[lxc-users] sysvinit with cgroup namespace

[Harald Dunkel]

On 04/06/2016 05:18 PM, Serge Hallyn wrote:
> Quoting KATOH Yasufumi (karma at jazz.email.ne.jp):
>>
>> Will we be able to start a container on sysvinit with cgroup namespace
>> in the future release?
> 
> mkdir /sys/fs/cgroup
> mount -t cgroup -o none,name=systemd systemd /sys/fs/cgroup
> 

Or was it

	sudo mkdir /sys/fs/cgroup/systemd
	sudo mount -t cgroup -o none,name=systemd /sys/fs/cgroup/systemd

?

> You could argue that this shouldn't be needed, but if you
> don't do it you won't be able to start any systemd-based
> containers, and I'd rather bail out early with a clear error
> and simple fix, rather than get cryptic reports about certain
> containers not working.

Unfortunately the error message is hidden, unless you run

	lxc-start -n mylxc -F

Without -F it just fails. There will be reports about broken
systemd containers for LXC, still. Maybe lxc-checkconfig
should complain?


I feel *very* bad about LXC becoming systemd-only.

Regards
Harri