KMonads crashes frequently

[MaxGyver83]

Recently, KMonad crashes very often but only for one external (USB) keyboard. That's why I'm running KMonad in an endless loop now.
Today I had 278 crashes in 9.5 hours. In most cases, I get these two error messages (for 2 subsequent restarts):

kmonad: /dev/input/by-id/usb-SONiX_USB_DEVICE-event-kbd: openFd: does not exist (No such file or directory)
kmonad: Could not perform IOCTL release on: /dev/input/by-id/usb-SONiX_USB_DEVICE-event-kbd

This might be a hardware problem, as these crashes don't happen with my other keyboard. Do you have an idea what could be the reason? Maybe the keyboard gets disconnected (although it's connected via a USB cable)?

[slotThe]

On Thu, Jan 28 2021 11:53, Max Schillinger wrote: ``` kmonad: /dev/input/by-id/usb-SONiX_USB_DEVICE-event-kbd: openFd: does not exist (No such file or directory) kmonad: Could not perform IOCTL release on: /dev/input/by-id/usb-SONiX_USB_DEVICE-event-kbd ``` This might be a hardware problem, as these crashes don't happen with my other keyboard. Do you have an idea what could be the reason? Maybe the keyboard gets disconnected (although it's connected via a USB cable)?

Yes, this does indeed seem like a hardware problem; the keyboard gets disconnected, kmonad tries to access the respective (now non-existant) `dev/input' directory, can't find it, and throws an exception. I can reproduce this by connecting a USB keyboard, starting kmonad, and then disconnecting that keyboard.

[MaxGyver83]

OK, then it's not caused by KMonad. This is the corresponding part in my syslog, btw:

Jan 28 21:55:39 kernel: [48518.879974] usb 3-1.1-port4: disabled by hub (EMI?), re-enabling...
Jan 28 21:55:39 kernel: [48518.880131] usb 3-1.1.4: USB disconnect, device number 19
Jan 28 21:55:39 acpid: input device has been disconnected, fd 7
Jan 28 21:55:39 upowerd[3328]: unhandled action 'unbind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.0/0003:05AC:024F.0109
Jan 28 21:55:39 upowerd[3328]: unhandled action 'unbind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.0
Jan 28 21:55:39 acpid: input device has been disconnected, fd 27
Jan 28 21:55:39 acpid: input device has been disconnected, fd 8
Jan 28 21:55:39 upowerd[3328]: unhandled action 'unbind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.1/0003:05AC:024F.010A
Jan 28 21:55:39 upowerd[3328]: unhandled action 'unbind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.1
Jan 28 21:55:39 upowerd[3328]: unhandled action 'unbind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4
Jan 28 21:55:39 kernel: [48519.280240] input: KMonad output as /devices/virtual/input/input588
Jan 28 21:55:39 acpid: input device has been disconnected, fd 7
Jan 28 21:55:40 kernel: [48519.364643] usb 3-1.1.4: new full-speed USB device number 20 using xhci_hcd
Jan 28 21:55:40 kernel: [48519.490046] usb 3-1.1.4: New USB device found, idVendor=05ac, idProduct=024f
Jan 28 21:55:40 kernel: [48519.490049] usb 3-1.1.4: New USB device strings: Mfr=1, Product=2, SerialNumber=0
Jan 28 21:55:40 kernel: [48519.490051] usb 3-1.1.4: Product: USB DEVICE
Jan 28 21:55:40 kernel: [48519.490053] usb 3-1.1.4: Manufacturer: SONiX
Jan 28 21:55:40 kernel: [48519.493143] input: SONiX USB DEVICE as /devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.0/0003:05AC:024F.010B/input/input589
Jan 28 21:55:40 kernel: [48519.553276] apple 0003:05AC:024F.010B: input,hidraw4: USB HID v1.11 Keyboard [SONiX USB DEVICE] on usb-0000:3e:00.0-1.1.4/input0
Jan 28 21:55:40 kernel: [48519.554146] apple 0003:05AC:024F.010C: Fn key not found (Apple Wireless Keyboard clone?), disabling Fn key handling
Jan 28 21:55:40 kernel: [48519.554187] input: SONiX USB DEVICE as /devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.1/0003:05AC:024F.010C/input/input590
Jan 28 21:55:40 kernel: [48519.612869] apple 0003:05AC:024F.010C: input,hidraw5: USB HID v1.11 Keyboard [SONiX USB DEVICE] on usb-0000:3e:00.0-1.1.4/input1
Jan 28 21:55:40 mtp-probe: checking bus 3, device 20: "/sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4"
Jan 28 21:55:40 mtp-probe: bus: 3, device: 20 was not an MTP device
Jan 28 21:55:40 upowerd[3328]: unhandled action 'bind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.0/0003:05AC:024F.010B
Jan 28 21:55:40 upowerd[3328]: unhandled action 'bind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.0
Jan 28 21:55:40 upowerd[3328]: unhandled action 'bind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.1/0003:05AC:024F.010C
Jan 28 21:55:40 upowerd[3328]: unhandled action 'bind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4/3-1.1.4:1.1
Jan 28 21:55:40 upowerd[3328]: unhandled action 'bind' on /sys/devices/pci0000:00/0000:00:1d.6/0000:06:00.0/0000:07:02.0/0000:3e:00.0/usb3/3-1/3-1.1/3-1.1.4
Jan 28 21:55:40 kernel: [48519.694742] input: KMonad output as /devices/virtual/input/input591

Maybe some power saving function!? Nevermind, I'm closing this.

[bard]

@MaxGyver83 FYI there's a systemd unit at https://github.com/david-janssen/kmonad/blob/master/startup/kmonad.service that will restart on exit. It won't solve the problem but at least it would save the trouble of managing a loop in a shell.

[MaxGyver83]

OK, meanwhile I found the root cause. I googled the first line of the syslog posted above: EMI stands for Electromagnetic interference and there are interesting stories about what could cause a USB hub to be disabled, like turning off an exhaust fan in the adjacent bath room. After reading this, I recognized a pattern. The USB port disconnects (and KMonad crashes) every time I lean towards the screen (only!) while standing up from my office chair. I would never have thought about that. I guess I need better isolated cables...

@bard : Thanks for you comment. I hope that I don't need this after changing the USB cable.

[david-janssen]

Just popping into the old issue to say: haha! Posture-induced bug, great fun :-)