Hi All, Some of you may not be aware yet that a new concern was raised regarding the extraroute improvement plans just after the last neutron session was closed on the PTG. It seems we have a tradeoff between the support for the use case of tracking multiple needs for the same extra route or keeping the virtual router abstraction as simple as it was in the past. I'm raising the question of this tradeoff here in the mailing list because this (I hope) seems to be the last cross-project question of this topic. If we could find a cross-project consensus on this I could continue making progress inside each project without need for further cross-project coordination. Please help me find this consensus. I don't want to unnecessarily repeat arguments already made. I think the question is clearly formulated in the comments of patch sets 5, 6 and 8 of the below neutron-spec: https://review.opendev.org/655680 Improve Extraroute API All opinions, comments, questions are welcome there. Thanks in advance, Bence (rubasov)