1. #1

    Halp! HTPC blue-screening at seeming random.

    I have an HTPC -

    ASUS B450M Pro 4
    Ryzen 2400G (stock cooler)
    8GB DDR4 (no XMP profile enabled, so running at 2133 or 2666, whichever B450 defaults to) - pretty sure its G.Skill (3200 rated, but since its just an HTPC i didnt want to futz with trying to get it to run at 3200)
    256GB SanDisk SATA SSD
    Win10 Pro (fully up to date)

    I often leave it on for long periods of time, as i help seed torrents (legal ones, for ex. i seed stuff for OCRemix.org).

    I would often swap back to that input to find a black screen and the computer apparently crashed/locked up.

    I hadn't, until recently, been able to see/know that it was blue screening, but just a few weeks ago, it finally did it while i was still on that input.

    It popped an "IRQL_NOT_LESS_OR_EQUAL" error, which the interwebs tells me is memory related.

    Thinking it might be the ram (it was a used kit that i picked up cheap on FB Market specifically for the HTPC, built somewhat on the cheap, back when RAM was spiking), i ran MemTest86 on it.. and forgot i was doing so, so it ran for several days and returned no errors.

    Im somewhat at a loss for what else it could be, memory related, and why it happens intermittently. Sometimes itll go weeks, then sometimes itll just be a few hours after i turn it on.

    I'd examine logs, but its one of the areas where i'm sorta Windows-Dumb - i have no idea where to look or what logs to look for.

    Under other circumstances id swap out the RAM, but i dont have any spare DIMMs lying around and i dont really want to spend money on more RAM if thats not the problem.

    If anyone has any ideas, and/or can tell me what logs to look for and where to find them, that'd be great. (Could it be an issue with the SSD? How would i even test it?)

    Its somewhat annoying because sometimes im the only person seeding some of this stuff, and ive also been trying to upload a large set of files to cloud backup, and its frustrating to make essentially no progress because the machine imploded 30 minutes after starting the upload.

    Thanks in advance and if there's any info i didnt post that would be needed or helpful, let me know.

  2. #2
    The IRQL... error can be related to drivers so I would check them before thinking about replacing hardware.

  3. #3
    Moderator chazus's Avatar
    10+ Year Old Account
    Join Date
    Nov 2011
    Location
    Las Vegas
    Posts
    17,222
    Yeah, most often its a driver issue.

    First, I'd check the drive on principle (I do this with -every- computer just to check). Open command prompt as admin, and type 'chkdsk' and see if it comes up with anything.

    The next step would be poking at eventviewer or running WhoCrashed (Or BlueScreenView) to see if it has a specific file/driver that failed
    Gaming: Dual Intel Pentium III Coppermine @ 1400mhz + Blue Orb | Asus CUV266-D | GeForce 2 Ti + ZF700-Cu | 1024mb Crucial PC-133 | Whistler Build 2267
    Media: Dual Intel Drake Xeon @ 600mhz | Intel Marlinspike MS440GX | Matrox G440 | 1024mb Crucial PC-133 @ 166mhz | Windows 2000 Pro

    IT'S ALWAYS BEEN WANKERSHIM | Did you mean: Fhqwhgads
    "Three days on a tree. Hardly enough time for a prelude. When it came to visiting agony, the Romans were hobbyists." -Mab

  4. #4
    While i was trying to do what Chazus suggested, it crashed a few more times relatively close together, with the following additional stop codes:

    Kernel Security Check Failure

    System Service Exception
    (what failed: dxgkrnl.sys)

    Video TDR Failure
    (what failed: amdkmdag.sys)

    Anything there jumping out at anyone?

    I just decided to take it down and clean it (it was due anyway) and ill just do a completely clean install of Win 10 (it being an HTPC, there's not a lot of setup that needs to be done, just a few basic apps to download like VLC, a torrent client, the NAS software, and 7-Zip).. .

    But i wanted to see if any of those errors jumped out at anyone as a possible hardware issue.

    Thanks again.

  5. #5
    Quote Originally Posted by Kagthul View Post
    While i was trying to do what Chazus suggested, it crashed a few more times relatively close together, with the following additional stop codes:

    Kernel Security Check Failure

    System Service Exception
    (what failed: dxgkrnl.sys)

    Video TDR Failure
    (what failed: amdkmdag.sys)

    Anything there jumping out at anyone?

    I just decided to take it down and clean it (it was due anyway) and ill just do a completely clean install of Win 10 (it being an HTPC, there's not a lot of setup that needs to be done, just a few basic apps to download like VLC, a torrent client, the NAS software, and 7-Zip).. .

    But i wanted to see if any of those errors jumped out at anyone as a possible hardware issue.

    Thanks again.
    Without googling i'm pretty sure the first is directx related, which would be GPU.
    But without digging deeping, like chaz said, its hard to tell you more.
    Start with reading event viewer and then analyze the blue screen dump, that will tell you exactly what happened and why. Everything else is just guessing.

  6. #6
    Quote Originally Posted by tomten View Post
    Without googling i'm pretty sure the first is directx related, which would be GPU.
    But without digging deeping, like chaz said, its hard to tell you more.
    Start with reading event viewer and then analyze the blue screen dump, that will tell you exactly what happened and why. Everything else is just guessing.
    There isn't one. Machine immediately hangs and never makes it past 0% on the blue screen dump/progress.

    Hokay, so:

    I did a complete nuke and pave, completely re-installing Windows fresh and wiping the drive with the most up to date installer.

    The ONLY things i installed were:
    The Ethernet Driver (Realtek)
    The AMD Chipset drivers for the B450M Pro 4
    The GPU drivers for the Vega 11 iGPU

    (many restarts involved, of course)

    Then, blamo, blue screen. Flashed past so fast i couldnt even see what it was, but after the machine rebooted (wich it never did before, it would always just hang there on the blue screen), i installed one of the programs Chazus recommended (WhoCrashed), and it turned up these:

    Crash dump directories:
    C:\Windows
    C:\Windows\Minidump

    On Sun 12/5/2021 1:43:07 AM your computer crashed or a problem was reported
    crash dump file: C:\Windows\Minidump\120521-4390-01.dmp
    This was probably caused by the following module: ntoskrnl.exe (nt+0x3F71B0)
    Bugcheck code: 0x3B (0xC0000005, 0xFFFFF804137FF0E7, 0xFFFF9D8032B92920, 0x0)
    Error: SYSTEM_SERVICE_EXCEPTION
    file path: C:\Windows\system32\ntoskrnl.exe
    product: Microsoft® Windows® Operating System
    company: Microsoft Corporation
    description: NT Kernel & System
    Bug check description: This indicates that an exception happened while executing a routine that transitions from non-privileged code to privileged code.
    This appears to be a typical software driver bug and is not likely to be caused by a hardware problem.
    The crash took place in the Windows kernel. Possibly this problem is caused by another driver that cannot be identified at this time.



    On Sun 12/5/2021 1:43:07 AM your computer crashed or a problem was reported
    crash dump file: C:\Windows\MEMORY.DMP
    This was probably caused by the following module: win32kfull.sys (win32kfull+0x2E22E)
    Bugcheck code: 0x3B (0xC0000005, 0xFFFFF804137FF0E7, 0xFFFF9D8032B92920, 0x0)
    Error: SYSTEM_SERVICE_EXCEPTION
    file path: C:\Windows\system32\win32kfull.sys
    product: Microsoft® Windows® Operating System
    company: Microsoft Corporation
    description: Full/Desktop Win32k Kernel Driver
    Bug check description: This indicates that an exception happened while executing a routine that transitions from non-privileged code to privileged code.
    This appears to be a typical software driver bug and is not likely to be caused by a hardware problem.
    The crash took place in a Microsoft module. Your system configuration may be incorrect. Possibly this problem is caused by another driver on your system that cannot be identified at this time.

    Both times, it says its likely not a hardware issue... but i haven't installed much of anything. Just the basic chipset and GPU drivers for the iGPU, basically, so im unsure what could be causing this.

    Any insights would be great.

    Otherwise... i might just scrap the core of the thing even though its plenty powerful and get an i5 11400 with an H510 board (can grab the pair for ~200$ right now at Micro Center) and a small M.2, thereby removing any hardware issues other than RAM (which came back clean after nearly 4 days of MemTest because i forgot i had it running).

    - - - Updated - - -

    Eh, might have (i stress might) have tracked it down to bad DIMM slots, though im unsure what might have caused them to fail suddenly after a few years of service.

    While i had it apart, i replaced the SATA cables to the SSD and installed a 4TB drive i had lying around but did not wire it in yet.

    Even after the nuke and pave it was still just packing it in, quite frequently.

    It wasnt leaving crash dumps, just poof black screen.

    When trying to reinstall VLC it would reliably trigger the crash.

    The only thing i hadnt done was try swapping the DIMMs to the other slots. So i did, and the problem seems to be cleared up - at least insofar as it hasnt crashed in the last few hours.

    This shizz is infuriating sometimes, and is why my usual MO (afforadability being a factor, of course) is to just swap the hardware.
    Last edited by Kagthul; 2021-12-05 at 07:48 AM.

  7. #7
    Dreadlord sunxsera's Avatar
    10+ Year Old Account
    Join Date
    Jan 2012
    Location
    Germany | Blackmoore-EU
    Posts
    904
    Is your bios updated ?

    Edit: If yes, which version ?

  8. #8
    Just try 1 stick of ram, if it fails, try the other stick, if that fails, try 1 stick in the other dimm slot.

  9. #9
    Quote Originally Posted by sunxsera View Post
    Is your bios updated ?

    Edit: If yes, which version ?
    Updated as far as ASRock says is safe if you're running a 2400G, which is 3.50

  10. #10
    Dreadlord sunxsera's Avatar
    10+ Year Old Account
    Join Date
    Jan 2012
    Location
    Germany | Blackmoore-EU
    Posts
    904
    Quote Originally Posted by Kagthul View Post
    Updated as far as ASRock says is safe if you're running a 2400G, which is 3.50
    Nice, that's why i asked.
    So, you ran Memtest; RAM seems to be fine.
    Bluescreens indicate problems with ram or drivers.

    Did you try an older/newer version of the drivers? especially chipset drivers.
    In which DIMM slot do you have the rams ? try slot 1+3.

  11. #11
    Quote Originally Posted by sunxsera View Post
    In which DIMM slot do you have the rams ? try slot 1+3.
    You may have missed, i swapped them as the "only thing i hadnt tried", and it *seems* to have worked.

    Its been up for a couple of hours now without crashing.

    To even get it to boot (before the RAM swap), i had to use the basic restore point that Windows does when you first boot it up after installation, so as of now, there are no chipset drivers installed (confirmed by i do not have the "Ryzen Balanced" power plan available), the Radeon/GPU drivers arent installed, and its running on the default windows ethernet driver.

    Even after that, though, before i swapped the DIMMs, it would still crash the moment i tried to install anything. I assume because it was trying to decompress that data through a bad RAM channel (if the DIMM slots went bad, presumably).

    It was instantly repeatable.

    Now that i've swapped the DIMMs (they are now in 1/3), its been stable since the reboot. I installed VLC, WhoCrashed, and the software for my NAS, and all *seems* well, at least for now.

  12. #12
    Dreadlord sunxsera's Avatar
    10+ Year Old Account
    Join Date
    Jan 2012
    Location
    Germany | Blackmoore-EU
    Posts
    904
    Quote Originally Posted by Kagthul View Post
    You may have missed, i swapped them as the "only thing i hadnt tried", and it *seems* to have worked.

    Its been up for a couple of hours now without crashing.

    To even get it to boot (before the RAM swap), i had to use the basic restore point that Windows does when you first boot it up after installation, so as of now, there are no chipset drivers installed (confirmed by i do not have the "Ryzen Balanced" power plan available), the Radeon/GPU drivers arent installed, and its running on the default windows ethernet driver.

    Even after that, though, before i swapped the DIMMs, it would still crash the moment i tried to install anything. I assume because it was trying to decompress that data through a bad RAM channel (if the DIMM slots went bad, presumably).

    It was instantly repeatable.

    Now that i've swapped the DIMMs (they are now in 1/3), its been stable since the reboot. I installed VLC, WhoCrashed, and the software for my NAS, and all *seems* well, at least for now.
    Ahh okay. I somehow overread that and thought it still wouldn't work.

    Here's a small explanation & the reason why i asked in which slots you have the ram:
    https://forums.tomshardware.com/thre...first.3195865/

    Let's hope it's fixed!

    Edit: Doesn't matter if its 1+3 or 2+4.

  13. #13
    So... didn't work.

    Though it is no longer crashing immediately (or whatever is happening), it went down last night.

    Had run fine for the better part of 2 days, then i went to open a Chrome window and POOF, immediate black screen.

    Power isnt off (the fans are still running) but there's just instantly no video output.

    Manually power the system down (holding power button) and reboot it...

    And no crash logs.

    Im sorta at a loss.

    So today i installed furmark, crystaldiskmark, and prime95 and ran them all at once (i had the brief thought that maybe the PSU was flaky and was causing the system to fail, so i put it under full load) but the thing just sat there for two hours running fine.

    Eerrrggghh.

    So, off to Micro Center it is i guess.

    Thankfully with in-store pricing, an i3-10105 + MoBo is only ~200$ and i can make use of/abuse the "free 240GB SSD" thing they have going on to grab a new SSD just in case that is somehow the issue (flaky controller maybe, i dunno, as it mostly seems to pack it in when doing something that requires reads/writes, though it sat through 4 runs of CDM, so... /shrug).

    Only parts that wont have been replaced are PSU and RAM, so if it continues with the new setup, it will have narrowed it down to one of those.

  14. #14
    Quote Originally Posted by Kagthul View Post
    .
    Not sure why so many people keep talking about hardware when the crash dump explicitly says that its not?
    "This appears to be a typical software driver bug and is not likely to be caused by a hardware problem."

    Both of them tell you that.
    And they reference:
    win32kfull.sys
    ntoskrnl.exe

    The former is where the issue is most likely to be.
    I dont have time to dig through it now but check this link:
    https://docs.microsoft.com/en-us/ans...ice-excep.html
    And quote me so i remember to check this thread again

    - - - Updated - - -

    Also the reason its not leaving full crash dumps is because of setting enabled.
    Try boot into safe mode and find crash dump and enable "full kernel dump".

    - - - Updated - - -

    Last note, make sure to enable full kernel dumps and then just make it blue screen again. This should tell you what the conflicting driver is that it says is unidentified atm.
    At least i think it should, haven't really read blue screen dumps in forever ever... But it will tell you a lot more at least vs just a minidump

    Just because its the latest driver doesn't mean its a good driver.. I've had wifi drivers that's been incompatibility causing blue screens when it's had a specific network improvement driver that kept switching between wifi and lan in a way windows didn't like. Took forever to solve that one. And the solution was to go into safe mode without networking, disable the wifi device and uninstall the software that came with while installing said drivers, while also using a generic lan driver... Fun times.
    My point being, I would never have solved it without digging deep into logs and crash dumps. You will have to do the same.
    Last edited by tomten; 2021-12-09 at 12:07 AM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •