**Session Date/Time:** 09 Nov 2022 09:30

# pim

## Summary

This PIM working group meeting covered several topics, including working group status, PIM light, a multicast address allocation proposal, an EVPN multicast YANG model presentation, and a discussion about documenting the history of multicast.  Key discussions revolved around TCP for PIM light, coordinating with other WGs for the multicast address allocation proposal, and the scope and location (PIM vs. MBONED) for the historical multicast document.

## Key Discussion Points

*   **Working Group Status:**  Progress on RFCs, need for document shepherds, review of several documents.
*   **PIM Light:** Clarification on message types, PMS parts mode, join attributes, DR selection, and failure detection. Discussion on using TCP (RFC 6559) for PIM light for reliability, particularly in beer deployments. Debate on whether PIM light implementations must support a default to Port 6559.
*   **Multicast Address Allocation:** Presentation on a proposal for zero-configuration multicast address allocation in boat networks. Concerns about mdns dependency and scope (Layer 2 only initially).  Discussion about coordinating with DNS and IPv6 experts and Brian Haberman.
*   **EVPN Multicast YANG Model:** Presentation of a new YANG model for EVPN multicast.  Feedback suggested that this work should be presented and potentially owned by the BEST working group due to its reliance on EVPN/BGP.
*   **Tree DN: A Tree-Based CDN Model:** Discussion on using SSM and AMT, deployment strategies. Discussion on the applicability of the same document in Mboned. Source Discovery requirements.
*   **History of Multicast and Lessons Learned:** Discussion on the need for a historical document to understand previous development work. Debate on its scope (protocol vs. deployment lessons) and its potential location (PIM vs. MBONED).

## Decisions and Action Items

*   **PIM Light:** Homan to review RFC 6559 and open a conversation. Cisco will verify if it is agnostic to whether TCP or UDP is used. Sandy will be reflected in the data tracker lister as the shepherd and made a delegate in the working group.
*   **Multicast Address Allocation:**  Stig to drive coordination with the DNS working group and other experts, including Brian Haberman.  Nate to update RFC 3307.
*   **EVPN Multicast YANG Model:** Stig and Mike will reach out to BEST chairs to determine the appropriate venue for this work.
*   **History of Multicast and Lessons Learned:** Discuss adoption on the PIM mailing list and discuss with the MBONED chairs to check what the best home is for the document.

## Next Steps

*   **PIM Light:** Homan will review RFC6559, other authors to work on draft based on the comments.
*   **Multicast Address Allocation:** Stig will connect with the DNS WG and other experts. Authors to address open issues and document the same.
*   **History of Multicast and Lessons Learned:** Begin Adoption call, discuss with MBONED chairs for the best venue.