Re: [j-nsp] TTL in MPLS

From: Jared Mauch (jared@puck.nether.net)
Date: Sun Dec 16 2001 - 18:42:28 EST


        IMHO,

        Unless it's important for you to hide your network topology
or you have really old tcp/ip stacks (windows or other folks w/ low
default ttl) that you need to work around i would personally recommend
against it as you may cause yourself more problems when it
comes to debugging where your packets are actually traversing
(if you are running dynamic lsps it can be worse) when the network fails

        
        - jared

On Sun, Dec 16, 2001 at 06:25:24PM -0500, Mark M. Forest wrote:
> I am making some assumptions here so bear with me. I do not have sniffer
> resources free to check this right now, so I am posting for commentary.
>
> My understanding of MPLS TTL is that the Ingress router copies the TTL from
> the IP packet into the TTL field in the MPLS header. This MPLS TTL will
> decrement per LSP hop in the same manner as a typical IP packet. When the
> packet reaches the final hop and the label is popped then the TTL from the
> MPLS header replaces the one in the IP packet.
>
> I see where JUNOS has the capability to disable TTL or Not Propagate it as
> well. With this I assume that the IP TTL will decrement only once seeing the
> entire LSP as a single hop.
>
> I think I am good to go on this. I have read through the docs on this, but
> again look to the fols on this list for value added commentary anf/or
> personal insight into the subject. I know it is not that complex of a
> subject, nonetheless important to me.
>
> Thanks again for all you help....happy holidays to all
>
> milo
>
>
>
> _________________________________________________________
> Do You Yahoo!?
> Get your free @yahoo.com address at http://mail.yahoo.com

-- 
Jared Mauch  | pgp key available via finger from jared@puck.nether.net
clue++;      | http://puck.nether.net/~jared/  My statements are only mine.



This archive was generated by hypermail 2b29 : Mon Aug 05 2002 - 10:42:38 EDT