windows normally.

Next, I tried to play the game on the 2nd drive, and got a BSOD.

I decided to remove everything from this drive and install a game on the first drive. No problems, it ran phenomenally.

So I decided to try stress test for CPU/GPU to mark those as not the problem. I ran AIDA64, 3D Mark Time Spy Extreme, and UserBenchmark, all of which reported that my hardware was performing at the top end of the spectrum, so I suspect there is no issue with the GPU/CPU/RAM at this time.

I have updated the minidump again in the OP, as I just received a BSOD of MULTIPLE_IRP_COMPLETE_REQUESTS.

I have also sent in a support ticket to Gigabyte trying to explain all the issues I've had and the errors I've run into, hoping they can tell me to hold out for the next bios or something that prevents me having to rebuild this thing.

If you could again look over my minidump files I would greatly appreciate it. You're being extremely helpful!

 

[johnbl]

I have conducted a full system reinstall again to try to debug the issue further. Here are my methods and findings:

1. While PC is off and switched off at PSU, uninstall all M.2 drives
2. Install single M.2 drive in slot 1
3. Boot into windows installation media
   1. Use SHIFT+F10 to access privileged CMD
   2. Enter diskpart and clean drive, convert to GPT, create primary partition
4. Install windows on clean drive
5. Once in windows, install the following:
   1. All AMD chipset drivers from AMDs site for X670E
   2. Nvidia GeForce Experience & Latest Driver
   3. Run windows update and install latest updates
   4. Install all C++ redistributable packs from 2005 to latest, both x86 and x64
   5. Install Java runtimes 8, 11, 17 both x86 and x64
   6. Install .NET 5, .NET 6, .NET 7 runtimes
   7. Install 7zip

At this point, things felt pretty stable, and I was ready to move on. There were no BSODs during all of this if you can believe it or not.

Next, I installed the 2nd M.2 drive, and things started getting interesting. The first drive I tried showed valid in windows, and I tried to install a game to it. Got a fairly instant BSOD from this, and on next boot the drive would not allow anything written to it. I looked in diskpart and the size was reported as 2mb. I tried to clean it, and diskpart froze, then the drive disappeared from diskpart and windows entirely.

I booted into BIOS and tried to see if the device was there, and sure enough it was. This was a brand new drive and it had the functionality of running self-test in BIOS, so I ran the short self test and this came back as PASS. I then ran the extended self-test and this came back as an instant FAIL. Then the short test would also FAIL.

I uninstalled this drive, and tried another one. Booted into windows again, installed a game to it, and everything was fine. So I tried a third drive, and the key thing to note here is that on this mobo, the 3rd and 4th M.2 slots share bandwidth with the PCIE x16 slot, causing it to run in x8 instead.

When I booted next with the 3rd stick in, there was no video output, but I did boot to windows, and could hear various windows sounds going on as normal, including discord notifications and device connect/disconnect sounds.

I went rebooted to bios and found that the mobo no longer recognized my GPU as being installed.

I tried powering down and removing the 3rd drive, and now the GPU was back. I confirmed this one more time with a drive in the 4th slot, to the same result. I looked around the bios and made sure the split for the x16 and m.2 drives was in the correct spot using the integrated graphics of the CPU on the mobo socket, and tried the setting for 8x 8x, and 8x 4x 4x, neither of which would bring my GPU back into existence.

When I shutdown and removed the 3rd drive, the GPU was back and I could enter windows normally.

Next, I tried to play the game on the 2nd drive, and got a BSOD.

I decided to remove everything from this drive and install a game on the first drive. No problems, it ran phenomenally.

So I decided to try stress test for CPU/GPU to mark those as not the problem. I ran AIDA64, 3D Mark Time Spy Extreme, and UserBenchmark, all of which reported that my hardware was performing at the top end of the spectrum, so I suspect there is no issue with the GPU/CPU/RAM at this time.

I have updated the minidump again in the OP, as I just received a BSOD of MULTIPLE_IRP_COMPLETE_REQUESTS.

I have also sent in a support ticket to Gigabyte trying to explain all the issues I've had and the errors I've run into, hoping they can tell me to hold out for the next bios or something that prevents me having to rebuild this thing.

If you could again look over my minidump files I would greatly appreciate it. You're being extremely helpful!

I looked at the bugcheck
MULTIPLE_IRP_COMPLETE_REQUESTS.
the function making the call was for a network driver.
you have these two adaptors installed:
\SystemRoot\System32\DriverStore\FileRepository\netwtw6e.inf_amd64_1000547336ad8b60\Netwtw10.sys Tue Jul 19 14:15:17 2022 (62D71EE5)
\SystemRoot\System32\drivers\aqnic650.sys Fri Oct 22 01:42:21 2021 (6172796D)

there were two status returned for one network packet. one indicating it was completed, the second indicated it was pending. So it can be a bug in one driver returning both indicators or both drivers are processing the same packet (which would be a bug in one of the drivers)

so, first you want to try to update both drivers to the most current version.
(my guess is the error will be in the aqnic650.sys driver)

second, you should go into windows control panel, device manager and find the two network devices, right mouse click to bring up properties. find the power management tab and tell windows not to turn off the device to save power. This could eliminate the case where one driver returns two status messages and handles it incorrectly. IE something like the device is asleep, returns pending then wakes and processes the packet and later ends up processing it again. (one adaptor processing one packet twice)

you can take turns disabling the network hardware to eliminate the case where both networks adapters are processing the same packet.

start with the update first. Look at the motherboard vendor website first. If there is not a update then look at chip vendors website.

most likely this update will fix this problem: new driver dated 2022/08/09

X670E AORUS XTREME (rev. 1.0) Support | Motherboard - GIGABYTE U.S.A.

but the wired lan, the wireless lan and the bluetooth drivers all have updates. It could be any network driver as the cause of this bugcheck.

 

[codelode]

I looked at the bugcheck
MULTIPLE_IRP_COMPLETE_REQUESTS.
the function making the call was for a network driver.
you have these two adaptors installed:
\SystemRoot\System32\DriverStore\FileRepository\netwtw6e.inf_amd64_1000547336ad8b60\Netwtw10.sys Tue Jul 19 14:15:17 2022 (62D71EE5)
\SystemRoot\System32\drivers\aqnic650.sys Fri Oct 22 01:42:21 2021 (6172796D)

there were two status returned for one network packet. one indicating it was completed, the second indicated it was pending. So it can be a bug in one driver returning both indicators or both drivers are processing the same packet (which would be a bug in one of the drivers)

so, first you want to try to update both drivers to the most current version.
(my guess is the error will be in the aqnic650.sys driver)

second, you should go into windows control panel, device manager and find the two network devices, right mouse click to bring up properties. find the power management tab and tell windows not to turn off the device to save power. This could eliminate the case where one driver returns two status messages and handles it incorrectly. IE something like the device is asleep, returns pending then wakes and processes the packet and later ends up processing it again. (one adaptor processing one packet twice)

you can take turns disabling the network hardware to eliminate the case where both networks adapters are processing the same packet.

start with the update first. Look at the motherboard vendor website first. If there is not a update then look at chip vendors website.

most likely this update will fix this problem: new driver dated 2022/08/09

X670E AORUS XTREME (rev. 1.0) Support | Motherboard - GIGABYTE U.S.A.

but the wired lan, the wireless lan and the bluetooth drivers all have updates. It could be any network driver as the cause of this bugcheck.

Currently running with latest win10 2h22 update, all drivers from mobo site and GCC installed except things that I shouldnt need that I noticed in some of the error messages like Gigabyte Monitor RGB controls (I dont have a gigabyte monitor) and VGA libs, DRAM libs that are for controlling RAM/GPU OC. Also installed all the chipset drivers from AMD directly and the Adrenaline software, and Ryzen Master.

Seems to be running stable. I downgraded the BIOS to F4, fewer issues here. For one, my original RAM runs at EXPO speeds without issues.

I also downloaded a firmware update to my 4090 that was hidden away behind specific search terms, and running it detected that my 4090 was eligible for a firmware update to prevent incompatibilities with my mobo. This brought even greater stability and I felt more confident in the PC overall.

The next thing I need to try is installing more than 1 M.2 drive. This is usually where the major problems arise.

Thanks for the advice on drivers again, getting those sorted is bringing more stability. I am still cautious to say it is 'stable' though. I feel like I'm wandering through a minefield and at any moment I could get into a BSOD loop of some kind again.

 

[codelode]

So I installed a second M.2, and things were fine. I could install games and run them just fine, and there were no issues. So, after a couple of hours playing games and generally enjoying my PC at the most stable its been since I got it, I decided I would risk it and install another M.2 drive.

This brought back all the issues, all at once.

First of all, i couldn't boot with video, not into bios or windows. Windows would load, and bios would load, but it would be an empty black screen and I could hear desktop notifications and such. Obviously unusable.

I switch to integrated graphics and boot to bios, and the PC is not detecting that my GPU is installed when there is a 3rd M.2 installed.

I go through the BIOS settings for like half an hour and finally stumble over some settings related to PCIe Bifurcation and try switching it from 'AUTO' to 8x8x. Still no GPU, so I try 8x4x4x, and now I have GPU!

I restart, plug my monitor into my GPU, and boot to windows. All seems fine.

I go to disk manager and see that my 3rd drive is showing up as unallocated and try to create simple volume on it.

PC freezes, then I get BSOD. The start of the loop once more.

After this BSOD, whenever I boot into windows, I get a new BSOD after a few moments, and my windows exhibits critical errors such as when I right click CMD or Powershell to open as admin, there is no option to do so. I only get 'pin to taskbar' and 'remove from list'.

I try opening CMD anyway, and I get some very weird cryptic windows error ive never seen before on any version, then BSOD.

On next boot, I try opening powershell, and it says that an administrator account on my PC has limited access to this program with a red warning box.

I had created restore points specifically for this moment, and shutdown the PC, turn of PSU, and take out the 3rd M.2, and revert the BIOS to before it was installed.

I boot into windows, and open powershell as admin, and try to run SFC and it comes up with no errors. DISM also comes up with no errors on scanhealth or restorehealth. I get BSOD again.

From then on, whenever I got into windows, I would get BSOD after just a few minutes, and random programs would constantly pop open saying they could not run properly, such as AMD Adrenaline, Geforce Experience, and other critical apps.


At this point, I think it is very safe to say the issue is the motherboard, but I am open to other suggestions. Unfortunately, I dont have access to the minidumps from this experience, since windows was entirely unstable and edge would crash or report no internet after just opening a new webpage. I am probably going forward with an RMA, though at this point I have the urge to look for other motherboard vendors that do no split PCIe lanes on M.2 use, I would rather lose SATA or other PCIE x16 slots than lose half my first PCIE x16 slot for another M.2.

I can probably get the system stable again by going through the entire process of cleaning the drives, installing win10, and installing all the drivers, but its tiring.

 

[drivinfast247]

So I installed a second M.2, and things were fine. I could install games and run them just fine, and there were no issues. So, after a couple of hours playing games and generally enjoying my PC at the most stable its been since I got it, I decided I would risk it and install another M.2 drive.

This brought back all the issues, all at once.

First of all, i couldn't boot with video, not into bios or windows. Windows would load, and bios would load, but it would be an empty black screen and I could hear desktop notifications and such. Obviously unusable.

I switch to integrated graphics and boot to bios, and the PC is not detecting that my GPU is installed when there is a 3rd M.2 installed.

I go through the BIOS settings for like half an hour and finally stumble over some settings related to PCIe Bifurcation and try switching it from 'AUTO' to 8x8x. Still no GPU, so I try 8x4x4x, and now I have GPU!

I restart, plug my monitor into my GPU, and boot to windows. All seems fine.

I go to disk manager and see that my 3rd drive is showing up as unallocated and try to create simple volume on it.

PC freezes, then I get BSOD. The start of the loop once more.

After this BSOD, whenever I boot into windows, I get a new BSOD after a few moments, and my windows exhibits critical errors such as when I right click CMD or Powershell to open as admin, there is no option to do so. I only get 'pin to taskbar' and 'remove from list'.

I try opening CMD anyway, and I get some very weird cryptic windows error ive never seen before on any version, then BSOD.

On next boot, I try opening powershell, and it says that an administrator account on my PC has limited access to this program with a red warning box.

I had created restore points specifically for this moment, and shutdown the PC, turn of PSU, and take out the 3rd M.2, and revert the BIOS to before it was installed.

I boot into windows, and open powershell as admin, and try to run SFC and it comes up with no errors. DISM also comes up with no errors on scanhealth or restorehealth. I get BSOD again.

From then on, whenever I got into windows, I would get BSOD after just a few minutes, and random programs would constantly pop open saying they could not run properly, such as AMD Adrenaline, Geforce Experience, and other critical apps.


At this point, I think it is very safe to say the issue is the motherboard, but I am open to other suggestions. Unfortunately, I dont have access to the minidumps from this experience, since windows was entirely unstable and edge would crash or report no internet after just opening a new webpage. I am probably going forward with an RMA, though at this point I have the urge to look for other motherboard vendors that do no split PCIe lanes on M.2 use, I would rather lose SATA or other PCIE x16 slots than lose half my first PCIE x16 slot for another M.2.

I can probably get the system stable again by going through the entire process of cleaning the drives, installing win10, and installing all the drivers, but its tiring.

Just a quick question. Is it the same m.2 drive that causes the issue or is it when that particular m.2 port gets used?

 

[codelode]

Just a quick question. Is it the same m.2 drive that causes the issue or is it when that particular m.2 port gets used?

Well, its a good question, and I asked myself that today as well.

I have 4x M.2 drives, 3x of them I know for a fact work perfectly fine, as I have since put 2x back inside my old PC and am typing this response on that PC right now. They report as healthy drives with between 98%-100% of life remaining. They're all identical models - SB-ROCKET-2TB - and I can put files on them and write to them all and read from all just fine.

Now, one thing that is for sure, is that when only 1 of these drives is installed in the first slot of the X670E motherboard, I can install windows and run everything just fine. When 2 of these are installed, it runs fine. I don't recall what happened with 3x, and since I have relegated one of the drives to my old PC as a backup so I'm not without a PC, I am not comfortable risking having to reinstall everything from square one on this PC again.

Another thing for sure, is that whenever the really bad issues happened, the other M.2 drive, the SB-RKTG-4TB, was installed. However, I have had minimal issues with it installed in the 2nd M.2 slot, since I do know for sure one of these resets I had it installed in the 2nd slot, installed Warzone 2.0 on it, and played a few matches. I also installed Hell Let Loose on it, and played for 3h without problems. But, I did ultimately run into BSOD issues on those 'runs' at some point in the day.

So, I can see that it is possible just the SB-RKTG-4TB has problems, but I wouldn't be able to know for sure unless I tried:

1. One good SSD in slot 1, 2, and 3. (Cannot do because need 1 as backup for working PC)
2. One good SSD in slot 1, and 3. (Can try)
3. One good SSD in slot 1, and 4. (Can try)

And I will probably try those solutions out tomorrow and see if it helps.

Another thing, I've decided that either I will RMA the motherboard, or get a refund through my credit card purchase protection, depending on if I want to bother with another of the same board.

However, if the problem is isolated to just the SB-RKTG-4TB, then I will RMA that drive instead, and be much happier.

 

[codelode]

Just a quick question. Is it the same m.2 drive that causes the issue or is it when that particular m.2 port gets used?

I put the new drive that we suspect is causing the issue into my old PC that I know has no issues, and I think we found the problem.

The new drive first of all was found in disk manager as a RAW data drive, so I deleted the volume and tried to create a simple volume without quick format, so that I could see if issues popped up in formatting.

The drive failed to format, and then disappeared from my windows list of drives. Then, in diskpart, the drive was still there, but after selecting it and running 'clean' command on it, diskpart told me that I have selected a drive that doesn't exist and it couldn't complete the task.

If this is truly the problem, then I should be able to also put one of the 'good' drives into those 3rd and 4th slots of my new pc and see it run properly, after I get windows up and running on it again.

Thanks for the idea to check on the drive itself! I am confused how the problem drive was able to run games for at least a few hours without issues, but maybe that was a fluke.

 

[drivinfast247]

I put the new drive that we suspect is causing the issue into my old PC that I know has no issues, and I think we found the problem.

The new drive first of all was found in disk manager as a RAW data drive, so I deleted the volume and tried to create a simple volume without quick format, so that I could see if issues popped up in formatting.

The drive failed to format, and then disappeared from my windows list of drives. Then, in diskpart, the drive was still there, but after selecting it and running 'clean' command on it, diskpart told me that I have selected a drive that doesn't exist and it couldn't complete the task.

If this is truly the problem, then I should be able to also put one of the 'good' drives into those 3rd and 4th slots of my new pc and see it run properly, after I get windows up and running on it again.

Thanks for the idea to check on the drive itself! I am confused how the problem drive was able to run games for at least a few hours without issues, but maybe that was a fluke.

Hopefully that was the issue the entire time.

 

[codelode]

Hopefully that was the issue the entire time.

I installed a good drive in the 3rd M.2 slot and all is fine so far. In fact, for the first time, my mobo did not fail to show my drive or gpu as installed, and while my gpu is running PCIE 5.0 x8 instead of x16 as expected, everything is fine right now.

This is great. ill update if anything changes.

 

[codelode]

Update:

I restarted my PC, and lost video output. I could get video output from iGPU and boot to windows there, but obviously I want my GPU to work.

So I tried updating bios to F6c, this broke things horribly.

I downgraded to F5, and its working now. I even get GPU output again.

But, I think to get GPU output, I have to do something really odd, which is I have to have a display plugged into my iGPU mobo slot at boot, otherwise it doesn't decide to power on my GPU? Not sure. But, we'll see how this goes now. F6c caused a lot of BSOD. Repaired with SFC and DISM, no problems doing that.

We will see how F5 goes.

 

[codelode]

Had a BSOD on this run: Link here

BIOS is F5, 2 M.2 Installed. Maybe this is just a driver issue though for this BSOD? Let me know if someone can analyze.

 

[drivinfast247]

Get rid of that Gigablow motherboard and grab one from another manufacturer.