Computer fails to reboot "watchdog did not stop" "waiting for process systemd-udevd"

Troubleshooting
1

System Information

  • DietPi version
    G_DIETPI_VERSION_CORE=8
    G_DIETPI_VERSION_SUB=16
    G_DIETPI_VERSION_RC=2
    G_GITBRANCH=‘master’
    G_GITOWNER=‘MichaIng’
    G_LIVE_PATCH_STATUS[0]=‘not applicable’
  • Distro version: bullseye
  • Kernel version: Linux DietPi 5.10.0-21-amd64 #1 SMP Debian 5.10.162-1 (2023-01-21) x86_64 GNU/Linux
  • Architecture: amd64
  • SBC model: Native PC (x86_64)

System reboots on a cronjob at 6am every day and sometimes it just fails to reboot. It happens randomly maybe 1-2 times per month and I can’t SSH to the system when it happens so I had to connect a monitor and wait for the problem to happen again. Took a picture of the last output, which showed the same thing both times it happened so far:

339924405_244817491405765_6034307996588466955_n
339924405_244817491405765_6034307996588466955_n1296×1728 336 KB

Anyone’s got any idea what this does specifically? Checked Google but the information I found wasn’t really helping with my level of knowledge. Is it possible it’s a kernel bug as per this post? Later kernel version though.

NOTE: No USB flashdrive connected (only a USB keyboard) and there is no dedicated GPU (There was one before but I removed it a while ago)

2

Please enable udev debug logging to hopefully get more output when this happens:

G_CONFIG_INJECT 'udev_log=' 'udev_log=debug' /etc/udev/udev.conf

This adds quite a large number of system logs (especially at boot), so might be worth it to try triggering the issue directly with some reboots or so.

3

Thanks friend,

Just came here to say I still don’t have any info yet since the freeze doesn’t happen often, but I’ll report back as soon as I get it :slight_smile:

1 Like
4

If it somehow doesn’t freeze anymore, even better. Otherwise jep, just report back with hopefully additional logs when it happens again.

5

Hello again,

server didn’t reboot again this morning but the error messages seem exactly like the old one, no more info:

image
image1128×2000 195 KB

Shouldn’t the command have been:

G_CONFIG_INJECT 'udev_log=debug' /etc/udev/udev.conf

instead?

The content of /etc/udev/udev.conf at the moment is:

# see udev.conf(5) for details
#
# udevd is also started in the initrd.  When this file is modified you might
# also want to rebuild the initrd, so that it will include the modified configuration.

#udev_log=info
#children_max=
#exec_delay=
#event_timeout=180
#timeout_signal=SIGKILL
#resolve_names=early
udev_log=

And I would have thought the level should be debug or info

1 Like
6

Ah, indeed I switched the arguments, it should have been (also corrected above):

G_CONFIG_INJECT 'udev_log=' 'udev_log=debug' /etc/udev/udev.conf

And then yes, the result should be that it contains udev_log=debug. “info” is the default, which does not reveal sufficient info, so we need debug logs.

7

Gotcha thanks haha! Fixed it, will report back in a few weeks/months when the problem happens again XD

1 Like