Hi, On Aug 23, 2023, at 20:42, Rishabh Parekh <risha...@gmail.com> wrote: > My point is this draft specifies behavior of a single replication segment > which can be deployed for ingress replication like service (similar to > End.DT2M for Layer 2 BUM) without needing a control plane - either by local > provisioning on root and leaf nodes or by using overlay automatic discovery > mechanisms like BGP MVPN procedures. Loops don't arise in this case unless > underlying unicast point-to-point paths between a Root and Leaf nodes have > loops, but this is out of scope of this document. > > Loops can only arise when replication segments are stitched together to form > a P2MP tree. Typically, this requires a control plane; > draft-ietf-pim-sr-policy is specification of one such control plane (using > PCE as centralized P2MP compute). I think it is appropriate to document loop > prevention or mitigation there. But it is not necessary to use a control > plane to deploy a single replication segment. Hence I don't think this > document needs to have a normative reference to the PIM WG draft.
thanks for the background! I did not fully understand that distinction when reading the document - would you be OK with adding something like the above to the document? Thanks, Lars
signature.asc
Description: Message signed with OpenPGP
_______________________________________________ spring mailing list spring@ietf.org https://www.ietf.org/mailman/listinfo/spring
