BSOD with no dump file due to SSD issue (...or motherboard or PSU)

[zatchmo137]

Hi All,

 

This is my first post here so hopefully I do it correctly! I want to try to be as brief as possible but provide as much explanation as I can on what I have done as this issue seems to be relatively complicated but close to a fix. So my computer was working great with the following build.

 

CPU: AMD Ryzen 7 7800X3D 4.2 GHz 8-Core Processor
CPU Cooler: Noctua NH-U12S chromax.black 55 CFM CPU Cooler 
Motherboard: MSI B650 GAMING PLUS WIFI ATX AM5 Motherboard  
Memory: Corsair Vengeance RGB 32 GB (2 x 16 GB) DDR5-6000 CL36 Memory
Storage: Western Digital Green 240 GB 2.5" Solid State Drive (Running my windows)
Storage: Samsung 970 Evo Plus 2 TB M.2-2280 PCIe 3.0 X4 NVME Solid State Drive *(Use this to run my games off of and it is plugged in to the upper nvme slot with the heatshield. Important for later)
Storage: Western Digital Blue 4 TB 3.5" 5400 RPM Internal Hard Drive
Storage: Seagate Barracuda Compute 2 TB 3.5" 7200 RPM Internal Hard Drive 
Video Card: XFX Speedster MERC 310 Black Edition Radeon RX 7900 XTX 24 GB Video Card 
Case: Corsair 4000D Airflow ATX Mid Tower Case  ($84.97 @ Amazon) 
Power Supply: NZXT C850 (2022) 850 W 80+ Gold Certified Fully Modular ATX Power Supply 
Case Fan: Corsair iCUE SP120 RGB ELITE 47.7 CFM 120 mm Fans 3-Pack 
Case Fan: Corsair iCUE AF120 RGB ELITE 65.57 CFM 120 mm Fans 3-Pack

 

Then one day I ordered an NVME for my wife's computer on amazon, it wasn't going to arrive in time, so I bought another and used the one that arrived in the mail for my PC. That SSD was the Samsung 990 EVO 1 TB. I installed the SSD and did a clean windows install on the SSD. At the same time (or almost so) I also updated my bios to get some faster boot times. For the next few weeks I encountered what seemed to be random crashes on my PC primarily when I was going from full load playing a game to when it was idling (as an example I would play Indiana Jones for 3 hours, close the game, and it would crash). That happened most of the times but it would also crash randomly. When it did crash (and yes I read the rules) it would almost always do a restart crash, giving me a BSOD for seconds and then restarting. When attempting to get the crash file, I was never able to. I made sure the settings were right for the dump file but no matter what I would do it would not generate a dump file. I also used the event viewer which was relatively useless, only telling me that it unexpectedly lost power and restarted. When I did get a BSOD it gave me a few different errors. When the BSOD's gave me descriptions it would be either UNEXPECTED STORE EXCEPTION or CRITICAL PROCESS DIED or nothing at all. Other times it would just start to crash and then restart the computer. 

 

At this point I tried everything (most of the advice told me it at least had to do with the SSD). I replaced the SSD with a new one. I rolled back the bios. I got a UPS with AVR (something I already wanted anyways). I reset windows multiple times. I cleared CMOS. I changed some BIOS settings. One thing did seem to help which was adjusting the c-power state (This seemed to help the most but it still occurred). Turned off and on EXPO. Reseated the RAM and the graphics card. Upgarded and downgraded drivers. I tried everything to no avail for like a month.

 

Finally I got an error that seemed to provide more information, KERNAL_DATA_INPAGE_ERROR. This put the nail in the coffin that it was an issue with the storage drive. So I did my last ditch effort which was plugging my SATA SSD back in. And this worked my computer works fine again with 0 crashing albeit slower due to the SATA connection. 

 

So since I used a different SSD the likelihood that it is actually an issue with the SSD is slim to none which provides two possible issues and one very unlikely issue.

• A. The motherboard is unable to handle the power correctly on the SSD and is crashing the windows system due to spikes in power level (which I couldn't verify due the power issue (most likely) being the reason the crash was happening).
• B. The PSU is supplying to much power and is causing the shorting
• C. Unlikely but the SSD is heating up to much due to the SSD being behind the graphics card. I didn't see any weird temps, however, there is some weird temp showing up on HWMonitor for the second temp sensor. As far as I could tell this was weird but normal.
• D. Even more unlikely, some other issue I haven't considered but somehow fixed by the old Sata SSD holding together the glue of infinity.

Weird situation but I'm 99.9% sure that its either the motherboard or the PSU since using the Sata SSD has fixed this issue. So my question to all ya'll is should I purchase a new PSU or a new motherboard or both? One other detail that might be important is that my house has some weird wiring (hence why I bought the AVR UPS) so potential shutdown or over/under voltage may have caused the issue at first but is hopefully resolved now. The Sata SSD is about 5 years old now so I'm hoping to move on from it and retire it in peace.

Thanks for all your help!

 

[BillBill]

16 minutes ago, zatchmo137 said:

Hi All,

 

This is my first post here so hopefully I do it correctly! I want to try to be as brief as possible but provide as much explanation as I can on what I have done as this issue seems to be relatively complicated but close to a fix. So my computer was working great with the following build.

 

CPU: AMD Ryzen 7 7800X3D 4.2 GHz 8-Core Processor
CPU Cooler: Noctua NH-U12S chromax.black 55 CFM CPU Cooler 
Motherboard: MSI B650 GAMING PLUS WIFI ATX AM5 Motherboard  
Memory: Corsair Vengeance RGB 32 GB (2 x 16 GB) DDR5-6000 CL36 Memory
Storage: Western Digital Green 240 GB 2.5" Solid State Drive (Running my windows)
Storage: Samsung 970 Evo Plus 2 TB M.2-2280 PCIe 3.0 X4 NVME Solid State Drive *(Use this to run my games off of and it is plugged in to the upper nvme slot with the heatshield. Important for later)
Storage: Western Digital Blue 4 TB 3.5" 5400 RPM Internal Hard Drive
Storage: Seagate Barracuda Compute 2 TB 3.5" 7200 RPM Internal Hard Drive 
Video Card: XFX Speedster MERC 310 Black Edition Radeon RX 7900 XTX 24 GB Video Card 
Case: Corsair 4000D Airflow ATX Mid Tower Case  ($84.97 @ Amazon) 
Power Supply: NZXT C850 (2022) 850 W 80+ Gold Certified Fully Modular ATX Power Supply 
Case Fan: Corsair iCUE SP120 RGB ELITE 47.7 CFM 120 mm Fans 3-Pack 
Case Fan: Corsair iCUE AF120 RGB ELITE 65.57 CFM 120 mm Fans 3-Pack

 

Then one day I ordered an NVME for my wife's computer on amazon, it wasn't going to arrive in time, so I bought another and used the one that arrived in the mail for my PC. That SSD was the Samsung 990 EVO 1 TB. I installed the SSD and did a clean windows install on the SSD. At the same time (or almost so) I also updated my bios to get some faster boot times. For the next few weeks I encountered what seemed to be random crashes on my PC primarily when I was going from full load playing a game to when it was idling (as an example I would play Indiana Jones for 3 hours, close the game, and it would crash). That happened most of the times but it would also crash randomly. When it did crash (and yes I read the rules) it would almost always do a restart crash, giving me a BSOD for seconds and then restarting. When attempting to get the crash file, I was never able to. I made sure the settings were right for the dump file but no matter what I would do it would not generate a dump file. I also used the event viewer which was relatively useless, only telling me that it unexpectedly lost power and restarted. When I did get a BSOD it gave me a few different errors. When the BSOD's gave me descriptions it would be either UNEXPECTED STORE EXCEPTION or CRITICAL PROCESS DIED or nothing at all. Other times it would just start to crash and then restart the computer. 

 

At this point I tried everything (most of the advice told me it at least had to do with the SSD). I replaced the SSD with a new one. I rolled back the bios. I got a UPS with AVR (something I already wanted anyways). I reset windows multiple times. I cleared CMOS. I changed some BIOS settings. One thing did seem to help which was adjusting the c-power state (This seemed to help the most but it still occurred). Turned off and on EXPO. Reseated the RAM and the graphics card. Upgarded and downgraded drivers. I tried everything to no avail for like a month.

 

Finally I got an error that seemed to provide more information, KERNAL_DATA_INPAGE_ERROR. This put the nail in the coffin that it was an issue with the storage drive. So I did my last ditch effort which was plugging my SATA SSD back in. And this worked my computer works fine again with 0 crashing albeit slower due to the SATA connection. 

 

So since I used a different SSD the likelihood that it is actually an issue with the SSD is slim to none which provides two possible issues and one very unlikely issue.

• A. The motherboard is unable to handle the power correctly on the SSD and is crashing the windows system due to spikes in power level (which I couldn't verify due the power issue (most likely) being the reason the crash was happening).
• B. The PSU is supplying to much power and is causing the shorting
• C. Unlikely but the SSD is heating up to much due to the SSD being behind the graphics card. I didn't see any weird temps, however, there is some weird temp showing up on HWMonitor for the second temp sensor. As far as I could tell this was weird but normal.
• D. Even more unlikely, some other issue I haven't considered but somehow fixed by the old Sata SSD holding together the glue of infinity.

Weird situation but I'm 99.9% sure that its either the motherboard or the PSU since using the Sata SSD has fixed this issue. So my question to all ya'll is should I purchase a new PSU or a new motherboard or both? One other detail that might be important is that my house has some weird wiring (hence why I bought the AVR UPS) so potential shutdown or over/under voltage may have caused the issue at first but is hopefully resolved now. The Sata SSD is about 5 years old now so I'm hoping to move on from it and retire it in peace.

Thanks for all your help!

 

KERNAL_DATA_INPAGE_ERROR can be due to faulty RAM or a failing storage device. You can download CrystalDiskInfo and check the health of each drive. You can also run a memory test to check on the RAM. 

[zatchmo137]

Just now, BillBill said:

KERNAL_DATA_INPAGE_ERROR can be due to faulty RAM or a failing storage device. You can download CrystalDiskInfo and check the health of each drive. You can also run a memory test to check on the RAM. 

I did the windows test on the ram and it didn't throw any issues. Samsung Magician also didn't have any problem with either of the SSDs.

[Cringe_Master245]

23 minutes ago, zatchmo137 said:

I did the windows test on the ram and it didn't throw any issues. Samsung Magician also didn't have any problem with either of the SSDs.

Like @BillBillsaid, run Crystaldiskinfo and memtest86. These are very well-known and good tools. I'm not saying the 'windows test" and Samsung magician are bad, just that Crystal and memtest would be helpful.

 

[zatchmo137]

38 minutes ago, Cringe_Master245 said:

Like @BillBillsaid, run Crystaldiskinfo and memtest86. These are very well-known and good tools. I'm not saying the 'windows test" and Samsung magician are bad, just that Crystal and memtest would be helpful.

 

Understood, how would an issue with the Ram manifest on the NVME SSD but not the SATA SSD, or similarly since I tested 2 different SSD's why would that be a factor? Genuinely, just wondering since I'm not having the issue on my current build?

[Bjoolz]

I'm feeling strong deja vu with this thread. Were you on the discord or on reddit with this issue? I'll reply here regardless.

 

I hate to be against everyone, but with NVMe SSDs checking SMART is completely useless. SMART is the self diagnostic of the SSD which CrystalDiskInfo reads. SMART was nerfed into the ground with NVMe SSDs. I've seen hundreds of faulty NVMe SSDs and less than five have showed any signs of failure in SMART. They also removed almost all of the useful parameters for checking the health. The general info in the top half of CDI isn't really trustworthy with HDDs or SATA SSDs either because it's up to the manufacturer when the status changes and most of them are scumbags.

 

From your BSOD errors, storage would be the main suspect. It can also be the M.2 port/motherboard, the SSD is just way more commonly faulty. Because you already tried replacing the SSD, the motherboard would be the next suspect. 

[zatchmo137]

13 hours ago, Bjoolz said:

I'm feeling strong deja vu with this thread. Were you on the discord or on reddit with this issue? I'll reply here regardless.

 

I hate to be against everyone, but with NVMe SSDs checking SMART is completely useless. SMART is the self diagnostic of the SSD which CrystalDiskInfo reads. SMART was nerfed into the ground with NVMe SSDs. I've seen hundreds of faulty NVMe SSDs and less than five have showed any signs of failure in SMART. They also removed almost all of the useful parameters for checking the health. The general info in the top half of CDI isn't really trustworthy with HDDs or SATA SSDs because it's up to the manufacturer when the status changes and most of them are scumbags.

 

From your BSOD errors, storage would be the main suspect. It can also be the M.2 port/motherboard, the SSD is just way more commonly faulty. Because you already tried replacing the SSD, the motherboard would be the next suspect. 

I've posted in a few different places without much success but I swear this is a newly written post written specifically for the forum! I appreciate the advice. I'm assuming its most likely just an issue with that specific port as I had never used it before and I just got unlucky. Thank you!

[synergistftw]

On theory - what are your NVme controller temperatures?

 

I also have a 4 TB Samsung 990 Pro on my X670E ProArt as my OS drive. The controller on the Samsung 990 Pro NVMe runs hot, if you don't have the heatsinked variant and you are relying on case airflow or the motherboard's heatsink cover, they may not be very effective. On your mobo, the NVMe heatsink covers aren't really designed for optimum cooling. I've built in the 4000D and it's a great case, but it's not got the most amount of airflow through it, which could be a factor.

 

For example, in my case, due to my GPU blocking a lot of the vertical airflow through my case (Fractal Torrent!) where the first NVMe slot is located (just above the GPU, like on your board), I ended up having to swap NVMes between slots. I also fitted a small finned NVMe heatsink on the 990 Pro, which keeps its temperatures at a better level. The 990 Pro still idles at ~50-55 Celsius - it was frequently over 65 before during the same workloads, which is getting close to Samsung's stated upper maximum operating temperature of 70 Celsius. A 2 TB Corsair MP600 Pro NH I have in the other slot runs around 44-48 Celsius.

 

Check your NVMe temperatures using HWiNFO64 or similar SMART monitoring tool. If the 990's controller is running very hot, it might be a contributing factor, despite the fact Samsung say it should be fine up to 70 degrees Celsius.

 

Other causes of a kernel inpage error can include an unstable overclock or very aggressive RAM timings, memory which is running at an EXPO profile which is not QVLed for the motherboard, a failing PSU or an overly aggressive undervolt or overclock. Is your system running stock clock speeds and voltages?

[Bjoolz]

9 hours ago, zatchmo137 said:

I've posted in a few different places without much success but I swear this is a newly written post written specifically for the forum! I appreciate the advice.

I found it, it was on r\Techsupport on reddit. To me it seemed like you were 50/50 on it being the PSU or motherboard so I just replied that I was very skeptical about it being the PSU. 

[zatchmo137]

On 1/15/2025 at 8:05 PM, synergistftw said:

On theory - what are your NVme controller temperatures?

 

I also have a 4 TB Samsung 990 Pro on my X670E ProArt as my OS drive. The controller on the Samsung 990 Pro NVMe runs hot, if you don't have the heatsinked variant and you are relying on case airflow or the motherboard's heatsink cover, they may not be very effective. On your mobo, the NVMe heatsink covers aren't really designed for optimum cooling. I've built in the 4000D and it's a great case, but it's not got the most amount of airflow through it, which could be a factor.

 

For example, in my case, due to my GPU blocking a lot of the vertical airflow through my case (Fractal Torrent!) where the first NVMe slot is located (just above the GPU, like on your board), I ended up having to swap NVMes between slots. I also fitted a small finned NVMe heatsink on the 990 Pro, which keeps its temperatures at a better level. The 990 Pro still idles at ~50-55 Celsius - it was frequently over 65 before during the same workloads, which is getting close to Samsung's stated upper maximum operating temperature of 70 Celsius. A 2 TB Corsair MP600 Pro NH I have in the other slot runs around 44-48 Celsius.

 

Check your NVMe temperatures using HWiNFO64 or similar SMART monitoring tool. If the 990's controller is running very hot, it might be a contributing factor, despite the fact Samsung say it should be fine up to 70 degrees Celsius.

 

Other causes of a kernel inpage error can include an unstable overclock or very aggressive RAM timings, memory which is running at an EXPO profile which is not QVLed for the motherboard, a failing PSU or an overly aggressive undervolt or overclock. Is your system running stock clock speeds and voltages?

Thanks for the really detailed explanation of everything. Yes, everything is running at stock clock speeds and voltages on my PC, besides EXPO. I saw some pretty high temps on the controller temp probe but the other temp was "technically" within range. I also have the 970 with the Mobo heatshield which seems to be just fine.