[Fixed] BSOD - - - Crashing out of nowhere - - I have no idea what to do next?
[Fixed] BSOD - - - Crashing out of nowhere - - I have no idea what to do next?
Hello, I have a problem with my computer few last days (I bought the PC at April this year). Just to describe the problem. One day I was peacefully playing Counter Strike with my brother and then it happened. The screen freeze and he scary BSOD shows up. So, I have just restarted PC and went back in game after few minutes it happened again... and again. I ended in the situation I personally called BSOD hell . I turned my PC on, log in to the system, 10 seconds is everything loading up, freeze, BSOD, and again. I tried some repairs (not relevant for now), nothing helped so I reinstalled Windows and I was setting everything up using Armoury Crate software from ASUS to get the latest drivers and then it happened again, with the clean install of the Windows, just updating a drivers. I ended in the BSOD hell again. So I reinstalled the computer again. I was more conscious about the next steps, so I let the Windows get everything (updates, security), then I started to update drivers one by one. I installed one driver and restart a computer, waited for a while and then the next driver and so on. I installed the latest drivers for everything. Everything worked just fine. Except that I found that one of the memory stick (DRAM 0 slot) shows up 0,5 degree temperature in the BIOS. Weird, right? I was like, cool I found the issue, it is the RAM stick with weird contact. I pulled out both sticks, switched them if the temperature will be now in the DRAM 1 slot. It was not, all looked great. So I took Prime95 and gave it 3 hours of stress testing. No crash, not a single issue. So I started installing next software like Steam, etc. After a day of happiness the stealthy BSODs came back. Just a few, not the whole BSOD hell. So I start debugging. Event viewer + Google if something suits my case. I found that the HyperX Cloud 7.1 USB sound card was making troubles so I updated firmware. One issue down but it was not relevant to the BSOD. So I went deeper the WinDbg + Google. I have two types of BSOD errors - CLOCK_WATCHDOG_TIMEOUT and MACHINE_CHECK_EXCEPTION . I googled that should be issue with the RAM, thus I run Memtest86 several times for each stick, all good, 0 errors. BTW the DRAM 1 slot (stick from DRAM 0 slot with 0,5 temp) shows the weird temperature again (was good). I have no clue what should I do next. Help me please. Everything was perfect since I bought this PC, but the few last days it is a struggle. HW INFO: MB: ASUS ROG STRIX B660-I GAMING WIFI CPU: Intel Core i5-13600KF RAM: Corsair 32GB KIT DDR5 5600MHz CL36 Vengeance SSD: Kingston KC3000 NVMe 2TB PSU: Corsair RM850 White (2021) Windows 11 Pro TLDR: My computer throws two types of BSOD out of nowhere. CLOCK_WATCHDOG_TIMEOUT and MACHINE_CHECK_EXCEPTION. One of the RAM stick shows weird temperature. No overclocking was done at all. All is stock. I have: Reinstalled Windows (twice) Prime95 stress tested whole system for a few hours with the latest drivers installed Memtested both memory sticks multiple times without errors. I have the latest drivers and updates everywhere. Checked with all the manufacturers . Google Drive folder with dump files <- DUMP files here I can buy a new part and return the broken one, no problem, but I can not buy a whole new PC.
Thanks a lot! I made all these parts myself. The BIOS is old (version 1620) and uses 64-bit code. My memory sticks are labeled "CMK32GX5M2B5600C36W".
Welcome to this forum! Reinstalled Windows (two times) Where did you get the installer for the operating system? Did you install it with offline mode and manually add all the needed drivers that are most recent, right-click on the installer and run it as an administrator?
Thanks! The installer was a factory reset. I tested the online option but it didn't work. Everything is set up through Armoury Crate on ASUS.
Thanks for getting back to me! Yesterday I updated the BIOS and got a blue screen twice. I tried focusing on the CPU, Intel Diagnostics said everything was fine, and stress tests passed too. But after some searching, someone who has had this exact problem with the same setup suggested turning off "CPU C States Support" in the BIOS. I'm testing it right now. This is super weird because many 13600K CPUs suddenly have these issues without any real work being done on them.
First things first; four of the dumps are (as you say) 0x101, CLOCK_WATCHDOG_TIMEOUT and these can only be fully analysed with the full kernel dump. Unfortunately there is only one of those recorded and it's always for the most recent BSOD, which isn't a 0x101 in your case. Sod's law of course! If/when you get another 0x101 BSOD please copy the file C:\Windows\Memory.dmp to a temp folder somewhere to prevent it being overwritten by a subsequent BSOD. Then upload it to the cloud - it will be large. The 0x101 BSOD is almost always a CPU issue, so if you're overclocking that K series CPU please remove it and run at stock frequencies until your problem is resolved. Same with your RAM, please disable XMP and run at stock frequencies. Any overclocking introduces instability, and 0x101 is the common excessive overclock BSOD (so too is the 0x9C, MACHINE_CHECK_EXCEPTION). The latest dump is a 0x3B, SYSTEM_SERVICE_EXCEPTION which means that some sort of exception occurred whilst running in kernel mode. In the call stack you can see that virtual memory operations were in progress... Code: STACK_TEXT: fffff086`a3fff490 fffff806`28749be4 nt!MiGetProtoPteAddress+0x38 fffff086`a3fff4e0 fffff806`2874ba2b nt!MiQueryAddressState+0x564 fffff086`a3fff700 fffff806`28bda75f nt!MiQueryAddressSpan+0x24b fffff086`a3fff7c0 fffff806`28bda005 nt!MmQueryVirtualMemory+0x73f fffff086`a3fff960 fffff806`288274e5 nt!NtQueryVirtualMemory+0x25 fffff086`a3fff9b0 00007ffb`43c0f854 nt!KiSystemServiceCopyEnd+0x25 00000039`95fffd58 00000000`00000000 0x00007ffb`43c0f854 The error occurred in the last call there (the top one)... Code: CONTEXT: ffffe001ddf7e900 -- (.cxr 0xffffe001ddf7e900) rax=ffffcc8720a80b20 rbx=00000007ffaef9ee rcx=ffffcc8720a80aa0 rdx=ffffcc8721a44520 rsi=0000000000000000 rdi=0000000000000000 rip=fffff8062874a948 rsp=fffff086a3fff490 rbp=fffff086a3fff5e0 r8=0000000000000004 r9=fffff086a3fff568 r10=00000007ffaebae0 r11=ffffcc8721a44568 r12=0000000000000004 r13=fffff43ffd77cf70 r14=fffff086a3fff568 r15=0000000000000000 iopl=0 nv up ei pl zr na po nc cs=0010 ss=0018 ds=002b es=002b fs=0053 gs=002b efl=00050246 nt!MiGetProtoPteAddress+0x38: fffff806`2874a948 f7413800000008 test dword ptr [rcx+38h],8000000h ds:002b:ffffcc87`20a80ad8=???????? Resetting default scope You can see the failing instruction is a TEST instruction using the RCX register as a memory pointer. However, the resulting memory address is invalid (=????????) meaning that the memory location was either not allocated, was paged-out (which is not allowed since we're running at an elevated IRQL), or the RAM page was bad. When you tested the RAM what tester did you use? If that 32GB kit is 2x16GB then try running on just one stick for a day or two (or until you get a BSOD) and then swap RAM sticks and run on just the other for a day or two (or until you get a BSOD). That is the 'gold standard' RAM test. I really would like to see the kernel dump for the next 0x101 BSOD however...
I don't know if this "gold standard RAM test" is actually real, but it can help find which stick is broken when it really is one. If the problem is about settings, software, or the motherboard instead of physical memory, this won't tell you much. When running memtest, make sure to check every single stick for four full passes using Memtest86 while all 11 tests pass on each run. Test them together again for another four full passes after that. This will take a long time if you choose to do everything manually. It is also okay to run the Windows memory diagnostic tools as well. After updating the BIOS, make sure to do a hard reset so nothing gets stuck in settings. Sometimes this happens a lot. Here is how to do it: Turn off the computer and switch the PSU off, then unplug the power cord from the wall or supply. Take out the motherboard CMOS battery for three to five minutes. In some cases you might need to remove the graphics card to get there. While the battery is gone, keep pressing the case button for 15 to 30 seconds to drain any leftover electricity in the circuit. After five minutes, put the battery back on and make sure it fits with the right side up as it came out. If you took out the graphics card now, you can put it back but also reconnect your power cables and display cable if they were attached. Plug the power cord back in, turn the PSU back on, and start the system. It should show a POST screen and let you choose between Windows or legacy boot options. Go into the BIOS setup program and change the boot settings to either Windows or legacy mode depending on your drive. Save everything and exit. If the computer starts up normally, move forward from there and come back into the BIOS if needed to set custom memory profiles like XMP, fan profiles, or any other settings you liked before. Sometimes after a reset, you might need to go back in and load default values so the hardware tables reset inside the boot manager. If anything needs a hard reset first thing, it is smart to try a different screen if there is no video signal. Some systems just won't work with certain display ports like DisplayPort for some reason. Try HDMI instead if you have no picture or can't see anything when trying in. Changing monitors too can help if there is no visual signal at all and that happens sometimes.
I really appreciate the help from your side! I still don't know how to fix these .dmp files, so thank you for all of your tips. I tried running memtest86 (free) already. They say there are not enough tests in that free version. But I run 2 full tests on each stick and 1 test with both sticks together. I ran one day with just one stick, then another day with the other stick. Both times I got a BSOD. I was almost giving up. BUT if I turned off CPU C States Support when I started my computer, there were no more BSODs. I run some stress tests like Counter Strike for 2 hours. Wish me luck! Hopefully that solved it. I wrote more about this in past replies. My next steps are: if another BSOD happens, please upload the whole Memory.dmp file so you can fix it again. Also, turn off XMP when you help and I will try harder to solve it.