298
The flood-and-prune process takes place periodically. A pruned state timeout mechanism is provided. A
pruned branch restarts multicast forwarding when the pruned state times out and then is pruned again
when it no longer has any multicast receiver.
NOTE:
Pruning has a similar implementation in IPv6 PIM-SM.
Graft
When a host attached to a pruned node joins an IPv6 multicast group, to reduce the join latency, IPv6
PIM-DM uses the graft mechanism to resume IPv6 multicast data forwarding to that branch. The process
is as follows:
1.
The node that needs to receive IPv6 multicast data sends a graft message toward its upstream node,
as a request to join the SPT again.
2.
After receiving this graft message, the upstream node puts the interface on which the graft was
received into the forwarding state and responds with a graft-ack message to the graft sender.
3.
If the node that sent a graft message does not receive a graft-ack message from its upstream node,
it keeps sending graft messages at a configurable interval until it receives an acknowledgment
from its upstream node.
Assert
On a shared-media network with more than one multicast router, the assert mechanism shuts off duplicate
IPv6 multicast flows to the network. It does this by electing a unique IPv6 multicast forwarder on the
shared-media network.
Figure 83
Assert mechanism
As shown in
, the assert mechanism is as follows:
1.
After Router A and Router B receive an (S, G) IPv6 multicast packet from the upstream node, both
of them forward the packet to the local subnet.
As a result, the downstream node Router C receives two identical multicast packets, and both
Router A and Router B, on their own downstream interfaces, receive a duplicate IPv6 multicast
packet that the other has forwarded.
2.
After detecting this condition, both routers send an assert message to all IPv6 PIM routers on the
local subnet through the interface that received the packet.
