BSOD - help troubleshooting my system

[Marvin_18]

Hello all,

First and foremost, whilst I have some basic understanding of computers, I am in no way, shape or forum knowledgeable so please forgive my stupidity!

I have a Dell Inspiron that I use for work (mainly office programmes). I've had this system for years. About a month ago it started hanging, randomly rebooting or throwing up BSOD's with varying codes at any time between 5 seconds and 1 hour of powering on. I'm having difficulty figuring out what's wrong and was wondering if anyone could help me. Specs are:

AMD Phenom II quad core something or other.
Nivea GeForce 440 GT
4 Gb (4x1gb) Samsung DDR-3 memory
Windows 10

It's a stock machine that hasn't been tampered with or changed in any way over the years.

Since the errors started I have:

1. Reinstalled Windows from scratch, but the errors simply returned.
2. Updated all drivers
3. Booted in safe mode, removed graphics card drivers and reinstalled.
4. Removed GPU and used on-board graphics to see if it was the GPU causing the issue. It wasn't.
5. Used Memtest86 to test memory overnight. Came back and had 16k errors. Have since tested each RAM module individually with no errors. I am currently using one stick on each DIMM slot to see if it's an issue with the MB.

I'm struggling on where to go next. If anyone has any suggestions then please let me know.

Thanks!

 

[notlim981]

If the computer has never been touched inside as you stated above, it's most likely running on AMD's stock cooler and the Phenom is a really hot processor. Have you been following the temps? It could be the cause of your problems. In case you need, you can install HW Monitor by clicking this link: http://www.cpuid.com/softwares/hwmonitor.html

 

[Obakasama]

Would that be good for checking temperatures of the Phenom? (I've gotten used to replying with AMD Overdrive and thermal margin readings...)

How many passes of Memtest86 did it make? With Memtest86+ it's suggested 8 passes are required for something close to certainty. Insofar it initially came back with that many errors despite the later tests, it still sounds dodgy.

Have you noted what blue screen errors you were getting? These could help as well.

 

[Marvin_18]

Hi notlim981,

The computer has been so unstable that I haven't managed the few minutes required to install and test using the software you recommended. I've just reinstalled Windows again in the hope hat I can get 10 minutes of stability to find out. In the meantime I'e removed the case side cover and used a desktop fan to prove cool air. It's still crashing. I'll keep you updated once this fresh install has finished.

Hi obakasama,

Me test finished 4 passes on each memory stick when tested individually. I then used one stick in each of the 4 DIMM slots for four passes each. I know it's not much and I suspect that memory may be the issue. I'll run further tests on each individual stick today and give each one 6 passes before testing them as a group again.

As for error messages these are the ones from the last 8 hours of repairing:

System_service_exception (win32kbase.sys)
System_thread_exception_not_handled
System_exit_owned_mutex (Wof.sys)
Kernel_security_check_failure

There have been many, many more since these issues started.

Thanks.

 

[Paul NZ]

Upload the small dmp files to onedrive then post the link

 

[Marvin_18]

Hi Paul,

As I've just reinstalled Windows, I can't get he dump files. If/when it happens again, I'll collect them.

Managed to run HWMonitor. all 4 core of CPU running at 33 degrees. Going to run blend95 to see what it's like under load. Will report back shortly.

 

[Paul NZ]

k

 

[Marvin_18]

CPU temp running Prime95 is 63 degrees. Is this okay?

I'm going to let this run for a few hours whilst I pop out. I'm assuming that if any errors occur then it will log them for em?

Thanks,

 

[Marvin_18]

Full/clean reinstall of Windows 10 and the system has been stable for a few hours (the longest it's been in a while). Max CPU temp of 63 degrees after 3 hours of Prime95 with no errors. Managed to update all drivers and complete a couple of reboots with no problems.

What should I do next, as I still haven't found what caused the issue in the first place? I'll continue running prime but don't know what to do next.

Any advice welcome.

 

[Obakasama]

A clean install of Windows should rid the system of any software errors it may have had. (If errors continue then that's a worry.)

If it's working now I don't think it's something to worry about now. Personally I'd still keep an eye on the RAM as memtest may have flagged a potential hardware issue there.