UMN:CLI
User Manual
V8102
528
[
A.B.C.D
]
A.B.C.D: multicast group or source address
A.B.C.D/M: range of multicast group addresses
static: static multicast route entry
summary: summary of multicast route entry
show ip pim mroute
[
A.B.C.D/M
|
static
|
summary
]
show ip pim nexthop
Shows the next hop information.
A.B.C.D | *: specific or any source address
A.B.C.D: multicast group address
vrf NAME: the name of the VPN routing/forwarding
instance
show ip
[
vrf NAME
]
pim nexthop
{
A.B.C.D
|
*
} [
A.B.C.D
]
10.3.3
PIM-SM
Rendezvous Point Tree (RPT)
PIM-SM mainly uses a shared tree to deliver multicast traffic, called the RP tree (RPT). As
its name implies, it relies on a core router called the Rendezvous Point (RP) that receives
all multicast traffic from the sources and forwards that traffic to the receivers. Other rout-
ers do not need to know the information of the sources. All they need to know is the ad-
dress of the RP, because the RP surely knows the information of the sources for all mul-
ticast groups. Thus, receivers who are interested in a certain multicast group only send
PIM join messages with (*, G) state toward the RP. That is, the RPT prevent each router
from maintaining source and group (S, G) states for every multicast source. This mecha-
nism shifts the burden of finding the multicast sources from each router to the network it-
self.
The shared tree is unidirectional, which means all multicast traffic flows only from the RP
to the receivers. Thus, there is no guarantee that the shared tree (RPT) is the shortest
path tree to the source, and most likely it is not, resulting in longer delays, but less for-
warding states to maintain. Each multicast group has only one RP that may be different;
each multicast group may have the different distribution tree.
Fig. 10.6 shows an example of the RPT network. The multicast traffic from the source A
flows through the router B to the router D which is the RP. Note that, even in the RPT,
RPs must receive multicast traffic from the sources via the shortest path. The RP then
distributes the traffic to the receiver E and F that indicate the interest in the multicast
group. Consequently, the distribution tree for the receiver E is
A
→
B
→
D
→
E
, and the one
for the receiver F is
A
→
B
→
D
→
C
→
F
.