Ah, thanks. I just added some more aggregates into vrf test within bgp and I
see how they all use same label.
Aaron
RP/0/0/CPU0:R8-xr#sh mpls for | in test
16082 Aggregate test: Per-VRF Aggr[V] \
test 0
RP/0/0/CPU0:R8-xr#sh bgp vpnv4 uni lab | in "16082|Network"
Fri Feb 13 10:07:44.074 UTC
Network Next Hop Rcvd Label Local Label
*> 172.16.0.0/16 0.0.0.0 nolabel 16082
s> 172.16.1.0/24 0.0.0.0 nolabel 16082
s> 172.16.2.0/24 0.0.0.0 nolabel 16082
*> 172.17.0.0/16 0.0.0.0 nolabel 16082
s> 172.17.1.0/24 0.0.0.0 nolabel 16082
s> 172.17.2.0/24 0.0.0.0 nolabel 16082
*> 192.168.0.0/16 0.0.0.0 nolabel 16082
* i 10.0.0.5 23 16082
*** Another PE on other side of network....
R1#sh bgp vpnv4 unicast all labels | in Network|16082
Network Next Hop In label/Out label
172.16.0.0 10.0.0.8 nolabel/16082
172.17.0.0 10.0.0.8 nolabel/16082
192.168.0.0/16 10.0.0.8 nolabel/16082
172.16.0.0 10.0.0.8 nolabel/16082
10.0.0.8 nolabel/16082
172.17.0.0 10.0.0.8 nolabel/16082
10.0.0.8 nolabel/16082
192.168.0.0/16 10.0.0.8 nolabel/16082
10.0.0.8 nolabel/16082
From: Adam Vitkovsky [mailto:[email protected]]
Sent: Friday, February 13, 2015 10:05 AM
To: Lukas Tribus; Aaron; [email protected]
Subject: RE: [c-nsp] aggregate lfib entry - what is the actual prefix ?
Lukas is right,
Think of it this way:
Consider data-plane perspective:
When a packet arrives on the PE destined to a 192.168.1.1 IP address -it's not
going to be switched to where the 192.168.0.0/16 points to as it would get
dropped.
Instead an additional FIB lookup in the VRF is needed to find out which
interface to use in order to forward the packet.
adam
> -----Original Message-----
> From: cisco-nsp [mailto:[email protected]] On Behalf Of
> Lukas Tribus
> Sent: 13 February 2015 15:03
> To: Aaron; [email protected]
> Subject: Re: [c-nsp] aggregate lfib entry - what is the actual prefix ?
>
> > Are you sure it doesn't map to a particular prefix ? I just found this.
> > This seems to be what it maps to...the aggregate route that I config'd
> > within bgp vrf test.
>
> Aggregate routes will point to *the* aggregate vrf label, yes. But you can
> have many aggregates and they all will point to the very same label.
>
> The aggregate label may only have been created because of your aggregate
> route, but that doesn't mean its per aggregate prefix. Its per vrf, really.
>
>
> Lukas
>
>
> _______________________________________________
> cisco-nsp mailing list [email protected]
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
_____
This email has been scanned for email related threats and delivered safely by
Mimecast.
For more information please visit http://www.mimecast.com
_____
_______________________________________________
cisco-nsp mailing list [email protected]
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/
