[c-nsp] mpls ipv6-tunneling for Cisco

[Daniel Stamatov]

Hi,

Is there Cisco alternative for the Juniper command "mpls ipv6-tunneling"?
Basically I have a network migration scenario from native IPv6 to 6PE and
in interim situation I will need the "native" IPv6 PEs (Cisco) to be able
to resolve the protocol next-hop of ::ffff:<IPv4 address>  for prefixes
originated by PE migrated to 6PE and learned from IPv6 unicast session with
RR (not labeled). I know that this works on Juniper as with the named
command the IPv4 prefixes from inet.3 will be converted and installed in
inet6.3 so a PE which is not migrated to 6PE session will
accept ::ffff:<IPv4 address> as valid next-hop. But from my tests it seems
that Cisco would do this only if the prefix with next hop ::ffff:<IPv4
address> comes from labeled IPv6 unicast session, not from IPv6 unicast
one.

Here is an example:

R4#sh bgp ipv6 unicast 2001:4323:5621::/64
BGP routing table entry for 2001:4323:5621::/64, version 19
Paths: (2 available, best #1, table Global-IPv6-Table)
  Not advertised to any peer
  Local
    ::FFFF:10.0.9.7 (metric 12) from 10.0.3.3 (10.0.3.3)
      Origin IGP, localpref 100, valid, internal, best
      Originator: 10.0.9.7, Cluster list: 10.0.3.3
      mpls labels in/out nolabel/2
  Local
    ::FFFF:10.0.9.7 (inaccessible) from 2001:3333::3 (10.0.3.3)
      Origin IGP, localpref 100, valid, internal
      Originator: 10.0.9.7, Cluster list: 10.0.3.3

Path #1 comes from ipv6 labeled-unicast sesson and ::FFFF:10.0.9.7 is valid
next-hop. Path #2 is comes from native IPv6 session and the next-hop is
marked as "inaccessible". I need this to work when the ipv6 labeled-unicast
sesson is not configured.

Regards,
Daniel