If you're just trying to fix the non-deterministic behavior, use

bgp deterministic-med

what this does is, order the oaths by peer-AS, then select the best path
in each group then find an overall bestpath from comparing the winners in 
each group. This will assure that the same path is always prefered regardless
of the order it was received in.


Patrick

> >
> >
> >I'm researching some strange behavior with some non-deterministic
> >path selection with BGP on a few routers. I thumbed through Radia
> >Perlman's second edition of Interconnections at the bookstore last
> >night and saw that in her paragraph where she noted some problems
> >with BGP, she referenced a whitepaper which described situations
> >where timing of updates can cause non-deterministic path selection.
> >I've been searching for that whitepaper, but I am just going off of
> >vague memories of the title. If anyone happens to have their copy
> >handy and could email me with the title of the whitepaper, I would
> >appreciate it. I unfortunately have a meeting this morning before
> >the book store opens again 8).
> >
> 
> There's a substantial amount of research in this area, and not yet a 
> consensus. These references are definitely research-oriented, beyond 
> what many people on the list call "theory" .  The names and keywords 
> below should help you find them with search engines.
> 
> Look at some of the back meeting presentations at www.merit.org, 
> where there are  Internet routing convergence studies by Craig 
> Labovits, Abba Ahuja, and others.
> 
> Some of the first work in weak synchronization of routing protocols, 
> initially with respect to RIP, is from Sally Floyd.  I'd have to look 
> at her dissertation to see if Radia Perlman considers timing part of 
> the Byzantine Corruption Problem.
> 
> Look through the Internet Drafts directory and look for network layer 
> and routing workshops.
> 
> Vern Paxson's paper and dissertation "End-to-end packet dynamics in 
> the Interernet."
> 
> A recent publication by Varadhan, Govindan, and Estrin on "Persistent 
> Route Oscillations in Inter-Domain Routing." I got this by navigating 
> to www.isi.edu/dev7 and going to Deborah Estrin's publications.
> 
> ___________________________________
> UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
> FAQ, list archives, and subscription info: http://www.groupstudy.com
> Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
> 

___________________________________
UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
FAQ, list archives, and subscription info: http://www.groupstudy.com
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to