[c-nsp] IPv6 connectivity on global unicast addresses between two directly connected interfaces if VRRPv3 is enabled

Martin T m4rtntns at gmail.com
Wed May 21 14:30:23 EDT 2014


Hi,

I have two routers r1 and r2 connected to a switch with GigE ports:

r1[Gi] <-> switch <-> [Gi]r2

r1 and r2 are in the same broadcast domain. According to "sh ipv6
interface" command, first router has IPv6 address 2001:10:7::4 and
second one has IPv6 address 2001:10:7::3:

  Global unicast address(es):
    2001:10:7::2, subnet is 2001:10:7::/64 [OOD/VIRT]
    2001:10:7::4, subnet is 2001:10:7::/64

  Global unicast address(es):
    2001:10:7::2, subnet is 2001:10:7::/64 [UNA/OOD/VIRT]
    2001:10:7::3, subnet is 2001:10:7::/64


In addition, as you can see above, GigE interfaces in both routers
have VRRPv3 configured:

 vrrp 1 address-family ipv6
  address FE80::250:56FF:FEA8:5E6C primary
  address 2001:10:7::2/64
  exit-vrrp

While overall IPv6 connectivity, VRRP and IPv4 connectivity between
the routers works fine, the IPv6 one does not:

r1#ping ipv6 2001:10:7::3 source 2001:10:7::4
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 2001:10:7::3, timeout is 2 seconds:
Packet sent with a source address of 2001:10:7::4
.H..H
Success rate is 0 percent (0/5)
r1#

Is it possible that VRRP restricts hosts to reach each other on IPv6?
Unfortunately, I do not have another pair of routers besides those
live ones which support VRRPv3..



regards,
Martin


More information about the cisco-nsp mailing list