On Fri, Apr 18, 2008 at 02:31:40PM -0400, Janne Karhunen wrote:
> On Fri, Apr 18, 2008 at 2:25 PM, J. Bruce Fields <[email protected]> wrote:
>
> > > /*
> > > * If a local hostname is given (-n option to statd), bind to
> > > the address
> > > * specified. This is required to support clients that ignore
> > > the mon_name in
> > > * the statd protocol but use the source address from the
> > > request packet.
> > > */
> >
> > Weird--I dont' see that comment.
>
> All original nfs-utils-1.0.12 tarball; indeed strange.
I was looking at recent git. Note the most recent tarball on
sourceforge.net is 1.1.2, and that has the relevant changes.
--b.
>
> But to be honest, I don't instantly understand how
> this would work without -n at all (apart blind luck).
> INADDR_ANY has no guarantees..
>
>
> --
> // Janne
On Fri, Apr 18, 2008 at 2:34 PM, J. Bruce Fields <[email protected]> wrote:
> I was looking at recent git. Note the most recent tarball on
> sourceforge.net is 1.1.2, and that has the relevant changes.
There you go, against 1.1.2.
--
// Janne
On Fri, Apr 18, 2008 at 2:34 PM, J. Bruce Fields <[email protected]> wrote:
> > > Weird--I dont' see that comment.
> >
> > All original nfs-utils-1.0.12 tarball; indeed strange.
>
> I was looking at recent git. Note the most recent tarball on
> sourceforge.net is 1.1.2, and that has the relevant changes.
So what has been done in 1.1.2 is that the
-n binding has been removed completely.
It would _probably_ work, but only if I'm lucky.
This way it will not only try to guess the interface
to be used (and may or may not get it right). That,
and it will expose NFS services on interfaces
that should not need them.
I'll rework the patch to 1.1.2 to see how well
it fits in.
--
// Janne