Am 26.01.21 um 13:53 schrieb [email protected]:
Hi Michael,

Shouldn't "/sys/fs/cgroup/unified" be mounted by systemd to be used lxc
to use cgroupv2?

Should we need to do "mount -t tmpfs cgroup /sys/fs/cgroup" for
cgroupv2 as well? Will it be done systemd ?

In cgroupv1, it is usually seen "/sys/fs/cgroup" mounted on tmpfs. In
case of cgroupv2 unified, nothing like that is seen.


df -h
Filesystem      Size  Used Avail Use% Mounted on
udev            3.4G     0  3.4G   0% /dev
tmpfs           692M  1.7M  691M   1% /run
/dev/sda8       125G   57G   62G  48% /
tmpfs           3.4G     0  3.4G   0% /dev/shm
tmpfs           5.0M  4.0K  5.0M   1% /run/lock
/dev/sda1        96M   62M   35M  65% /boot/efi
tmpfs           692M  212K  692M   1% /run/user/1000



Please note that lxc and lxcfs is malfunctioning for fresh installs
(Linux coffee 5.10.0-1-amd64 #1 SMP Debian 5.10.4-1 (2020-12-31) x86_64
GNU/Linux) as well. Not sure which package is having problem. But
its(lxc) not working.

I can not confirm that.
lxc is working fine here ootb with cgroupv2.
No changes necessary, no manual mounts or anything.

The only known issue I'm aware of, is if you want to run jessie containers on a bullseye/sid host. Since jessie's systemd does not yet support cgroupv2, you need to switch the host back to unified if you want to use such containers.

Maybe you have some custom config, which breaks lxc. The lxc maintainers will be in a better situation to help you debug that.

Regards,
Michael



Attachment: OpenPGP_signature
Description: OpenPGP digital signature

Reply via email to