18-3
Catalyst 6500 Series Switch Cisco IOS Software Configuration Guide—Release 12.1 E
78-14099-04
Chapter 18 Configuring IP Multicast Layer 3 Switching
Understanding How IP Multicast Layer 3 Switching Works
The Layer 3 switching cache contains flow information for all active Layer 3-switched flows. After the
switching cache is populated, multicast packets identified as belonging to an existing flow can be
Layer 3 switched based on the cache entry for that flow. For each cache entry, the PFC maintains a list
of outgoing interfaces for the IP multicast group. From this list, the PFC determines onto which VLANs
traffic from a given multicast flow should be replicated.
These commands affect the Layer 3 switching cache entries:
• Clearing the multicast routing table (using the clear ip mroute command) clears all multicast
Layer 3 switching cache entries.
• Disabling IP multicast routing on the MSFC (using the no ip multicast-routing command) purges
all multicast Layer 3 switching cache entries on the PFC.
• Disabling multicast Layer 3 switching on an individual interface basis (using the no mls ip
multicast command) causes flows that use this interface as the RPF interface to be routed only by
the MSFC in software.
IP Multicast Layer 3 Switching Flow Mask
IP multicast Layer 3 switching with PFC1 supports only the multicast source-destination-VLAN flow
mask. PFC1 maintains one multicast Layer 3 switching cache entry for each {source IP, destination
group IP, source VLAN}. The multicast source-destination-VLAN flow mask differs from the IP unicast
MLS source-destination-ip flow mask in that, for IP multicast Layer 3 switching, the source VLAN is
included as part of the entry. The source VLAN is the multicast RPF interface for the multicast flow.
Flows are based on the IP address of the source device, the destination IP multicast group address, and
the source VLAN. The MSFC uses the RPF interface to send a unicast packet back to the source.
Layer 3-Switched Multicast Packet Rewrite
Note Only ARPA rewrites are supported for IP multicast packets. Subnetwork Address Protocol (SNAP)
rewrites are not supported.
When a multicast packet is Layer 3 switched from a multicast source to a destination multicast group,
PFC1 performs a packet rewrite based on information learned from the MSFC and stored in the Layer 3
switching cache. In the case of PFC2 and the DFCs, the packet rewrite is based on information learned
from the MSFC2 and is stored in the adjacency table. The format of the packet rewrite is the same for
PFC1, PFC2, and DFCs.
For example, Server A sends a multicast packet addressed to IP multicast group G1. If there are members
of group G1 on VLANs other than the source VLAN, the PFC must perform a packet rewrite when it
replicates the traffic to the other VLANs (the switch also bridges the packet in the source VLAN).
When the PFC receives the multicast packet, it is formatted (conceptually) as follows:
Layer 2 Frame Header Layer 3 IP Header Data FCS
Destination Source Destination Source TTL Checksum
Group G1 MAC1
1. In this example, Destination B is a member of Group G1.
Source A MAC Group G1 IP Source A IP n calculation1