7.15.2
simplest way i could route leak (to effectively append BGP AS) was with inter-VRF BGP peering. or at least for ipv4.
ip ad add address=172.16.0.0 interface=lo
ip vrf add name=AS65001 interface=bridge-AS65001
ip ad add address=172.16.0.1 interface=AS65001
ip route add dst=172.16.0.1 gateway=AS65001@AS65001
ip route add dst=172.16.0.0 gateway=lo routing-mark=AS65001
both 172.16.0.0 and .1 can ping each other and BGP peering is successful.
however, exact same setup doesn't work for ipv6, ping returns "hop limit exceeded" but only when trying to ping from VRF AS65001 to main VRF, the other way around does ping.
is this a known bug or do i need to do something extra?
i'm assuming it's a bug since it only pings one-way if i don't specify the source. if i do it doesn't ping.
ipv6 ad add address=fd00:: interface=lo
ip vrf add name=AS65001 interface=bridge-AS65001
ipv6 ad add address=fd00::1 interface=AS65001
ipv6 route add dst=fd00::1 gateway=AS65001@AS65001
ipv6 route add dst=fd00:: gateway=lo routing-mark=AS65001
> ping 172.16.0.1 src-address=172.16.0.0
SEQ HOST SIZE TTL TIME STATUS
0 172.16.0.1 56 64 38us
> ping 172.16.0.0 src-address=172.16.0.1 vrf=AS65001
SEQ HOST SIZE TTL TIME STATUS
0 172.16.0.0 56 64 45us
> ping fd00::1 src-address=fd00::
SEQ HOST SIZE TTL TIME STATUS
0 fd00:: 104 64 124us hop limit e...
> ping fd00:: src-address=fd00::1 vrf=AS65001
SEQ HOST SIZE TTL TIME STATUS
0 fd00:: 104 64 52us hop limit e...
the weird ones indicating a bug, since ipv4 works the same just fine even without specifying src address.
> ping fd00::1
SEQ HOST SIZE TTL TIME STATUS
0 fd00::1 56 64 33us echo reply
> ping fd00:: vrf=AS65001
SEQ HOST SIZE TTL TIME STATUS
0 fd00:: timeout