[isp-bgp] Re: BGP metric weight issue

[Tycho Eggen <tycho@...]

Hi,

when trying to adjust preferred egress connections,
one can use route-map like the one below.

If you make a prefix-list or access-list and use it in the route-map,
you can certainly change egress preferences.

However, a nifty feature of BGP is the local-preference part.

Weight is a local Cisco proprietory BGP attribute.
Local in the sense that it's only working on the router with the egress connection.
It's advisable to use local-preference,
because this is an standard BGP attribute and thus gets distributed through all bgp connections.

I think the config ought to look a bit like the following:

## Example configuration - Start ## 

ip prefix-list genuity-egress-in seq 10 permit 208.57.25.1/32

route-map genuity-in permit 10
 set local-preference 110
 match ip address prefix-list genuity-egress-in

router bgp <as#>
 neighbour <ip> remote-as <as#>
 neighbour <ip> route-map genuity-in in 
end

## Example configuration - End ## 


Just my EUR 0.02 ;-)

Grtz,	
	Tycho Eggen


On Wed, Jan 08, 2003 at 02:30:00PM -0600 William R. Charnock(will@...) claimed:
> Try using a route-map to adjust the MED's and localprefs as you want them:
> 
> !
> route-map set-weight-lpref permit 10
>     set local-preference 100
>     set metric 100
> !
> 
> Then apply this as an inbound route-map to your bgp neighbors... Note that
> localpref is an all or nothing proposition (if you localpref one providers
> routes higher than anothers, their routes will be preferred).  In your case,
> you might just need to change the metric to accomplish what you're trying to
> do (without messing with localpref)
> 
> --
> William R. Charnock
> Director of Core Technology
> Allegiance Telecom, Inc
> 
> 
> ----- Original Message -----
> From: "Peter David" <PDavid@...>
> To: <isp-bgp@isp-bgp.com>
> Cc: "George Sepeda" <GeorgeS@...>
> Sent: Monday, January 06, 2003 2:40 PM
> Subject: [isp-bgp] BGP metric weight issue
> 
> 
> >
> > I am trying to figure out how to force route ip address 208.57.25.1 to our
> > LA EBGP gateway which is Genuity.
> > But Genuity is forcing us to route it IGRP wise to our NY EBGP gateway.
> >
> > Problem:
> > When trace the address 208.57.25.1 which is located in Aneheim CA, the
> > average is 85ms since it is being force to NY and Back to CA.
> > The Metric weight in LA for Genuity is 1390
> > The metric weight in NY for Genuity is 0
> >
> > Hence,  It is being forced to go to NY and insteadt of LA
> >
> >
> > Question:
> > Is there a way to force 208.57.25.1 to go thru LA's EBGP Peer instead of
> > NY's EBGP peer without having to change the reomte peer config?
> >
> > Any help is much Appreciated.....
> > Thank you.
> >
> >




> >
> 
> 
> 





-- 
Tycho Eggen             (Unix|Network|Social) Engineer
BOFH/NCIA/NOFH          tycho@...
"Don't worry over what other people are thinking about you."
"They're too busy worrying over what you are thinking about them."