So asr2 is pinging from it's global table to a loopback inside a vrf on asr1 (or something like that)?
Sounds like you may need some static routes in there to tell the vrf table how to find the global address, and vice-verse. -- Sent from my mobile device. On 2011-02-13, at 7:26 AM, marc williams <[email protected]> wrote: > setup: > > asr1-----asr2 > > both act as PE routers doing back to back BGP/VRF to customers. they also act > a P routers facing each other. > Currently have an ibgp session per vrf between asr1-asr2 to learn external > routes between asr's. > > Moving to MP-BGP between asr1/2. mpls and ldp session active between asr1/2 > via a vlan/connected /30. > BGP VPNv4 setup and active with next hop self and update source looback0. > show ip bgp vpnv4 vrfname shows the external routes from asr1 are being > learnt on asr2. > > problem i have is asr2 cannot ping asr1 loopback within a vrf. so packets to > external asr1 routes get blackholed. > > on asr2: > show ip bgp vpnv4 vrf name shows the route from asr1 is learnt and is > preferred. > show ip route vrf name shows the route correctly with next hop of loopback of > asr1 > show ip cef vrf name shows the next of of the route is the connected /30 > that connected /30 isnt in the cef vrf table. > > My understanding is that shouldn't be an issue as we should be label > switching. > Any ideas? > > -- > Marc > > > > > > > > > > _______________________________________________ > cisco-nsp mailing list [email protected] > https://puck.nether.net/mailman/listinfo/cisco-nsp > archive at http://puck.nether.net/pipermail/cisco-nsp/ _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
