But from the output it shows that is mounting a fake cgroupfs, no?
Hello,
Have exactly the same issue after upgrade from 3.21 to 3.22 (on arm64). Tried both versions from snap (rev 13718 3.22/stable and rev 13764 3.22/candidate).
Inside containers I see only 1 process init:
root@ns01:~# ps aux
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 0.0 0.0 12116 3068 ? Ss 14:13 0:00 /sbin/init
root 205 0.0 0.0 7300 3364 ? Ss 14:13 0:00 bash
root 215 0.0 0.0 8680 2764 ? R+ 14:13 0:00 ps aux
In logs:
systemd 237 running in system mode. (+PAM +AUDIT +SELINUX +IMA +APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD -IDN2 +IDN -PCRE2 default-hierarchy=hybrid)
Detected virtualization lxc.
Detected architecture arm64.
Welcome to Ubuntu 18.04.1 LTS!
Set hostname to <ns01>.
Failed to create compat systemd cgroup /lxc.payload/ns01/init.scope: Invalid argument
Failed to attach 1 to compat systemd cgroup /lxc.payload/ns01/init.scope: No such file or directory
Failed to open pin file: No such file or directory
Failed to allocate manager object: No such file or directory
[ESC[0;1;31m!!!!!!ESC[0m] Failed to allocate manager object, freezing.
Freezing execution.
Set to “security.privileged” and server reboot didn’t help.
Tentative fix merged in lxcfs and being included in the candidate snap now.
We’ll validate on CentOS 7 before pushing to stable later today.
Seems to work fine here. Just need to wait for arm64 to be done building before releasing to stable.
Rolled out to stable.
Confirmed, arm64 version
stable: 3.22 2020-03-16 (13771) 62MB -
works now without issues.
Thanks for fix that.
Good to hear, sorry about that.
Thanks for the fix.