Computer BSOD occurs due to ntoskrnl.exe crashing.
Computer BSOD occurs due to ntoskrnl.exe crashing.
I've recently changed some components on my PC and have experienced repeated but unpredictable crashes. I've attempted various fixes such as updating drivers, removing peripherals, checking for corrupted files, repairing drives, running memtest86, reseating the CPU and RAM, but none resolved the issue. The crashes occur during booting, while playing games, using a browser, or even when the computer is idle. They also happen when waking up from sleep mode after an irregular period. My system specifications are as follows: motherboard MSI pro A260M-E, CPU AMD Ryzen 5 8500G, graphics card Nividia GeForce RTX 3070 ROG STRIX GAMING OC, RAM two Patriot Signature Line DDR5 8gb sticks, and I've moved my main drive to a 2TB SSD. The crashes are often linked to the ntoskrnl.exe process. I wasn't sure whether to save the logs as text files, but uploading DMP files wasn't possible. The crashes began shortly after the PC started up, so it's unclear if they stem from outdated drivers, new software, or hardware problems.
Welcome to the forums, newcomer!
I've recently made some changes to my PC components and have been experiencing frequent but random crashes.
When sharing a troubleshooting thread, it's standard to provide your full system specifications. Please list them clearly as follows:
CPU:
CPU cooler:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
OS:
Monitor:
Include the age of the PSU along with its make and model.
Also, provide the BIOS version for your motherboard at this time.
Since you've upgraded or replaced parts, feel free to mention the components you've discarded from your previous build.
You mentioned selecting an incorrect board for your system—consider choosing one with a superior VRM design.
I also transferred my main drive to a 2TB SSD.
If you cloned your older OS drive onto the new one, that was an error; please reinstall the OS and you should be fine.
Regarding file formats, I wasn't sure if saving them as txt files would work, but it prevented me from uploading the dmp files directly.
You can store your .dmp files on a service like DropBox and share the link here.
note: your cpu came out jan 2024
windows version is strange:
Major Version : 15
Minor Version : 19041 (may 2020 release)
not sure why major version is 15 rather than 10
minor version, first part indicates a 2019 build
maybe updates not being applied? (microsoft amd cpu specific dll might not be loaded for your newer cpu)
you have to ignore the bugchecks that have access violation for the error code. (number 1 and 5 below)
ignore the bugcheck with the undocumented error code #2
that leaves #3 and #4 as the most helpful bugchecks
number 4 is most common with a overheated cpu or a bad overclock or underclock. first thing I would look for is a cpu fan that is controlled by a usb port. IE your system goes to sleep but the port does not wake up and the fan does not restart.
Second thing would be having more than one CPU overclocking drivers installed. (debug command lmiftsm shows the list of drivers installed)
third thing to look for: make sure the GPU supplemental power connectors are connected to the GPU. if the GPU does not get proper power, it will pull to much power thru the PCI/e bus and the motherboard power protection circuit will reset the CPU. if this is the case you want to look in the debugger and look at the system up time. If the time is short (under 15 seconds) it indicated the motherboard reset the CPU but the power supply did not hold the cpu from restarting, this causes the system to reboot and immediately restart with low power on the CPU pins. It can then lead the cpu to detect errors and call a bugcheck.
the errors relating to the PTE or PFN could be cpu errors or errors caused by power fluctuations but then saved to virtual memory (pagefile.sys) you will want to google "how to make windows delete the pagefile.sys on system reboot" make the registry changes and reboot. (just to prevent confusion as to the cause of the error)
The virtual memory error can also happen with overheated cpu, bad overclock to cpu, bad voltage to the CPU/gpu and bad firmware in a ssd or need to update CPU chipset drivers.
best guess is overheated cpu (assuming you do not have two copies of a overclocking driver installed, and no bios overclock and your voltages from your PSU are within spec.
if you have a all in one cooling for your CPU that is connected to a usb port. go into windows control panel device manager, find the usb port, right mouse click to bring up properties then find the power management tab and uncheck the option to let windows turn off the device to save power. short burst of no cooling fan on a cpu could cause overheating of cpu.
providing actual minidump.dmp file can help eliminate some of the causes of this error. But the interesting bugcheck are panic bugcheck from the CPU and may not have all of the proper data stored.
notes:
1) 040325-5859-01.txt error code = access violation
2) 040325-5781-01.dmp bugcheck 0x139 error code undocumented
3) 040325-5718-01.dmp bugcheck 0x1a error code 0x888a
Internal memory management structures (likely the PTE or PFN) are corrupted.
4) 040325-5609-01.dmp bugcheck 0x1000007f UNEXPECTED_KERNEL_MODE_TRAP error code 8
error code 8 is a double fault. (error happened then a error handler also got a error)
(could be a stack overflow/corruption)
5) 040325-5515-01.dmp bugcheck because a service died due to access violation
note: also, if you can not find a cause of this problem, make sure you update the bios and motherboard drivers. some versions of the CPU were unstable during low power mode that new windows sleep functions use. (two low of voltage, bios updates fix this, also the fixes are put in certain version of cpu specific drivers, (ryzen master driver can override bios default voltages)
note: bugcheck with access violation error code:
can indicate a bad value for memory timing. best to update the bios, set your memory profile and run memtest86 to confirm your memory timings.
(I would skip this if you have run the test before and it passes. (since you mentioned that problem happen mostly after sleep)
I have discovered the problem! There were corrupted files on my C disk, and I needed to run run chkdsk /r to locate them. I was misled by error messages about driver and kernel issues that the sfc couldn't resolve. I repeatedly checked my drivers, but while trying to delete something in the control panel, it would crash whenever I opened the uninstall program. I'm hoping this will fix the issue or if it's just a temporary fix, since my PC has been working fine for a while now.
corrupted files frequently result from a bugcheck process. chkdsk is typically set to execute after such a check. if you suspect the corruption occurred before the bugcheck, consider updating the CPU chipset drivers and disabling virtual memory to remove the pagefile.sys. then restart it to create a fresh one. download and execute a utility such as crystaldiskinfo.exe, which analyzes the drive’s SMART data and provides insights into its condition. if the bugcheck led to corruption, it is usually caused by enabled drive cache and lazy writes. certain drives may require firmware updates.