)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":5890,"name":"Doug Goldstein","email":"cardoe@cardoe.com","username":"cardoe"},"change_message_id":"7d8e8a8faba303cf8e125ca076769cb65b851784","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":4,"id":"1de652ed_15553dc3","updated":"2026-05-07 22:59:24.000000000","message":"I’m against this VNI allocation implementation. It’s not expandable at all. This is what I’ve been saying on the spec and others where you worked on the BGP. I’ve been saying over and over that there are multiple pools of VNIs that’s why I created the VXLAN-EVPN spec at first. But really would like to see the VXLAN type support multiple pools. I discussed multiple pools being needed at the PTG. During the spec for this feature it was mentioned that multiple pools will be needed. This patch even says multiple pools will be needed in the future. This week on IRC and the mailing list another user showed up saying they needed multiple pools.\n\nThe way this is done is hardcoded to one pool via a config file entry and not via the API. This will not be an easy upgrade to support that.\n\nThe thing that makes neutron so great are generically designed interfaces with extensibility possible. So let’s do that for this.","commit_id":"1d98d3aabc5df31751a77b157d68349d40010501"}]}