A media company deploys PIM-SM across campus and WAN with many new groups weekly and no tolerance for single Auto-RP mapping failure. Engineers need dynamic discovery with resilience. Which rendezvous point discovery method using static, Auto-RP, or BSR best provides scalable redundancy?

The resilient design works because Bootstrap Router protocol provides standards-based dynamic RP election and distribution, where candidate RPs unicast their group ranges to the elected bootstrap router which then floods bootstrap messages domain-wide, and with multiple candidate BSRs (BSR has no mapping agents; that term belongs to Auto-RP) and multiple candidate RPs there is no single mapping failure while new groups are learned automatically without manual reconfiguration, making it scalable across campus and WAN. The static Anycast approach fails on dynamic discovery because although identical RP addresses with unicast routing give fast failover and are widely deployed, every router still needs manual RP definition so weekly new groups increase operational burden and risk inconsistency. The Auto-RP design with multiple RPs but one mapping agent fails on resilience because candidate RPs still depend on that single agent to arbitrate and advertise the mapping, recreating the prohibited single point of failure and requiring sparse-dense mode for autodiscovery groups. The centralized BSR design with only one candidate RP fails on redundancy because dynamic learning still functions but loss of that RP halts mapping for affected groups despite BSR itself being resilient. As an exam caveat, distinguish transport and scope: Auto-RP is Cisco-proprietary and uses 224.0.1.39 and 224.0.1.40 with dense flooding, while BSR is open-standard and uses 224.0.0.13 hop-by-hop, a favorite test discriminator. As a practical check, verify with show ip pim bsr-router, show ip pim rp mapping, and testing failover by isolating the primary BSR or RP and confirming receivers rejoin without manual changes.