258
•
The traffic rate of the VPN multicast data has fallen under the switchover threshold and stayed
lower than the threshold for a certain length of time (namely, the switch-holddown period).
•
The associated switch-group-pool is changed and the switch-group address for encapsulating
the VPN multicast data is out of the new address pool.
•
The ACL rule for controlling the switching of VPN multicast traffic from the share-MDT to the
switch-MDT is changed and the VPN multicast data fails to pass the new ACL rule.
Multi-AS MD VPN
If nodes of a VPN network are allocated in multiple autonomous systems (ASs), these VPN nodes
must be interconnected. To implement multi-AS VPN, VRF-to-VRF PE interconnectivity and
multi-hop EBGP interconnectivity are available.
VRF-to-VRF PE interconnectivity
As shown in
, a VPN involves AS 1 and AS 2. PE 3 and PE 4 are the autonomous system
boundary router (ASBR) for AS 1 and AS 2 respectively. PE 3 and PE 4 are interconnected through
their respective VPN instance and treat each other as a CE device.
Figure 81 VPN instance-VPN instance interconnectivity
In the VPN instance-to-VPN instance interconnectivity approach, a separate MD must be
established within each AS, and VPN multicast traffic between different ASs is transmitted between
these MDs.
NOTE:
Because only VPN multicast traffic is forwarded between ASBRs, different PIM modes (PIM-DM,
PIM-SM, BIDIR-PIM, or PIM-SSM) can run within different ASs. However, the same PIM mode must
run on all interfaces that belong to the same VPN (including interfaces with VPN bindings on
ASBRs).
Multi-hop EBGP interconnectivity
As shown in
, a VPN network involves AS 1 and AS 2. PE 3 and PE 4 are the autonomous
system boundary router (ASBR) for AS 1 and AS 2 respectively. PE 3 and PE 4 are interconnected
through their respective public network instance and treat each other as a P device.
