Question BSOD again (looking for someone to read minidump file)

[mant25]

Hello,

I've posted a thread a month ago but the suggestions that were offered didn't solve my problem as it still persists till this day so I hope moderators don't mind if I create a new one.

If anyone can read this and tell me where the issue lies, I will appreciate it.



UNEXPECTED_STORE_EXCEPTION (154)
The store component caught an unexpected exception.
Arguments:
Arg1: ffff8683fa2d5000, Pointer to the store context or data manager
Arg2: fffff68d56345930, Exception information
Arg3: 0000000000000002, Reserved
Arg4: 0000000000000000, Reserved

Debugging Details:
------------------


KEY_VALUES_STRING: 1

Key : Analysis.CPU.mSec
Value: 1890

Key : Analysis.DebugAnalysisManager
Value: Create

Key : Analysis.Elapsed.mSec
Value: 1954

Key : Analysis.Init.CPU.mSec
Value: 359

Key : Analysis.Init.Elapsed.mSec
Value: 8083

Key : Analysis.Memory.CommitPeak.Mb
Value: 88


FILE_IN_CAB: 050122-8796-01.dmp

DUMP_FILE_ATTRIBUTES: 0x8
Kernel Generated Triage Dump

BUGCHECK_CODE: 154

BUGCHECK_P1: ffff8683fa2d5000

BUGCHECK_P2: fffff68d56345930

BUGCHECK_P3: 2

BUGCHECK_P4: 0

EXCEPTION_RECORD: fffff68d563468b8 -- (.exr 0xfffff68d563468b8)
ExceptionAddress: fffff8011e522c50 (nt!RtlDecompressBufferXpressLz+0x0000000000000050)
ExceptionCode: c0000006 (In-page I/O error)
ExceptionFlags: 00000000
NumberParameters: 3
Parameter[0]: 0000000000000000
Parameter[1]: 00000204bf53e170
Parameter[2]: 00000000c0000185
Inpage operation failed at 00000204bf53e170, due to I/O error 00000000c0000185

EXCEPTION_PARAMETER1: 0000000000000000

EXCEPTION_PARAMETER2: 00000204bf53e170

CONTEXT: fffff68d563460f0 -- (.cxr 0xfffff68d563460f0)
rax=fffff8011e522c00 rbx=ffff9a01193d7000 rcx=ffff9a01193d7000
rdx=ffff9a01193d7000 rsi=0000000000000002 rdi=00000204bf53e170
rip=fffff8011e522c50 rsp=fffff68d56346af8 rbp=00000204bf53e2dc
r8=00000204bf53e170 r9=00000000000001c2 r10=ffff9a01193d7ea0
r11=00000204bf53e332 r12=fffff68d56346d68 r13=ffff868403bb6000
r14=ffff9a01193d8000 r15=0000000000000000
iopl=0 nv up ei pl zr na po nc
cs=0010 ss=0000 ds=002b es=002b fs=0053 gs=002b efl=00050246
nt!RtlDecompressBufferXpressLz+0x50:
fffff8011e522c50 418b08 mov ecx,dword ptr [r8] ds:002b:00000204bf53e170=????????
Resetting default scope

BLACKBOXBSD: 1 (!blackboxbsd)


BLACKBOXNTFS: 1 (!blackboxntfs)


BLACKBOXPNP: 1 (!blackboxpnp)


BLACKBOXWINLOGON: 1

CUSTOMER_CRASH_COUNT: 1

PROCESS_NAME: MemCompression

ERROR_CODE: (NTSTATUS) 0xc0000006 - The instruction at 0x%p referenced memory at 0x%p. The required data was not placed into memory because of an I/O error status of 0x%x.

EXCEPTION_CODE_STR: c0000006

EXCEPTION_PARAMETER3: 00000000c0000185

EXCEPTION_STR: 0xc0000006

STACK_TEXT:
fffff68d56346af8 fffff8011e4d26c0 : ffff9a01193d7000 ffff9a01193d7000 0000000000000002 00000204bf53e170 : nt!RtlDecompressBufferXpressLz+0x50
fffff68d56346b10 fffff8011e43b800 : 0000000000000001 0000000000000000 0000000000000000 00000000001193d6 : nt!RtlDecompressBufferEx+0x60
fffff68d56346b60 fffff8011e43b68d : 0000000000000004 fffff8011e43ba46 0000000000000000 0000000000000001 : nt!ST_STORE<SM_TRAITS>::StDmSinglePageCopy+0x150
fffff68d56346c20 fffff8011e43bfb8 : 0000000000000001 000000000001e170 ffff8683fa2d5000 ffff868400001000 : nt!ST_STORE<SM_TRAITS>::StDmSinglePageTransfer+0xa5
fffff68d56346c70 fffff8011e539c0c : fffff801ffffffff ffff868403bb6000 fffff68d56346d50 ffff8683f6592c10 : nt!ST_STORE<SM_TRAITS>::StDmpSinglePageRetrieve+0x180
fffff68d56346d10 fffff8011e539a59 : fffff801188f5730 0000000000000001 0000000000000000 0000000000000000 : nt!ST_STORE<SM_TRAITS>::StDmPageRetrieve+0xc8
fffff68d56346dc0 fffff8011e539911 : ffff8683fa2d5000 ffff8683f6592c10 ffff868403bb6000 ffff8683fa2d69c0 : nt!SMKM_STORE<SM_TRAITS>::SmStDirectReadIssue+0x85
fffff68d56346e40 fffff8011e4274b8 : ffff8684051bb640 ffff8683fa2d5000 0000000000000000 ffff868400f959c0 : nt!SMKM_STORE<SM_TRAITS>::SmStDirectReadCallout+0x21
fffff68d56346e70 fffff8011e53ade7 : fffff8011e5398f0 fffff68d56346f10 0000000000000003 0000000000000000 : nt!KeExpandKernelStackAndCalloutInternal+0x78
fffff68d56346ee0 fffff8011e4364dc : fffff68d56346fe0 0000000031526d73 00000000000003ff fffff8011ef239c0 : nt!SMKM_STORE<SM_TRAITS>::SmStDirectRead+0xc7
fffff68d56346fb0 fffff8011e435f10 : 000000000000000c 00000000000003ff fffff68d56347060 fffff8011ef239c0 : nt!SMKM_STORE<SM_TRAITS>::SmStWorkItemQueue+0x1ac
fffff68d56347000 fffff8011e53a047 : 000000000000000c 0000000000000001 ffff8683f6592c10 ffff868400f959c0 : nt!SMKM_STORE_MGR<SM_TRAITS>::SmIoCtxQueueWork+0xc0
fffff68d56347090 fffff8011e5324bf : ffff868400000001 ffff868400f95a80 0000000000000000 ffff8683fa2d5000 : nt!SMKM_STORE_MGR<SM_TRAITS>::SmPageRead+0x167
fffff68d56347100 fffff8011e532460 : 0000007f00000100 0000000000000000 fffff68d56347358 fffff8011e4739b0 : nt!SmPageRead+0x33
fffff68d56347150 fffff8011e47350d : 0000000000000002 fffff68d563471e0 fffff68d56347358 ffff868400f95970 : nt!MiIssueHardFaultIo+0x10c
fffff68d563471a0 fffff8011e4a6b58 : 00000000c0033333 0000000000000001 00007ff8b8612008 0000000000000000 : nt!MiIssueHardFault+0x29d
fffff68d56347260 fffff8011e605e5e : ffff8684000001e2 ffff8684051bb640 000001dc78210b80 0000000000010000 : nt!MmAccessFault+0x468
fffff68d56347400 00007ff8b86017cf : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 : nt!KiPageFault+0x35e
000000a55c3ff6c0 0000000000000000 : 0000000000000000 0000000000000000 0000000000000000 0000000000000000 : 0x00007ff8`b86017cf


SYMBOL_NAME: nt!RtlDecompressBufferXpressLz+50

MODULE_NAME: nt

IMAGE_NAME: ntkrnlmp.exe

IMAGE_VERSION: 10.0.19041.1682

STACK_COMMAND: .cxr 0xfffff68d563460f0 ; kb

BUCKET_ID_FUNC_OFFSET: 50

FAILURE_BUCKET_ID: 0x154_c0000006_c0000185_nt!RtlDecompressBufferXpressLz

OSPLATFORM_TYPE: x64

OSNAME: Windows 10

FAILURE_ID_HASH: {009a6bdf-fe23-c911-b99a-985b5dcd1647}

Followup: MachineOwner

 

[johnbl]

My initial virtual memory size was 32gb, after I turned off and turned it back on the virtual memory size is now 4096mb. I heard that the virtual memory size should be double your RAM size and no more than 3 times your RAM. So if my pc has 32gb ram installed and virtual memory is 4096mb is that not going to cause me problems since I'm always using heavy RAM usage programs doing multitasking?

The peripherals were connected while this happened:
Sony a6400 camera
SteelSeries Prisma QCK cloth

I haven't connected the SteelSeries Prisma QCK cloth peripheral for a month and I just did it yesterday and the BSOD happened right after a few hours of use. Maybe this was the peripheral that was trying to be installed (Windows said that it is trying to set up the device and after a minute it said installation was complete) but failed and somehow conflicted something and made my laptop to freeze and crash? This is like the 5-6 time that this happens.
I was also editing Photos on Photoshop with my camera connected to a laptop. Once the laptop froze I started to unplug all the devices and plug them back in but as I mentioned earlier once the system freeze happens and you unplug the peripherals and try to plug them back in, nothing happens. Programs also crash, I can move my mouse but you have to wait for atleast 10-15minutes until the BSOD happens and the laptop restarts.

So regarding virtual memory size should I set it up to more than 32000mb (something like 48000mb) or just leave it at 4096mb? (I have 1.3tb on my ssd so I have plenty of space for virtual memory) I feel like that is too little...

the system reported a bad block in the virtual memory. this means it was unable to either read or write to the block and is considered a hardware error.

The total size of the virtual memory is not related to the number of bad blocks. SSD have limited write endurance. This means after the system writes to a single location a limited number of times the entire block fails and becomes read only and the drive firmware must move the block to a new block and mark the old one as bad. ( and the drive size is reduced by the block size, so your drive space shrinks)
You want to make sure that you are using good devices to store your pagefile.sys on. for a ssd you want to make sure the firmware is up to date.
The firmware runs its trim functions 5 minutes after the system goes idle
these functions repair ssd drives and copy failing blocks to new locations and mark the old block as bad so it is not used again.

you can download and run crystaldiskinfo.exe and see the firmware versions and the health of your SSD.

windows might create several pagefile.sys files on various drives. You should look to make sure where they were created. You do not want them on slow memory devices like a flash card. I did not look to see how many pagefiles you had on your system.

 

[mant25]

the system reported a bad block in the virtual memory. this means it was unable to either read or write to the block and is considered a hardware error.

The total size of the virtual memory is not related to the number of bad blocks. SSD have limited write endurance. This means after the system writes to a single location a limited number of times the entire block fails and becomes read only and the drive firmware must move the block to a new block and mark the old one as bad. ( and the drive size is reduced by the block size, so your drive space shrinks)
You want to make sure that you are using good devices to store your pagefile.sys on. for a ssd you want to make sure the firmware is up to date.
The firmware runs its trim functions 5 minutes after the system goes idle
these functions repair ssd drives and copy failing blocks to new locations and mark the old block as bad so it is not used again.

you can download and run crystaldiskinfo.exe and see the firmware versions and the health of your SSD.

windows might create several pagefile.sys files on various drives. You should look to make sure where they were created. You do not want them on slow memory devices like a flash card. I did not look to see how many pagefiles you had on your system.

I only have one drive 2tb ssd.

Here is crystaldiskinfo: https://prnt.sc/ZYp8OPfAV3Wo 100% health I don't see anything bad with my drive.

That means the pagefile.sys can only be stored on my main drive as I don't have any more drives installed.

Regarding SSD firmware any other ways to check for updates other than going to device manager? I might have old firmware but I have no idea how to actually check that...

EDIT: I found some sites saying that the new firmware for my drive is GXB7601Q crystaldiskinfo shows I have GXB7301Q that means I'm running old firmware but I cannot find the updates anywhere on OEM websites..

 

[johnbl]

maybe use
Samsung Magician & SSD Tools & Software Update | Samsung Semiconductor Global

device manager will not install drive firmware updates.

 

[mant25]

maybe use
Samsung Magician & SSD Tools & Software Update | Samsung Semiconductor Global

device manager will not install drive firmware updates.

Samsung magician doesn't even report my SSD health status it shows N/A even tho I'm running samsung NVME drive.

anyways I updated the firmware to GXB7601Q (here is screenshot https://prnt.sc/EUW-RRmqE3Cq )

One more question here are the fixes that come with new firmware

Changes:
1. Fix Pyrite spec violation
Fix the FW flow not to erase the user data when TCG revert operation happened at Locking SP inactive status

2. Improvement of write performance drop when over 70% of SSD capacity is used
Foreground GC in the SSB area is performed to create a free block in the CDM
measurement section, so that the maximum SLC performance can be measured

My question is do any of these changes have anything to do with fixing my issue with bad block or does it mention anything related to that? Or do I just have to wait and see if this fixes it?

 

[johnbl]

Samsung magician doesn't even report my SSD health status it shows N/A even tho I'm running samsung NVME drive.

anyways I updated the firmware to GXB7601Q (here is screenshot https://prnt.sc/EUW-RRmqE3Cq )

One more question here are the fixes that come with new firmware

Changes:
1. Fix Pyrite spec violation
Fix the FW flow not to erase the user data when TCG revert operation happened at Locking SP inactive status

2. Improvement of write performance drop when over 70% of SSD capacity is used
Foreground GC in the SSB area is performed to create a free block in the CDM
measurement section, so that the maximum SLC performance can be measured

My question is do any of these changes have anything to do withth fixing my issue with bad block or does it mention anything related to that? Or do I just have to wait and see if this fixes it?

this one might:
Foreground GC in the SSB area is performed to create a free block in the CDM
gc is garbage collection. the drive only had 10 seconds to get a free block before the bugcheck. This could mean that more free blocks are available at a faster rate. old firmware often waited too long to run its garbage collection routines to free up previously used memory blocks. one of the "fixes" for a problem like this was to boot into bios and leave the drive powered but not in use for a few hours so the firmware garbage collection routines completed.
(this is just old files being deleted and put into a queue to complete the deletion later by the garbage collection routines in the drives firmware, it creates a lag where windows thinks the file is gone but the drive still knows about the file until the drive firmware completes its run of cleanup. this generally starts 5 minutes after the system goes idle. running a foreground gc process will do the cleanup even if the system is not idle. Most firmware started doing this because so many system are set to sleep very fast now.
ie start the cleanup at 5 minutes but the drive might be turned off at 10 minutes.

---
I am assume windows did not try to use your camera as storage and the bad block was in the camera memory card. if so you would want to do a full format of the memory card (not a quick format)

 

[mant25]

this one might:
Foreground GC in the SSB area is performed to create a free block in the CDM
gc is garbage collection. the drive only had 10 seconds to get a free block before the bugcheck. This could mean that more free blocks are available at a faster rate. old firmware often waited too long to run its garbage collection routines to free up previously used memory blocks. one of the "fixes" for a problem like this was to boot into bios and leave the drive powered but not in use for a few hours so the firmware garbage collection routines completed.
(this is just old files being deleted and put into a queue to complete the deletion later by the garbage collection routines in the drives firmware, it creates a lag where windows thinks the file is gone but the drive still knows about the file until the drive firmware completes its run of cleanup. this generally starts 5 minutes after the system goes idle. running a foreground gc process will do the cleanup even if the system is not idle. Most firmware started doing this because so many system are set to sleep very fast now.
ie start the cleanup at 5 minutes but the drive might be turned off at 10 minutes.

---
I am assume windows did not try to use your camera as storage and the bad block was in the camera memory card. if so you would want to do a full format of the memory card (not a quick format)

So I formatted my SD card that was inside my camera.

Regarding my SSD should I still boot into bios even after firmware update and let it sit there for a few hours and do it's cleanup? Also is there any rule of thumb as if how many hours my computer has to be booted up in bios for ssd to do it's cleanup? If this process runs automatically in the background when is the right time to assume SSD done its thing?

 

[johnbl]

So I formatted my SD card that was inside my camera.

Regarding my SSD should I still boot into bios even after firmware update and let it sit there for a few hours and do it's cleanup? Also is there any rule of thumb as if how many hours my computer has to be booted up in bios for ssd to do it's cleanup? If this process runs automatically in the background when is the right time to assume SSD done its thing?

for the sd card did you do a quick format? (take a second or two) or a full format (can take hour or more and finds all the bad blocks and marks them as bad.

for the drive, cleanup, I am not sure how fast it will go now. 10 years ago it would take several hours and depended on how much free space was on the drive. (faster cleanup with more free space)
With the firmware update you might not have to leave it in bios but it will cleanup much faster if windows is not running. IE powered on but no files in use.

 

[mant25]

for the sd card did you do a quick format? (take a second or two) or a full format (can take hour or more and finds all the bad blocks and marks them as bad.

for the drive, cleanup, I am not sure how fast it will go now. 10 years ago it would take several hours and depended on how much free space was on the drive. (faster cleanup with more free space)
With the firmware update you might not have to leave it in bios but it will cleanup much faster if windows is not running. IE powered on but no files in use.

I forgot to uncheck the "quick format". Tho my full format process got cut in the middle and finished without any messages of successful formating, guess it failed so I'l try it again one more time later.

Thanks for the insights on SSD cleanups, I'll keep that in mind and report with any new BSOD's or crashes if they might happen.

 

[johnbl]

I forgot to uncheck the "quick format". Tho my full format process got cut in the middle and finished without any messages of successful formating, guess it failed so I'l try it again one more time later.

Thanks for the insights on SSD cleanups, I'll keep that in mind and report with any new BSOD's or crashes if they might happen.

lots of counterfeit sd cards for sale. 8 gb card tweaked to look like 64 gb cards until you try to full format them. plus windows never used to verify data on usb connections unless you turn verify on (which no one ever does) years ago usb was considered too slow and so the default was verify off to save time.

 

[mant25]

lots of counterfeit sd cards for sale. 8 gb card tweaked to look like 64 gb cards until you try to full format them. plus windows never used to verify data on usb connections unless you turn verify on (which no one ever does) years ago usb was considered too slow and so the default was verify off to save time.

Now that you mentioned it and the fact that I bought it from amazon I had to double check. Seems like mines legit (checked a lot of articles about it) but I should be careful in case I purchase these from third party sellers.