How can I kill a stuck LXC container?

I have a new unprivileged LXC container that I just cannot seem to stop. lxc-stop -n dl hangs. lxc-stop -k --nolock -n dl hangs. lxc-ls hangs. lxc-attach -n dl hangs. Nothing works on this container.

I have restarted lcxfs. All that did was upset a privileged container. This is what the current processes for this container look like (from the host).

lxc-start,24434 └─systemd,24446 --system --deserialize 9 ├─agetty,25258 --noclear --keep-baud console 115200 38400 9600 vt220 ├─agetty,25846 --noclear --keep-baud pts/1 115200 38400 9600 vt220 ├─agetty,25878 --noclear --keep-baud pts/3 115200 38400 9600 vt220 ├─agetty,25906 --noclear --keep-baud pts/2 115200 38400 9600 vt220 ├─agetty,25934 --noclear --keep-baud pts/0 115200 38400 9600 vt220 ├─rsyslogd,26016 -n │ ├─{rsyslogd},26043 │ └─{rsyslogd},26044 └─systemd-journal,25962

And the process states:

$ ps aux | grep -P '24434|24446|25258|25846|25878|25906|25934|25934|26016|26043|26044|25962'
oli 24434 0.0 0.0 43400 3552 ? Ss 10:56 0:00 [lxc monitor] /home/oli/.local/share/lxc dl
100000 24446 0.0 0.1 36896 4972 ? Ss 10:56 0:00 /lib/systemd/systemd --system --deserialize 9
100000 25258 0.0 0.0 15676 2504 pts/0 Ss+ 10:56 0:00 /sbin/agetty --noclear --keep-baud console 115200 38400 9600 vt220
100000 25846 0.0 0.0 15676 2496 pts/1 Ss+ 11:33 0:00 /sbin/agetty --noclear --keep-baud pts/1 115200 38400 9600 vt220
100000 25878 0.0 0.0 15676 2380 pts/3 Ss+ 11:33 0:00 /sbin/agetty --noclear --keep-baud pts/3 115200 38400 9600 vt220
100000 25906 0.0 0.0 15676 2416 pts/2 Ss+ 11:33 0:00 /sbin/agetty --noclear --keep-baud pts/2 115200 38400 9600 vt220
100000 25934 0.0 0.0 15676 2332 pts/0 Ss+ 11:33 0:00 /sbin/agetty --noclear --keep-baud pts/0 115200 38400 9600 vt220
100000 25962 0.0 0.0 35256 3500 ? Ss 11:33 0:00 /lib/systemd/systemd-journald
100104 26016 0.0 0.0 182640 3244 ? Ssl 11:33 0:00 /usr/sbin/rsyslogd -n

2 Answers

Force killing the container's systemd process worked for me.

In your example,

kill -9 24446

I keep this command line around to quickly kill all processes with a very high UID (which are in my container):

ps -ef | grep "^100[0-9][0-9][0-9]" | tr -s " " | cut -f2 -d " " | xargs -I {} kill -9 {}

Then I can use lxc-start again.

I also found out that I have less problems with lxc-stop when I use lxc-start -d (in background mode) and then connect using lxc-attach or ssh.

Your Answer

Sign up or log in

Sign up using Google Sign up using Facebook Sign up using Email and Password Submit

Post as a guest

Post Your Answer Discard

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy