The MPLS-OPS Archive

[Marcelo Blanes <mblanes@larc.usp.br>]

Robert,

Here goes my comments and concerns :) ..

On Fri, 19 Oct 2001, Robert Raszuk wrote:

> Marcelo,
> 
> Here goes my answer ;) 
> 
> > In a MPLS environment router will not fragment IP packets. So let's
> 
> Not true. We can fragment even in P routers but this is process switched
> and HIGHLY not recommended !

I have PE - P - PE router. If I do a ping in PE to PE if datagram size for
example 2000 (PE to P router has ether interface if tag mut 1526) with DF
bit not set, I got I timeout answer. For 1500 it work fine.

 > 
> > suppose that you have a MPLS backbone where MTU is 1500 and 4470.
> 
> OK. 
> 
> > If you are using BGP and BGP messages can be larger than 1500 ( the
> > maximum BGP message is around 4K ). Suppose that you are sending a IP
> > packet from a router that supports a large MTU value to a router that
> > supports a small MTU value carring BGP message. The packet will be
> > dropped?!
> 
> You have to make sure that the first router already fragments to a
> lowest MTU in your core. tag mtu 1500 for example if your max MTU is
> 1500 anywhere in the core.

Ok Robert, but iBGP messages ( session VPNv4 ) is exchange between
loopacks. Will iBGP looks to MTU in the loopback?!

> 
> The right solution is to get rid of 1500 MTU in the core. Most of good
> ethernet switches support giant packets, The same goes for ethernet
> interfaces on the routers. 

OK, although I dont have L2 devices between any router.

Marcelo.

> 
> R.
> 
> > Marcelo Blanes wrote:
> > 
> > Hi,
> > 
> > Here goes my question.
> > 
> > In a MPLS environment router will not fragment IP packets. So let's
> > suppose that you have a MPLS backbone where MTU is 1500 and 4470.
> > 
> > If you are using BGP and BGP messages can be larger than 1500 ( the
> > maximum BGP message is around 4K ). Suppose that you are sending a IP
> > packet from a router that supports a large MTU value to a router that
> > supports a small MTU value carring BGP message. The packet will be
> > dropped?!
> > 
> > Thanks,
> > Marcelo.
> > 
> > -------
> > The MPLS-OPS Mailing List
> > Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
> > Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
> 

-------
The MPLS-OPS Mailing List
Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
Archive: http://www.mplsrc.com/mpls-ops_archive.shtml