The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-Oct> msg00003



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

MPLS/BGP routing question

  • From: GUESDON Herve FTRD/DAC/ISS <herve.guesdon@rd.francetelecom.fr>
  • Date: Mon, 2 Oct 2000 08:04:28 +0200
  • Cc: mpls@UU.NET, "'nbvpn@bbo.com'" <nbvpn@bbo.com>
  • X-MIME-Autoconverted: from quoted-printable to 8bit by cell.onecall.net id JAA00020

Hi

If a network event breaks the MPLS LSP between two PE in a BGP/MPLS VPN
design, then I agree that the rerouting is an IGP issue. But there is also a
MP-BGP-MPLS synchronisation issue if this rerouting leads to a change in the
MP-BGP next hop. The switch to the LSP corresponding to the new MP-BGP net
hop has to be transparent to the customer VPN traffic.

Regards

hervé

****************************************************************
Hervé Guesdon                                    DAC/CPN/RRI
Research and Development Engineer
IP Routing and VPN lab
France Télécom - R&D
38-40 rue du Général-Leclerc  
92794 Issy Moulineaux Cedex 9  France
phone : +33.1.45.29.43.74  fax : +33.1.45.29.54.11
****************************************************************

"L'avenir c'est du passé en préparation."


>-----Message d'origine-----
>De : S.Matsushima [mailto:satoru@japan-telecom.co.jp]
>Envoyé : samedi 30 septembre 2000 03:24
>À : erosen@cisco.com; Satoru Matsushima
>Cc : mpls@UU.NET
>Objet : Re: MPLS/BGP routing question 
>
>
>on 00.9.30 0:32 AM, Eric Rosen at erosen@cisco.com wrote:
>
>> 
>> Matsushima> When a  LSP of PE to  PE broken, BGP  has no way 
>of  LSP broken.
>> Matsushima> Then, BGP keep  up of VPN routes and VPN  
>traffic going to black
>> Matsushima> hole, until LSP available.
>> 
>> Matsushima> As a result,  VPN customer can not back up  
>their traffic to any
>> Matsushima> link.
>> 
>> Matsushima> I think this is one of most seriously problem of 
>BGP/MPLS VPN.
>> 
>> The  situation you  are worried  about is  where there  is  
>IGP connectivity
>> between the edges,  but for some reason labeled packets  
>cannot make it from
>> one edge to another.
>
>Yes, exactly.
>
>> I guess we don't really see this as a realistic failure
>> scenario.  Sure, buggy software could cause this, but 
>there's a million ways
>> in which buggy software could cause undetected packet loss.
>> 
>
>I think that LSP failure was caused by not only buggy software but also
>oparation failure.
>For example, i) erase a interface as LDP ID ;-< , ii) routes 
>summarization
>failure on ospf area,...
>
>IMO, BGP which on PE should has some way to know of LSP failure.
>This is for customer.
>
>--
>Satoru Matsushima
>