[SOLVED] Random memory related BSOD

[bloodymeli]

Hi,

I have recently built a new rig, which is composed of parts of my old rigs:
Samsung SSD 850 + Samsung SSD 860
2 X 8 gb DDR4 2133mhz (Corsair Vengeance)

With a new MB + CPU:
Ryzen 7 3700X
Asus B450 Tuf Plus
MSI rtx 2070

I keep getting random memory-related BSODs (my guess!) such as:
IRQL_NOT_LESS_OR_EQUAL
MEMORY MANAGMENT
kernel security check failure
etc.

They are completely random. The computer could be idle and the crash would just occur out of nowhere.
In fact, they were also present during Windows setup. I was only able to complete the setup after I changed the SSD connection from the SATA 5/6 connection to one of the other 4 slots on the motherboard.

I have run memtest86+ and windows memory diagnostic and did not detect any error. I also use Linux, and I didn't detect any errors using it as well either (although I didn't use it for a long duration). I'm using the latest bios & drivers.

Any idea what should I do? Any additional tests? My guess is that it is a problem with the bios / drivers / windows system, but not sure how to resolve it. Quite frustrating, the MB + CPU are completely new.

 

[PC Tailor]

Hello my friend, please see the BSOD thread in my signature below for enabling and uploading the minidump files for us to analyse. If it is software related, we can hopefully identify the point of cause through them.

• IRQL and Mem Mgmt can often be driver related, but not always.
• How long did you run the memtest for?
• Does the issue still occur in safe mode?
• Are you using a driver updater program to update your drivers?
• What is the PSU make and model?
• Are your RAM modules from the same physical pack?
• Are you doing anything in particular when the issues occur?
• Are you using a legitimate version of Windows downloaded using the Media Creation tool at Microsoft?

 

[bloodymeli]

Thanks.
Here Are some answers:

• How long did you run the memtest for?

Not too long. About 1 hour (1 pass). I'll do a night scan tonight.

• Does the issue still occur in safe mode?

I haven't tested. I'll reboot to safe mode and try.

• Are you using a driver updater program to update your drivers?

Not really. Direct download & install, except perhaps for NVIDIA experience

• What is the PSU make and model?

Corsair CX650M

• Are your RAM modules from the same physical pack?
• Are you doing anything in particular when the issues occur?
• Are you using a legitimate version of Windows downloaded using the Media Creation tool at Microsoft?

Yes X 3.
I'll configure Windows according to the link and upload the next minidumps.

 

[bloodymeli]

Hi,

Some mini dumps are in here.

Thanks!

 

[PC Tailor]

I have run the dump files and you can see the full reports here:

Dump1: https://pste.eu/p/3Gmc.html (MEMORY_MANAGEMENT)
Dump2: https://pste.eu/p/Qflr.html (SYSTEM_SERVICE_EXCEPTION)
Dump3: https://pste.eu/p/fv9p.html (IRQL_NOT_LESS_OR_EQUAL)

Being as there are multiple dump files, you can see the explanation of each one in each report as opposed to on this specific post.

Some things to consider:
I would highly advise you to view the full report above, as this will contain much more detail as to the bugcheck and modules running at the time.

• Each of the dump files tend to blame hardware but a couple do state the bucket as WIN8_DRIVER_FAULT. Which can mean either frankly.
• Being as each of the dump files seem to be memory corruption based, I would absolutely run memtest first, however 1 pass is not enough, it should be left for at least 4 passes, Ideally for as long as possible. The test has to be fully completed.
• There is nothing obvious in the stack text apart from some page faults and maybe some store interaction, but from a software side, I would consider these:

mudaxp.sys

02/06/2015

C-MEDIA

C-Media Audio Interface Driver

umpusbvista.sys

07/09/2016

Texas Instruments

USB 2 Serial Driver

dbx.sys

11/02/2019

DBX

DBX Professional Audio driver

AMDPCIDev.sys

11/09/2019

AMD

AMD Ryzen OC Utility

I would update where possible and also consider disabling the Ryzen OC utility to be sure.

I will await you results from memtest and Safe mode before carrying on, but this at least gives us a starting point.

 

[bloodymeli]

Hi,

I have completed four cycles of memtest and no errors were found.
I actually don't have AMD Ryzen OC utility installed. I guess this is just a driver that comes along with the chipset drivers. Should I install it?
Also, I've uploaded 5 additional minidumps to the same folder.

I'm a bit puzzled of what to do next.
If the memtest completed successfully, does it mean there is no physical problem in the RAM modules? likewise, for the BIOS/MB? I can try to downgrade the bios version, but I'm not sure if will be useful as the memtest was completed successfully.

What are you suggestions?

 

[PC Tailor]

I have just run the last one due to the quick succession of the stop errors, and just to see if it gives any more insight:

Have you verified if it still occurs in safe mode?
Have you previously clean installed windows to see if the issue persisted?

I have run the dump file and you can see the full report here: https://pste.eu/p/TAqQ.html

Summary of findings:

BugCheck 1A
Probably caused by : memory_corruption ( ONE_BIT )

Bugcheck Description:
MEMORY_MANAGEMENT
"This indicates that a severe memory management error occurred. Parameter 1 identifies the exact violation. For full details of each P1:

https://docs.microsoft.com/en-us/windows-hardware/drivers/debugger/bug-check-0x1a--memory-management"

About your bugcheck:
"Running the Windows Memory Diagnostic tool or memtest could be useful as well to exclude any kind of problem affecting the physical memory modules.

This can be driver or hardware based. If you are encountering other hardware issues, you may have to look at hardware diagnosis.

• Confirm that hardware is compatible with your system.
• View the third party modules of the report to view which drivers may be misbehaving."

Some things to consider:

• Have you checked all the drivers I stated in the previous post?
• Have you run HD Sentinel or Seagate tools to check if the storage drives are in OK health?

If the memtest completed successfully, does it mean there is no physical problem in the RAM modules?

No, no software can ever be 100% guaranteed for hardware problems, they can only do their best to try and pull out faults, but there are plenty of times where software tests can pass, when the hardware is still faulty.

Now memory corruption does not directly mean RAM, as memory is both part of RAM and Storage (virtual memory).

 

[bloodymeli]

The problem occurs in safe mode.
The issue occurs with a clean install of windows. It also occurs during the installation of windows.
It also occurs when I disconnect the hard drives, and it occurs when I install windows on two different SSDs.
All the drivers seemed legit.
I'm now running a fresh setup with an older BIOS version.
The thing that puzzles me is that memtest passed and linux works. Had I known this is a memory or a hardware problem, I would just go and buy a replacement. But I hate to buy a new RAM just to get to the same point.

I have just run the last one due to the quick succession of the stop errors, and just to see if it gives any more insight:

Have you verified if it still occurs in safe mode?
Have you previously clean installed windows to see if the issue persisted?

I have run the dump file and you can see the full report here: https://pste.eu/p/TAqQ.html

Summary of findings:


Some things to consider:

• Have you checked all the drivers I stated in the previous post?
• Have you run HD Sentinel or Seagate tools to check if the storage drives are in OK health?

No, no software can ever be 100% guaranteed for hardware problems, they can only do their best to try and pull out faults, but there are plenty of times where software tests can pass, when the hardware is still faulty.

Now memory corruption does not directly mean RAM, as memory is both part of RAM and Storage (virtual memory).

 

[PC Tailor]

Well you can return the RAM under warranty usually. I'd still be tempted to RMA them and get a replacement if all else fails.
I would be tempted as a last resort to remove NVIDIA drivers and roll back to a previous version as the latest has caused all manner of issues.

The fact that it occurs in safe mode and a clean install tells me it's hardware or a seriously coincidental Windows conflict with a hardware driver.
Linux is a good test, but also Linux is much more lightweight.

Does the BSOD occur in any circumstance? (i.e. not pattern to when it happens).
When you were in safe mode, did you disable networking?