[j-nsp] vpn-apply-export

harry harry at juniper.net
Mon Nov 10 12:43:23 EST 2003


Odd. Can you confirm that the static route is active in the VRF?

It seems that:

 policy-statement vpn-telemach-export {
        term a {
            then {
                community add vpn-telemach;
                accept;
            }
        }
        term b {
            then reject;

Should match on all active routes in the VRF and export them. Once a route
is accepted it cannot be later rejected (as I understand), so I am not sure
why modifying your PE-PE export policy should have any effect. Your AS path
reg x seems to match on a null AS path, so this explains why adding static
allows the export of the static despite no as path parameters being
configured for that static. You may want to open a case with JTAC as this
does not seem to be the desired behavior.



> -----Original Message-----
> From: Blaz Zupan [mailto:blaz at inlimbo.org] 
> Sent: Monday, November 10, 2003 9:20 AM
> To: harry
> Cc: juniper-nsp at puck.nether.net
> Subject: RE: [j-nsp] vpn-apply-export
> 
> 
> > I think I need to see the complete config to get a better 
> handle. It 
> > sounds like you are saying:
> >
> >  term limited-routes {
> >         from {
> >             protocol [ bgp aggregate static ];
> >             as-path limited-routes;
> >         }
> >         then accept;
> >     }
> >
> >
> > Is part of your unicast PE-PE BGP routing policy. I assumed 
> this was 
> > part of your VRF related policies.
> 
> Below is the relevant part of the configuration. This is the 
> configuration that worked with 5.5. With 5.7, I need to add 
> "static" to the "limited-routes" term of the 
> "to-ibgp-amis-routes" policy, otherwise the static default 
> route does not propagate to the Cisco PE. I believe this is 
> supposed to be needed only when I configure vpn-apply-export, 
> but I did not configure it.
> 
> protocols {
>     mpls {
>         interface e3-0/0/2.0;
>         interface fe-0/1/0.308;
>     }
>     bgp {
>         group ibgp {
>             type internal;
>             local-address 212.18.32.166;
>             peer-as 8591;
>             neighbor 212.18.32.175 {
>                 family inet {
>                     unicast;
>                 }
>                 family inet-vpn {
>                     unicast;
>                 }
>                 export to-ibgp-amis-routes;
>                 cluster 212.18.32.166;
>             }
>         }
>     }
>     ldp {
>         interface e3-0/0/2.0;
>     }
> }
> policy-options {
>     policy-statement to-ibgp-amis-routes {
>         term as8591 {
>             from community [ from-customer from-lix from-six ];
>             then accept;
>         }
>         term limited-routes {
>             from {
>                 protocol [ bgp aggregate ];
>                 as-path limited-routes;
>             }
>             then accept;
>         }
>         term everything-else {
>             then reject;
>         }
>     }
>     policy-statement vpn-telemach-import {
>         term a {
>             from {
>                 protocol bgp;
>                 community vpn-telemach;
>             }
>             then accept;
>         }
>         term b {
>             then reject;
>         }
>     }
>     policy-statement vpn-telemach-export {
>         term a {
>             then {
>                 community add vpn-telemach;
>                 accept;
>             }
>         }
>         term b {
>             then reject;
>         }
>     }
>     community from-customer members 8591:1000;
>     community from-lix members 8591:1002;
>     community from-six members 8591:1001;
>     community vpn-telemach members target:12644:1;
>     as-path limited-routes "[0-65535]{0,2}";
> }
> 
> routing-instances {
>     telemach {
>         instance-type vrf;
>         interface fe-0/1/0.308;
>         route-distinguisher 12644:1;
> 	vrf-import vpn-telemach-import;
> 	vrf-export vpn-telemach-export;
>         routing-options {
>             static {
>                 route 0.0.0.0/0 next-hop 213.143.77.41;
>             }
>         }
>     }
> }
> 



More information about the juniper-nsp mailing list