The MPLS WG Archive

[Grenville Armitage <gja@research.bell-labs.com>]

robert,

aside from the (lack of)ethics associated with reposting
private email back to the list, please at least do me the
courtesy of taking my comments in context. quite clearly
i made a restricted observation - viz. when the LSPs are up
they essentially function as tunnels since the inside
packets are isolated from the intermediate LSRs' forwarding
machinery.

gja

Robert Raszuk wrote:
> 
> Grenville,
> 
> > on what basis an LSP is established
> > doesn't really change how it behaves
> 
> >From operational expereince with them for a decent time I see that it
> really does change how it behavies depending on how it got setup/build.
> Of course in steady state both are fine and doing the same, but when
> things start to melt or break that is when you see big differences.
> 
> R.
> 
> > Grenville Armitage wrote:
> >
> > Robert Raszuk wrote:
> >         [..]
> > > I just don't know anymore
> > > what one means by term "mpls tunnel".
> >
> > generically an LSP is always a tunnel of sorts, since the
> > IP header of the MPLS payload is thereby always hidden from
> > the intermediate LSR hops.  on what basis an LSP is established
> > doesn't really change how it behaves as a form of 'tunnel'
> > in the data path....
> >
> > gja

-- 
________________________________________________________________________
Grenville Armitage                    http://members.home.net/garmitage/
Bell Labs Research Silicon Valley