Question Need some guidance figuring out what's causing BSoDs

[moshan]

Hello,

I've recently build a new PC and I am struggling a bit with some crashes.

Specs:
GIGABYTE B650 GAMING X AX
AMD Ryzen 5 7600X 4.7GHz box + cooler NH-U12S redux
Sapphire Radeon RX 6900 XT TOXIC Limited Edition
Corsair RM1000e, 80+ Gold, 1000W
G.Skill Trident Z5 Neo RGB 32GB DDR5 6000MHz CL30 Dual Channel Kit (F5-6000J3038F16GX2-TZ5NR)
SSD Kingston NV2 2TB PCI Express 4.0 x4 M.2 2280
+ SSD ADATA Ultimate SU650 480GB SATA-III 2.5 (that's just connected to the PC, nothing stored there)

OS: Windows 10 Pro (current version 10.0.19045 Build 19045)

Unfortunately I got some random BSoDs during it's short usage (I have the system from the start of March and the first BSoD hit a few days after a fresh install). I have uploaded the mini dumps accumulated in the past months here: https://drive.google.com/file/d/1nDQGwCI8-r6FrR-S-AkR4boAW-PzoKYJ/view?usp=sharing

I need some guidance on figuring out what's the cause of all those ntoskrnl.exe crashes. I researched online the failure codes I got and my understanding is this is just the Windows component that crashed, but I lack the technical knowledge to extract some valuable information out of the dumps to understand the root cause.

So far I've ran a memory test (the windows one) which passed with no error. I have AMD EXPO enabled in BIOS (and I run with the first revision of the BIOS - F1), but I am doubtful this is related to a memory problem.

Can someone please take a look and let me know if you can point me into the right direction with troubleshooting this?

Thanks!

 

[Bob.B]

Hello,

I've recently build a new PC and I am struggling a bit with some crashes.

Specs:
GIGABYTE B650 GAMING X AX
AMD Ryzen 5 7600X 4.7GHz box + cooler NH-U12S redux
Sapphire Radeon RX 6900 XT TOXIC Limited Edition
Corsair RM1000e, 80+ Gold, 1000W
G.Skill Trident Z5 Neo RGB 32GB DDR5 6000MHz CL30 Dual Channel Kit (F5-6000J3038F16GX2-TZ5NR)
SSD Kingston NV2 2TB PCI Express 4.0 x4 M.2 2280
+ SSD ADATA Ultimate SU650 480GB SATA-III 2.5 (that's just connected to the PC, nothing stored there)

OS: Windows 10 Pro (current version 10.0.19045 Build 19045)

Unfortunately I got some random BSoDs during it's short usage (I have the system from the start of March and the first BSoD hit a few days after a fresh install). I have uploaded the mini dumps accumulated in the past months here: https://drive.google.com/file/d/1nDQGwCI8-r6FrR-S-AkR4boAW-PzoKYJ/view?usp=sharing

I need some guidance on figuring out what's the cause of all those ntoskrnl.exe crashes. I researched online the failure codes I got and my understanding is this is just the Windows component that crashed, but I lack the technical knowledge to extract some valuable information out of the dumps to understand the root cause.

So far I've ran a memory test (the windows one) which passed with no error. I have AMD EXPO enabled in BIOS (and I run with the first revision of the BIOS - F1), but I am doubtful this is related to a memory problem.

Can someone please take a look and let me know if you can point me into the right direction with troubleshooting this?

Thanks!

You might want to update the bios just to get it out of the mix also the chipset driver if that's old.

 

[ubuysa]

As above.

With AMD processors the BIOS also contains AMD specific AGESA microcode and you generally want to keep that updated. There are three BIOS versions dated 2023 for your motherboard, one of them (F5 - 30th March 2023) contains AGESA updated code. I would get on to the latest BIOS.

There are also chipset driver updates dated 31st March 2023, you certainly want to install those.

From the dumps, two of the BSODs happened during graphics operations, one happened during a networking operation, and one is a buffer overrun but with no third-party driver indicated - that suggests that bad RAM might be the cause. Bad RAM could easily be the cause of all of these BSODs.

The Windows memory test is pretty poor but it's important that we test your RAM to be sure it's not the root cause. Download Memtest86 and use the extracted tool to make a bootable USB drive containing Memtest86 (the USB drive can be less an 1GB in size). Then boot that USB drive and Memtest86 will start running.

If you need help creating the bootable USB drive see here: https://www.memtest86.com/tech_creating-window.html

Allow Memtest86 to complete all four iterations of the 13 different tests, this may take a few hours. If no errors are found after the first four iterations then run Memtest86 again and do four more iterations. Even a single error means that you RAM is faulty, on the other hand no errors does not mean your RAM is good, it just means it's much less likely to be the root cause.

 

[moshan]

Thanks for the replies!

I will start with the memory tests using the recommended tool and then proceed with the BIOS and chipset updates. I hope I can post the memory tests results here.

Much appreciated your insight into this matter!

 

[moshan]

As above.

With AMD processors the BIOS also contains AMD specific AGESA microcode and you generally want to keep that updated. There are three BIOS versions dated 2023 for your motherboard, one of them (F5 - 30th March 2023) contains AGESA updated code. I would get on to the latest BIOS.

There are also chipset driver updates dated 31st March 2023, you certainly want to install those.

From the dumps, two of the BSODs happened during graphics operations, one happened during a networking operation, and one is a buffer overrun but with no third-party driver indicated - that suggests that bad RAM might be the cause. Bad RAM could easily be the cause of all of these BSODs.

The Windows memory test is pretty poor but it's important that we test your RAM to be sure it's not the root cause. Download Memtest86 and use the extracted tool to make a bootable USB drive containing Memtest86 (the USB drive can be less an 1GB in size). Then boot that USB drive and Memtest86 will start running.

If you need help creating the bootable USB drive see here: https://www.memtest86.com/tech_creating-window.html

Allow Memtest86 to complete all four iterations of the 13 different tests, this may take a few hours. If no errors are found after the first four iterations then run Memtest86 again and do four more iterations. Even a single error means that you RAM is faulty, on the other hand no errors does not mean your RAM is good, it just means it's much less likely to be the root cause.

I am coming back with results from the memory test:

MemTest86-Report-20230426-133158.html

drive.google.com

The first four iterations have passed the test, so hopefully I might not have to deal with faulty memory sticks.

I will update the chipset driver + BIOS as advised above and then re-run the test. I'll return with results once it's done.

 

[moshan]

I am coming back with results from the memory test:

MemTest86-Report-20230426-133158.html

drive.google.com

The first four iterations have passed the test, so hopefully I might not have to deal with faulty memory sticks.

I will update the chipset driver + BIOS as advised above and then re-run the test. I'll return with results once it's done.

I''ve updated the chipster driver along with the BIOS, which now runs on version F6b.

The memory tests ran slower this time (approximately 7h), but the results showed no errors. I also re-enabled AMD Expo after doing the BIOS upgrade so the same options as in the previous memory test.

So far I haven't hit any BSoDs, but I'll give it a few weeks as the crashes were not very frequent.