+++ This bug was initially created as a clone of Bug #1841526 +++ Description of problem: right now VXLAN is not supported by OVN for regular inter-chassis communication. There is VXLAN support for "vtep" ovsdb schema compliant external switches (aka ramp switches). This RFE is to extend core OVN to support VXLAN for regular in-cluster traffic. Initial upstream discussion: https://www.mail-archive.com/ovs-discuss@openvswitch.org/msg06771.html Demo implementation: https://patchwork.ozlabs.org/project/openvswitch/patch/20200320050711.247351-1-ihrachys@redhat.com/ Related links: https://blog.russellbryant.net/2017/05/30/ovn-geneve-vs-vxlan-does-it-matter/ Core OVN work is tracked in bug 1841526. This bug is to adjust Neutron to support the new network type for OVN driver. There are two elements to it: 1) allow VXLAN type (as simple as https://review.opendev.org/734889); 2) read max_tunid from NB_Global object and use it to adjust vni_ranges used by VXLAN type driver. Since max_tunid is written in the latest OVN only, its presence can be used as a tag that VXLAN support is present.
*** Bug 1916091 has been marked as a duplicate of this bug. ***
*** Bug 1916092 has been marked as a duplicate of this bug. ***
According to our records, this should be resolved by openstack-tripleo-heat-templates-11.5.1-2.20210603174823.el8ost.9. This build is available now.