[lxc-devel] [Not A Patch] [POC] Proof of concept code for using devtmpfs for autodev and more...

[Serge Hallyn]

Quoting Michael H. Warfield (mhw at WittsEnd.com):
> On Fri, 2013-11-01 at 17:18 -0500, Serge Hallyn wrote: 
> > Quoting Michael H. Warfield (mhw at WittsEnd.com):
> > > On Thu, 2013-10-31 at 13:00 -0500, Serge Hallyn wrote: 
> > > > Quoting Michael H. Warfield (mhw at WittsEnd.com):
> > > > > I did incorporate your suggestion of using the hash of the rootfs path
> > > > > as the subdirectory under the hosts /dev/ for the container.  I also
> > > > 
> > > > (Printed this out to look it over, just putting all my comments together
> > > > here) :
> > > > 
> > > > 1. I think if /dev is not devtmpfs, we should just bail on this.
> > > > 
> > > > 2. You say in comments that you're using the cgroup name, but it seems
> > > >    you're actually just using the container name?
> > > > 
> > > > 3. The cgroup name used to be unique, but now each mounted cgroupfs
> > > >    can actually have a different name for the same container (if some
> > > >    of them didn't get cleaned out well).
> > > 
> > > Maybe I misunderstood something here but I thought you told me at
> 
> > You did (or I wasn't clear :).  The container names need to be unique within
> > a given lxcpath.  So you can
> > 
> > 	lxc-create -t fedora -n Fedora19
> > 	lxc-create -t fedora -n Fedora19 -P /opt/lxc1
> > 	for i in `seq 2 201; do
> > 		lxc-create -t fedora -n Fedora19 -P /opt/lxc$i
> > 	done
> 
> > and start them all by passing -P <lxcpath> to lxc-start.
> 
> Yeah, that worked.  Creates some, errr, cough, interesting
> ambiguities...
> 
> [root at hydra ~]# lxc-ls 
> Alcove   Audience  CentOS6  Faces     Oracle Suse Y2
> Alpine   BusyBox   Chaos    Fedora19  Platform  Ubuntu  Yelm
> Anteroom  CentOS5   Debian   Fourier   Plover Vault
> [root at hydra ~]# lxc-ls --active
> Alcove   Audience  Faces     Fedora19-1  Oracle    Plover  Vault  Yelm
> Anteroom  Chaos     Fedora19  Fourier   Platform  Suse    Y2
> [root at hydra ~]# 
> 
> So, Fedora19-1 shows up in "lxc-ls --active" but not in plain "lxc-ls".
> So it's an active container but not a container???  Interesting...

That should definately not be happening.  The lxcpath is encoded
in the abstract unix socket for controlling the container.  So
I'm guessing lxc-ls --active is looking at cgroups and doing the
wrong thing there.  Bugger.