[libvirt] [RFC] Create ptmx as a device

[Daniel P. Berrange]

On Thu, Sep 01, 2011 at 02:13:16PM +0800, Daniel Veillard wrote:
> On Wed, Aug 31, 2011 at 01:18:33PM -0500, Serge Hallyn wrote:
> > Hi,
> > 
> > I'm seeing an issue with udev and libvirt-lxc.  Libvirt-lxc creates
> > /dev/ptmx as a symlink to /dev/pts/ptmx.  When udev starts up, it
> > checks the device type, sees ptmx is 'not right', and replaces it
> > with a 'proper' ptmx.
> > 
> > In lxc, /dev/ptmx is bind-mounted from /dev/pts/ptmx instead of being
> > symlinked, so udev sees the right device type and leaves it alone.
> > 
> > A patch like the following seems to work for me.  Would there be
> > any objections to this?
> > 
> > >From 4c5035de52de7e06a0de9c5d0bab8c87a806cba7 Mon Sep 17 00:00:00 2001
> > From: Ubuntu <ubuntu at domU-12-31-39-14-F0-B3.compute-1.internal>
> > Date: Wed, 31 Aug 2011 18:15:54 +0000
> > Subject: [PATCH 1/1] make ptmx a bind mount rather than symlink
> > 
> > udev on some systems checks the device type of /dev/ptmx, and replaces it if
> > not as expected.  The symlink created by libvirt-lxc therefore gets replaced.
> > By creating it as a bind mount, the device type is correct and udev leaves it
> > alone.
> > 
> > Signed-off-by: Serge Hallyn <serge.hallyn at canonical.com>
> > ---
> >  src/lxc/lxc_container.c |   20 ++++++++++----------
> >  1 files changed, 10 insertions(+), 10 deletions(-)
> > 
> > diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c
> > index e425328..6991aec 100644
> > --- a/src/lxc/lxc_container.c
> > +++ b/src/lxc/lxc_container.c
> > @@ -543,18 +543,18 @@ static int lxcContainerPopulateDevices(void)
> >          }
> >      }
> >  
> > +    dev_t dev = makedev(LXC_DEV_MAJ_TTY, LXC_DEV_MIN_PTMX);
> > +    if (mknod("/dev/ptmx", S_IFCHR, dev) < 0 ||
> > +        chmod("/dev/ptmx", 0666)) {
> > +        virReportSystemError(errno, "%s",
> > +                             _("Failed to make device /dev/ptmx"));
> > +        return -1;
> > +    }
> > +
> >      if (access("/dev/pts/ptmx", W_OK) == 0) {
> > -        if (symlink("/dev/pts/ptmx", "/dev/ptmx") < 0) {
> > -            virReportSystemError(errno, "%s",
> > -                                 _("Failed to create symlink /dev/ptmx to /dev/pts/ptmx"));
> > -            return -1;
> > -        }
> > -    } else {
> > -        dev_t dev = makedev(LXC_DEV_MAJ_TTY, LXC_DEV_MIN_PTMX);
> > -        if (mknod("/dev/ptmx", S_IFCHR, dev) < 0 ||
> > -            chmod("/dev/ptmx", 0666)) {
> > +        if (mount("/dev/pts/ptmx", "/dev/ptmx", "ptmx", MS_BIND, NULL) < 0) {
> >              virReportSystemError(errno, "%s",
> > -                                 _("Failed to make device /dev/ptmx"));
> > +                                 _("Failed to bind-mount /dev/ptmx to /dev/pts/ptmx"));
> >              return -1;
> >          }
> >      }
> 
>   Hum, if we do a mount, I would expect to do an unmount somewhere.
> Also the lifetime of the mount and the symlink is really different,
> a recursive remove when destroying the container would lead to no
> resource leak but I think that for a bind mount we absolutely have
> to clean it up.

No need to worry about this, everything in 'lxc_container.c' is
executed inside the container filesystem namespace. When the last
process in the container exits, the kernel automatically cleans
up all the mounts for us.

Regards,
Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|