Hi quicwg, I just published draft-iet-quic-load-balancers-20 with all the changes described in the last email.
Since then, there has been a new Issue <https://github.com/quicwg/load-balancers/issues/235>and PR <https://github.com/quicwg/load-balancers/pull/236> reflecting feedback that the current rules about what to do with unroutable CIDs, and servers that don't have any config loaded, are not exactly airtight for certain edge cases, resulting in many connections being rerouted. The PR, in my opinion, solves this issue, though it dramatically revises the concepts relevant to those edge cases. My experience is that taking agenda time is a waste because very few people care about these details, but I invite you to have a look at the PR, and I'll merge it shortly after IETF 122 modulo any feedback that comes in. Martin
