<p>On Jul 18, 2011 2:54 PM, "Serge E. Hallyn" <<a href="mailto:serge.hallyn@canonical.com">serge.hallyn@canonical.com</a>> wrote:<br>
><br>
> Quoting Joerg Gollnick (<a href="mailto:code4lxc%2Blist@wurzelbenutzer.de">code4lxc+list@wurzelbenutzer.de</a>):<br>
> > Hello Serge,<br>
> > I think that the main point is the initial setup of the cgroup (directory)<br>
> > structure.<br>
> ><br>
> > systemd<br>
> > tmpfs on /sys/fs/cgroup type tmpfs (rw,nosuid,nodev,noexec,relatime,mode=755)<br>
> > cgroup on /sys/fs/cgroup/systemd type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,release_agent=/lib/systemd/systemd-cgroups-<br>
> > agent,clone_children,name=systemd)<br>
> > cgroup on /sys/fs/cgroup/cpuset type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,cpuset,clone_children)<br>
> > cgroup on /sys/fs/cgroup/ns type cgroup (rw,nosuid,nodev,noexec,relatime,ns)<br>
> > cgroup on /sys/fs/cgroup/cpu type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,cpu,clone_children)<br>
> > cgroup on /sys/fs/cgroup/cpuacct type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,cpuacct,clone_children)<br>
> > cgroup on /sys/fs/cgroup/memory type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,memory,clone_children)<br>
> > cgroup on /sys/fs/cgroup/devices type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,devices,clone_children)<br>
> > cgroup on /sys/fs/cgroup/freezer type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,freezer,clone_children)<br>
> > cgroup on /sys/fs/cgroup/net_cls type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,net_cls,clone_children)<br>
> > cgroup on /sys/fs/cgroup/blkio type cgroup<br>
> > (rw,nosuid,nodev,noexec,relatime,blkio,clone_children)<br>
><br>
> That looks an awful lot like the default setup with cgroup-bin installed on<br>
> a ubuntu oneiric upstart system. Actually, I see ns cgroup is mounted<br>
> (separately). If you can find a way to not have that mounted, that may<br>
> solve the issue.<br>
><br>
> I wonder if systemd actually uses ns cgroup (perhaps to lock consoles into a<br>
> cgroup)?</p>
<p>I don't believe it does -- not certain though.</p>
<p>IIRC from some of Lennert's writings/posts they only mount them all so they are available? Perhaps in case a unit needs it ... if another process were to mount ahead of time systemd might not be able to fulfill the unit's request properly.</p>
<p>By default I don't think systemd even uses any of the cgroups except the `name` group for process tracking, but I remember some rants about the fact that you can't really discover new groups easily, and a bare mount of cgroup blindly mounts all subsystems.</p>
<p>C Anthony [mobile] </p>