[j-nsp] Juniper best path selection process

[Richard A Steenbergen]

On Thu, Apr 01, 2010 at 05:53:51PM -0500, Joe Metzger wrote:
> Does anybody have a reference to the real best path selection process
> actually used by an M7i running 9.3 that documents all of the
> tie-breaking criteria, and when it might decide to just blow off
> recomputing a new best route if the important attributes didn't
> change?

What you see documented is what it should be doing, unless there is a 
bug. The only way that Juniper should prefer paths based on age is if 
you configure "path-selection cisco-non-deterministic". 

http://www.juniper.net/techpubs/software/junos/junos93/swconfig-routing/how-the-active-route-is-determined.html

Not to say that there haven't been a couple bugs in the path selection 
algorithm over time, but all the ones I know of were fixed by 9.3.

> The underlying system feeding routes to the 2 black hole
> servers guarantees that both servers are sending identical
> lists of prefixes, with identical local-pref, as-path,
> origin code, med value, source protocol, etc. The import
> policy explicitly sets the next-hop to discard for all
> routes from both servers so IGP distance shouldn't be a factor.

One of the best features of JUNOS is it will tell you why a route is or
isn't active. Look at "show route detail" on a route you believe is 
incorrecting picking a wrong path and see for yourself. :)

-- 
Richard A Steenbergen <ras at e-gerbil.net>       http://www.e-gerbil.net/ras
GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)