Re: revised draft of BGP-4

Dennis Ferguson <dennis@mci.net> Thu, 15 June 1995 19:11 UTC

Message-Id: <199506151857.AA33789@interlock.ans.net>
To: Paul Traina <pst@cisco.com>
Cc: curtis@ans.net, "Michael F. Nittmann" <nittmann@wis.com>, Yakov Rekhter <yakov@cisco.com>, bgp@ans.net
Subject: Re: revised draft of BGP-4
In-Reply-To: Your message of "Thu, 15 Jun 1995 10:24:22 PDT." <199506151724.AA34347@interlock.ans.net>
Date: Thu, 15 Jun 1995 14:56:57 -0400
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Dennis Ferguson <dennis@mci.net>

Paul,

>   Partition repair (accepting your AS in the path exactly once) can be
>   configured but should not be the default.  If you have an agreement
>   with your provider to do partition repair, then configure it.
>
> Which is precicely what the BGP specification allows one to do, and if
> you have out-of-band information to eliminate static routing loops, have
> a party.

You don't need any out-of-band information to terminate routing loops,
it is sufficient that the BGP correctly update the routes it readvertises
and that it decline to use routes in which its AS path appears N or
more times.  N==1 is a good default but setting N>1 does not affect
the correctness of converged routing.  It just makes the transient
sloppier (a drawback which one might wish to correct with policy based
on out-of-band knowledge, to be sure).

Dennis

revised draft of BGP-4 Yakov Rekhter
Re: revised draft of BGP-4 Vadim Antonov
Re: revised draft of BGP-4 Paul Traina
Re: revised draft of BGP-4 Michael F. Nittmann
Re: revised draft of BGP-4 Paul Traina
Re: revised draft of BGP-4 Curtis Villamizar
Re: revised draft of BGP-4 Michael F. Nittmann
Re: revised draft of BGP-4 Paul Traina
Re: revised draft of BGP-4 Dennis Ferguson