System Specs before starting:
Fedora 43
KDE Plasma
Wayland
i7-6700K
GeForce GTX 1660 super
32 GB memory (unsure if ddr3 or ddr4)

Hi! Recently I convinced my partner to make the switch to linux, specifically, Fedora. However, sometimes her computer will just completely freeze, and requires a hard reboot, I haven’t gotten the chance to test if reisub works while it’s frozen. Also, very, very rarely, it will come back to life, or show some signs of life; today, it froze, but I could still hear her over our call, and then it slowly advanced the display over the next couple minutes- as in the mouse was moving very slowly according to how she was moving her mouse when it froze- before stopping again and requiring a reboot.

I’m pretty sure this has been happening since day 1, fresh install, so I don’t think it’s anything we did/installed that broke things.

My gut tells me that it’s an issue with her memory, but it seems like the majority of the time, it happens while using graphically intensive apps/games, but that’s not always consistent. This leads me to think it’s actually an nvidia issue, we’ve tried both the neuveau and proprietary drivers, but it happens on both.

Does anyone have any ideas about this?

  • x00z@lemmy.worldEnglish
    1·
    3 hours ago

    I have this exact same issue because of my GFX card.

    I think it overheats but it could just be borked too.

  • eleijeep@piefed.socialEnglish
    3·
    8 hours ago

    I just want to second (third?) the possibility that it’s an OOM. My system freezes in the same way when I run out of memory.

    Generally it’s because I have too many tabs open in Firefox, so I tweaked the Firefox settings to make it unload tabs when below a certain amount of free memory:

    browser.tabs.unloadOnLowMemory	true  
browser.low_commit_space_threshold_mb	8192	
browser.low_commit_space_threshold_percent	25  
browser.tabs.fadeOutUnloadedTabs	true  
browser.tabs.min_inactive_duration_before_unload	86400000

    The above settings in about:config will start to unload tabs when you get below 8Gb or 25% of free memory. The “inactive duration” is set to 1 day so that Firefox doesn’t just start unloading tabs unnecessarily.

    Since I changed these settings, I always have at least 8Gb memory free for other programs. You could reduce those values if you want to use more of your memory for Firefox.

  • Possibly linux@lemmy.zipEnglish
    2·
    9 hours ago

    My guess would be a hardware failure

    Check all of the system components. That machine is quite old and could be having issues.

    If it is really an Nvidia issue you could always pull the GPU.

    Another troubleshooting thing you could do is setup SSH and monitor dmesg

  • Riskable@programming.devEnglish
    18·
    15 hours ago

    I had this same thing happen a while back. You know what it was? A bad USB device!

    I had a little USB debug probe that went bad (somehow) and it totally screwed up my USB hub’s ability to… Stay stable? Haha, that’s the best way to put it.

    Anyway, the fix was to remove the device and disconnect the USB hub (and its power) for a few seconds. If I ever reconnected the probe, the problem would recur within an hour or two.

    Here’s how you can check for something similar: Run dmesg and look for regular messages like, “unable to enumerate device”. It’ll tell you which bus and port it’s on but that’s not easy to figure out so just keep unplugging things until you get the one matching the device that’s regularly throwing errors in dmesg. Keep it disconnected, power everything off (PC, USB hub’s, etc) for a few seconds and then try running without that device for a while. It might be the culprit!

    • Lucy :3@feddit.org
      2·
      9 hours ago

      Same thing, but with an external NVMe enclosure and it crashing the whole USB Bus of my MB it was plugged into until poweroff and poweron (not reboot!), meaning my peripherals disconnected too. Basically fixed by moving it to the front USB-Ports, now it crashes much rarer (previously a minute of playing MC with the profile residing on the NVMe would crash it, but not hours of RDR2 or similar. Very weird. Not it does after a few hours of MC), and can reliably be recovered by replugging.

      Still looking forward to a new PC with more that one NVMe-PCIe Slot though.

    • guynamedzero@piefed.zeromedia.vipOPEnglish
      6·
      15 hours ago

      Interesting, I have noticed that every time this happens, her camera usually gets fucked up for a little while. I guess it’s possible that it has something to do with that, I’ll have to check that next time it happens. That’s the only usb device she has plugged in that I’m suspicious of. The only other things she has are her mouse, keyboard, microphone, wireless headset dongle, and wifi adapter.

  • ohulancutash@feddit.ukEnglish
    41·
    10 hours ago

    This sounds very much like a swapfile issue. Check you have swap enabled and it is big enough for your needs.

    If the swapfile is filled, the OS will lock up and advance very slowly. You can monitor swap in top or whatever you prefer.

    Good news is this is trivial to fix.

    If your system keeps losing time that may point to the need to replace the CMOS battery, but its rare for a mobo to be in use long enough to need that.

  • paris@lemmy.blahaj.zone
    4·
    13 hours ago

    I’ve had this issue and for me using earlyoom helped. My computer ram would fill up and the kernel oom killer wouldn’t trip in time to be helpful. It could just be your memory filling up and your computer not killing the offending process in time.

  • SpikesOtherDog@ani.socialEnglish
    6·
    15 hours ago

    This sounds like a hardware issue. If you are lucky, it is a keyboard, mouse, or something. Doesn’t sound like memory because you would possibly have memory errors or kernel panics.

        • rem26_art@fedia.io
          2·
          13 hours ago

          Probably is worth the piece of mind to let memtest86+ run for a few cycles overnight, so you can rule memory out. Memory issues are never fun. May also want to check out the BIOS and play with the XMP settings for memory, if you’ve got any. I had a bunch of weird freezing issues (though mostly related to sleep mode) that stopped once I set my memory to a different XMP profile

  • who@feddit.orgEnglish
    6·
    15 hours ago

    it froze, but I could still hear her over our call, and then it slowly advanced the display over the next couple minutes-

    This means the system is not entirely frozen. The kernel is probably still running, but something is interfering with the display or some other subsystem(s). If SysRq key commands (reisub) worked on this system before the freeze, there’s a good chance they will still work after it happens.

    as in the mouse was moving very slowly according to how she was moving her mouse when it froze-

    I have seen behavior like this caused by a buggy video driver. You might want to experiment with different Nvidia driver versions.

    If that doesn’t bear fruit, you might experiment with different kernel versions.

    Edit 1: Also consider watching for kernel log messages when the problem happens (sudo dmesg --follow) or noting the time when it happens and then reviewing syslog/journalctl from that time after rebooting.

    Edit 2: Also take note of anything that consistently happens shortly before the problem turns up, like the screen going to sleep and waking up. Some devices have trouble with power saving features.

    • guynamedzero@piefed.zeromedia.vipOPEnglish
      1·
      15 hours ago

      If SysRq key commands (reisub) worked on this system before the freeze

      Yeahh, I only got SysRq commands working like an hour ago, because I didn’t realize that fedora had it off by default.

      different kernel versions.

      Assuming you literally just mean different versions of the kernel, it’s highly unlikely this would help, this issue has persisted over multiple kernels and even after upgrading from fedora 42 to fedora 43.

      reviewing syslog/journalctl from that time after rebooting.

      Ahhh, for some reason I had it in my head that rebooting cleared the syslog

      I’m not sure if changing the nvidia driver version would help, because like I said, it happens on both neuveau and proprietary.

      • who@feddit.orgEnglish
        1·
        14 hours ago

        Assuming you literally just mean different versions of the kernel,

        I do.

        it’s highly unlikely this would help, this issue has persisted over multiple kernels and even after upgrading from fedora 42 to fedora 43.

        It could help if you’re experiencing a kernel bug that has not been present forever, or was fixed upstream more recently than the kernels you have tried. If you establish that the problem does not occur when using a much older or newer kernel, it greatly narrows down where a developer would have to look for the bug.

        For what it’s worth, I’m not just throwing random ideas out there. I have some experience in this area. ;)

        In any case, good luck.

  • brucethemoose@lemmy.world
    51·
    15 hours ago

    Fedora

    GTX 1660 Super

    freezing

    Almost certainly a Nvidia issue, and that’s such a deep rabbit hole, I don’t even know where to start.

    …It’s not a fun troubleshooting/learning process.

    The most expedient thing would be to try Bazzite or Nobara for a while. Both are Fedora based distros (so they’re familiar), but they have better support for older Nvidia cards out of the box.

    This would be a better long term solution anyway, as whatever’s going wrong, you wont have to maintain it yourself through updates.

    • Ooops@feddit.org
      2·
      3 hours ago

      As someone with an ancient 750ti happily running on the regular nvidia drivers…

      Dedicated support for “older cards” as in “requiring different drivers” usually starts much later (Kepler and before), so about 4 generations before an 1660Super.

    • guynamedzero@piefed.zeromedia.vipOPEnglish
      1·
      15 hours ago

      I did consider trying that, but I’m just not sure how easy it would be to switch distros. I did make sure she set up her home directory as a separate partition, so that should make it easier, but the rest is still unfamiliar to me.

      • MagicShel@lemmy.zipEnglish
        3·
        2 hours ago

        I’ve had good luck keeping the home partition and just reinstalling the OS. Set it up with the same user name and home directory and you’re done.

      • brucethemoose@lemmy.world
        1·
        14 hours ago

        Also, random thing, but I did not get a notification for your reply.

        I don’t think that’s a piefed thing, as it happens a lot to me, even with other .world users.

      • brucethemoose@lemmy.world
        1·
        14 hours ago

        Nuke everything and start over? I guess you could keep the home directoy, but TBH I’d back it up and nuke it too, just in case.

        Or, as a shorter term solution, run it off a USB drive.

  • sudo@programming.dev
    3·
    15 hours ago

    Last time for me it was a bad CPU. Lived with it until I upgraded my CPU and recycled the old one into a new build. Then that one was having the same issue.

    • guynamedzero@piefed.zeromedia.vipOPEnglish
      1·
      14 hours ago

      Let’s say her CPU has been water cooled for a long, long time, without ever receiving maintenance, she’s had that computer for at least 7 years, and has never opened it once, and it was a hand-me-down from her dad, so I can’t say whether he took good care of it. Do you think that could have damaged it in some way?

      • sudo@programming.dev
        1·
        12 hours ago

        Idk about that. In my case I believe my CPU was defective from the start and I lived with it because I always assumed it was my OS in some way.

        If your CPU has seven years of not randomly freezing and its just doing that now then I wouldn’t suspect the CPU.

        However, unless you find some clues from journalctl -xeb1 or dmesg I would assume its faulty hardware somewhere.

      • Whostosay@sh.itjust.works
        1·
        13 hours ago

        Is it dusty as a mf in there? If so, yes. The fan very well could be stopped by dust and overheating the CPU or GPU which would cause a it to cut power and potentially freeze to attempt to stop damage to the component.

        I would at the very least open it and clean it, even if that isn’t the issue as it’s good to do regularly to increase longevity of the components. Make sure you don’t let the fans spin out of control while shooting air through them, it could damage them.

        Hope this helps.

        • sudo@programming.dev
          2·
          12 hours ago

          To build on this, it would help to install some sort of system monitoring to check temps, fanspeed, system usage and have those constantly going so OP can check for any red flags during a freeze.

  • potatoguy@lemmy.eco.br
    1·
    13 hours ago

    I had this in KDE (I have Nvidia too…), it was when I was sharing or downloading linux ISOs. If you see this problem when downloading or moving files (like KDE itself freezing, changing volume, but it staying the same, changing brightness, but it actually staying the same), I recommend going after another DE while using nvidia, because I could never figure out why it only freezes on KDE (I changed a lot of things over the years, only KDE gave me this problem).

  • e0qdk@reddthat.com
    1·
    14 hours ago

    Are you doing anything like network mounts? I’ve had freezes happen when home was mounted remotely. I’ve also had freezes before – though not recently – when doing extremely intensive I/O operations. I’ve also seen it with failing hardware, but I don’t remember what hardware was failing specifically (if we ever found out – that was an always-on display system at work, and we might have just replaced it since it was ancient at that point…).

    See if you can switch from GUI to console (either on the system or SSH-ing in). Usually the Ctrl-Alt-Fn keys switch between GUI and full screen console modes if you’re physically there to try it. If you can get in, you might be able to get some insight into what’s going on from logs, top, etc. while it’s happening.