Skip to main content

Multicast - Rendezvous Point (5)

Rendezvous Point (RP) Deployment Methods

A Rendezvous Point (RP) is a router that serves as the common shared root for a multicast shared tree. Multiple RPs can be configured to cover different multicast group ranges. However, for a network to operate correctly, every router within a PIM domain must map a given multicast group address to the exact same RP.

There are three primary methods used to define and distribute RP information across a PIM domain: Static RP, Bootstrap Router (BSR), and Auto-RP.

image.png

1. Static RP

Static RP requires manual configuration of the RP IP address on every single router within the multicast domain.

image.png

  • Mechanics: As shown in the image, specific routers are manually designated as the RP for a group range. If an RP address changes, every router in the network must be updated manually.

  • Precedence Rule: Static RP can coexist with dynamic mechanisms, but dynamically learned RP information takes precedence over manually configured static RPs by default.

  • Pros: Simple to implement in small networks, requiring only one or two lines of configuration per router.

  • Cons: 

    • Highly laborious and error-prone in large, complex environments.

    • Lacks a native failover mechanism; if a static RP fails, traffic drops until manual intervention occurs (unless combined with an advanced feature like Anycast RP).

    • Does not inherently support load balancing.

2. Bootstrap Router (BSR)

BSR is an industry-standard, automated mechanism (defined in RFC 5059) used to distribute RP information dynamically.

image.png

  • Mechanics: As shown in the image, the architecture relies on an elected BSR and Candidate RPs (C-RPs):

    1. Candidate RPs send unicast "Candidate-RP Advertisements" to the elected BSR.

    2. The BSR collects these advertisements into a list called the "RP set."

    3. The BSR floods this complete RP set to all routers in the domain using hop-by-hop BSR Messages.

    4. Each individual router runs a identical, common algorithm locally to select the exact same RP address for any given multicast group.

  • Pros:

    • Nonproprietary open standard that works across multi-vendor hardware.

    • Reduces administrative overhead since configuration is restricted only to the candidate routers.

    • Highly robust with built-in failover; if a primary RP fails, a backup RP from the set automatically takes over.

  • Cons: Cannot coexist with Auto-RP; only one dynamic protocol must be configured in the network.

3. Auto-RP

Auto-RP is a legacy, Cisco-proprietary automation mechanism designed to distribute RP mapping information without manual per-router configuration.

image.png

  • Mechanics: As shown in the image, Auto-RP relies on two separate functional components and two dedicated multicast groups:

    1. Candidate RPs (C-RPs): Announce their availability by broadcasting periodic "RP-announcement" messages to the well-known multicast group 224.0.1.39 (CISCO-RP-ANNOUNCE).

    2. RP-Mapping Agents: Join group 224.0.1.39 to listen to all candidate offers. The Mapping Agent resolves any conflicts, determines the authoritative group-to-RP mappings, and broadcasts the final decision to group 224.0.1.40 (CISCO-RP-DISCOVERY).

    3. PIM Routers: All routers in the domain join group 224.0.1.40 to receive and cache the authoritative RP mappings.

  • Pros: Fully automated discovery, easy updates, and native support for backup RPs and failover profiles.

  • Cons:

    • Cisco proprietary protocol.

    • Not supported for IPv6 multicast.

    • Cannot coexist with BSR in the same network.

Back to top