Protocol Independent Multicast (PIM) sparse mode is the most scalable protocol for IP multicasting. This relies on an explicit join model to forward multicast traffic. The multicast sources and receivers register with a device functioning as a Rendezvous Point (RP) to form the multicast distribution tree. Every router in the sparse mode network has to know the address of the device functioning as RP.
To configure an RP, choose one of these options:
To configure a static RP address, issue the ip pim rp-address command. This method is not scalable and does not provide redundancy for RP. It is recommended only in very small, simple networks. All the routers are configured to operate in sparse mode by issuing the ip pim sparse-mode command under the interfaces.
To distribute the RP information to all the routers, use the Cisco proprietary Auto-RP mechanism. This also provides redundancy. If an RP is known, Auto-RP uses sparse-dense mode, which allows a group to be treated in sparse mode. If an RP is not known, the group defaults to dense mode. RP information is conveyed to the routers by the Auto-RP mechanism. This mechanism uses two well-known groups operating in dense mode: 22.214.171.124 and 126.96.36.199.
Devices functioning as mapping agents listen to the 188.8.131.52 group address to announce candidate RPs. This is done by issuing the ip pim send-rp-announce command. The mapping agents issued with the ip pim send-rp-discovery command listen to the 184.108.40.206 address, choose the RP for the group based on the one with the highest RP address, and advertises it to the 220.127.116.11 group address to which all other routers listen. The advantage of this method is that any change to the RP designation needs to be configured only on the RP routers, and not on the leaf routers, as in the static configuration. All the routers need to be configured to operate in both sparse and dense mode. Issue the ip pim sparse-dense-mode command under the interfaces.
Use the standards based Cisco Bootstrap Router (BSR) mechanism available in PIM version 2. This method also provides the same benefit of Auto-RP. Devices configured as candidate BSRs by issuing the ip pim bsr-candidate command are equivalents of mapping agents in Auto-RP. These devices announce themselves to other routers using the 18.104.22.168 group address. Since a message sent to this address is forwarded hop-by-hop throughout the network, all the routers learn about the candidate BSRs and choose one of them as the BSR based on the highest priority or IP address.
Similar to Auto-RP, candidate RPs issued with the ip pim rp-candidate command advertise themselves to devices functioning as a BSR through the unicast mechanism. The RP set information is then advertised by the BSR to the 22.214.171.124 group address for other routers to learn about the RPs. The difference from Auto-RP is that the interfaces can be configured to operate only in sparse mode by issuing the ip pim sparse-mode command, since BSR messages are flooded hop-by-hop. Another difference is that each router receiving the RP set information from the BSR uses a hash algorithm to select the RP for a group. In Auto-RP mechanism, the mapping agent selects and advertises the group to RP mapping to leaf routers.
For more information on configuring an RP for PIM sparse mode, refer to these documents:
We'd like to learn a little about your network, your pain points with monitoring an enterprise network, and your preferred solution and workflow to solve issues.
We ask that you complete our brief survey: https://ciscoux.az1.qualtrics.com/jfe/form/SV...
Good day !I'm in the process of building a common configuration that I'm using to bootstrap multiple xrv 9k instances. The configuration comprises groups that are then applied in the relevant section. While I have verified that the groups are defined on s...
The first error we get is an error message that states: "SYS-2-MALLOCFAIL - Memory allocation of 962820 bytes failed from 0x8000561D10652750, alignment 8 Pool: Processor Free: 71887584 Cause: Memory fragmentation Alternate Pool: None Free: 0 Cause: N...
How to check Nexus 9K switches can be managed with Dnac current version 126.96.36.199 or not. Is there any possibility of checking compatibility with our existing version ? We want to deploy Nexus switches in DNAC tool with version 188.8.131.52, So please sugge...
Have you ever wondered why pluggable optics exist? Have you ever wondered what acronyms like QSFP, LR4, FEC, and PAM4 actually mean? In this episode I continue my conversation with my colleague Ray Nering. He explains to me how a lot of these came to be, ...