Re: [PATCH v2] nbd/server: Add --selinux-label option

[Daniel P . Berrangé]

On Wed, Aug 25, 2021 at 02:35:04PM -0500, Eric Blake wrote:
> On Fri, Jul 23, 2021 at 05:38:06PM +0100, Daniel P. Berrangé wrote:
> > On Fri, Jul 23, 2021 at 06:18:55PM +0200, Kevin Wolf wrote:
> > > Am 23.07.2021 um 12:33 hat Richard W.M. Jones geschrieben:
> > > > Under SELinux, Unix domain sockets have two labels.  One is on the
> > > > disk and can be set with commands such as chcon(1).  There is a
> > > > different label stored in memory (called the process label).  This can
> > > > only be set by the process creating the socket.  When using SELinux +
> > > > SVirt and wanting qemu to be able to connect to a qemu-nbd instance,
> > > > you must set both labels correctly first.
> > > > 
> > > > For qemu-nbd the options to set the second label are awkward.  You can
> > > > create the socket in a wrapper program and then exec into qemu-nbd.
> > > > Or you could try something with LD_PRELOAD.
> > > > 
> > > > This commit adds the ability to set the label straightforwardly on the
> > > > command line, via the new --selinux-label flag.  (The name of the flag
> > > > is the same as the equivalent nbdkit option.)
> > > > 
> > > > A worked example showing how to use the new option can be found in
> > > > this bug: https://bugzilla.redhat.com/show_bug.cgi?id=1984938
> > > > 
> > > > Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1984938
> > > > Signed-off-by: Richard W.M. Jones <rjones@redhat.com>
> > > 
> > > I suppose this would also be relevant for the built-in NBD server,
> > > especially in the context of qemu-storage-daemon?
> > 
> > It depends on the usage scenario really. nbdkit / qemu-nbd are
> > not commonly run under any SELinux policy, so then end up being
> > unconfined_t. A QEMU NBD client can't connect to an unconfined_t
> > socket, so we need to override it with this arg.
> > 
> > In the case of qemu system emulator, under libvirt, it will
> > already have a svirt_t type, so in that case there is no need
> > to override the type for the socket.
> > 
> > For qsd there's not really any strong practice established
> > but i expect most current usage is unconfined_t too and
> > would benefit from setting label.
> > 
> > > If so, is this something specific to NBD sockets, or would it actually
> > > make sense to have it as a generic option in UnixSocketAddress?
> > 
> > It is applicable to inet sockets too in fact.
> 
> So now that 6.2 is open, should I queue the patch as is, or wait for a
> v3 that makes the option more generic to all socket usage?

My gut feeling is that it makes sense to have a more generic option,
with the selinux label specified in the "SocketAddress" QAPI type,
and then have util/qemu-sockets.h be setting the context in
socket_listen().

I don't think this invalidates the patch that Richard proprosed
here, as we'll still need the command line argument he's added.
All that will differ is that the setsockcreatecon_raw will get
moved down.

So from that POV, I don't think we need the general solution to
be a blocker, it can be additive.

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|