Question BSOD Memory_management, after nvme installed

[giffmetango]

So i bought new nvme g3 plus 1tb, install it into pcie 3 slot, i just using phison to check it nand, after awhile it just dismount itself, bsod, restart and check, it disappeared from bios. I just pull it out, put it pack, and tightly the nvme screw (was 90% tight before) this time it show up, but from that day, pc now have randomly BSOD memory_management, until today it have 4times.

https://imgur.com/a/mWm2B4V

View: https://imgur.com/a/mWm2B4V


Before install new nvme, i just using lexar 128GB SSD, and it worked pretty well without any BSOD.

Windows ram test is fine, nothing wrong,

2697v3 (Unlocked bios, undervolt -0.041v)
32GBx2 2133mhz ECC hynix

This morning i just tried to remove undervolt, to see if that causes bsod or not.

By looking into HWInfo, i saw abnormal in CPU

https://imgur.com/Q9Gu74o

View: https://imgur.com/Q9Gu74o

https://imgur.com/GqwdHKq

View: https://imgur.com/GqwdHKq

https://imgur.com/iDL9xDY

View: https://imgur.com/iDL9xDY




Does it also causes the BSOD ? because it much more too high for 2697v3, should i lock cpu state in bios again ?

 

[Lutfij]

i bought new nvme g3 plus 1tb
You mean this;
https://apac.kioxia.com/en-apac/personal/ssd/exceria-plus-g3-nvme-ssd.html
?

install it into pcie 3 slot
+
2697v3 (Unlocked bios, undervolt -0.041v)
For the sake of relevance, we'll need to know;
a| the make and model of your motherboard
+
b| the BIOS version for said motherboard.

If you're using an adapter akin to a PCIe riser board, please pass on a link to said riser. Pass on a screenshot of what you see in Disk Management.

 

[giffmetango]

i bought new nvme g3 plus 1tb
You mean this;
https://apac.kioxia.com/en-apac/personal/ssd/exceria-plus-g3-nvme-ssd.html
?

install it into pcie 3 slot
+
2697v3 (Unlocked bios, undervolt -0.041v)
For the sake of relevance, we'll need to know;
a| the make and model of your motherboard
+
b| the BIOS version for said motherboard.

If you're using an adapter akin to a PCIe riser board, please pass on a link to said riser. Pass on a screenshot of what you see in Disk Management.

yes here's model, cpu and ram

https://imgur.com/a/nPyla85

View: https://imgur.com/a/nPyla85


also i let it idle for awhile, the cpu hit beyond max core once again
when i'm playing game, i just set the game into windowed mode, enough space to show CPU Core value in HWMonitor

i noticed it suddenly push all core into 4138.4Mhz then back to normal, i dont think it normal hmm

guide i use to unlock bios

View: https://www.youtube.com/watch?v=vpCgdtRhabw

 

[johnbl]

all three bugchecks were due to virtual memory getting corrupted.
generally, you would update bios and chipset drivers, firmware of any drive and retest.
in your case I would be looking at overheated drives and bad connections. IE put a fan on the drive and see if it helps.
The last bugcheck might indicate general corruption or a old version of a special drive device driver or old firmware.
error code was:
An error occurred in the process of unsecuring memory that previously was secured. This error might happen if the caller mistakenly invoked MmUnsecureVirtualMemory in the wrong process context.

(you could get this if the drive overheated and could not reconnect)

i would also run crystaldiskinfo.exe and read the smart data from the drive that is holding your pagefile.sys
you might consider removing your pagefile.sys rebooting and creating a new one to dump potential errors due to bad spots on the drive.
(mechanical drive with 2.8 years on time might have smart errors and poor health due to wear)

for other causes I would have to look at the actual memory dumps. The fact you got a memory dump at all is useful depending on where it is being stored (which physical drive)

one note: since you installed a new drive, most people do not do a full verified format so you should turn off sleep functions and leave the computer on over night. After 5 minutes of idle, the system will start a background thread and will try to verify the locations on the new drive for bad spots. This process can take a very long time (12 hours) if the drive sleeps quickly (10 minutes) then the process that does the checking gets starved and the drive ends up using bad locations on the drive and you have problems. Old drives are repaired as sectors fail but this leads to rapid failure at the drives end of life.

 

[giffmetango]

all three bugchecks were due to virtual memory getting corrupted.
generally, you would update bios and chipset drivers, firmware of any drive and retest.
in your case I would be looking at overheated drives and bad connections. IE put a fan on the drive and see if it helps.
The last bugcheck might indicate general corruption or a old version of a special drive device driver or old firmware.
error code was:
An error occurred in the process of unsecuring memory that previously was secured. This error might happen if the caller mistakenly invoked MmUnsecureVirtualMemory in the wrong process context.

(you could get this if the drive overheated and could not reconnect)

i would also run crystaldiskinfo.exe and read the smart data from the drive that is holding your pagefile.sys
you might consider removing your pagefile.sys rebooting and creating a new one to dump potential errors due to bad spots on the drive.
(mechanical drive with 2.8 years on time might have smart errors and poor health due to wear)

for other causes I would have to look at the actual memory dumps. The fact you got a memory dump at all is useful depending on where it is being stored (which physical drive)

one note: since you installed a new drive, most people do not do a full verified format so you should turn off sleep functions and leave the computer on over night. After 5 minutes of idle, the system will start a background thread and will try to verify the locations on the new drive for bad spots. This process can take a very long time (12 hours) if the drive sleeps quickly (10 minutes) then the process that does the checking gets starved and the drive ends up using bad locations on the drive and you have problems. Old drives are repaired as sectors fail but this leads to rapid failure at the drives end of life.

my mobo dont have sleep function, hibernate already disabled
pagefile always in C drive which are windows partition

Currently i just disabled undervolt and pc still running (24/24) without any bsod from that time i posted,

tomorrow i think i should swap RAM slot from 1-2 to 2-4 instead