Ask Your Question

Do we effectively still have 4K VLAN limit with VXLAN?

asked 2015-03-18 11:57:37 -0500

kxs81 gravatar image

I understand one of the big advantage of VXLAN is that it increases the number of VLAN ids (VNI) to 16 Million. However, if my understanding is correct , internally Neutron — at least with OpenvSwitch -- still uses traditional VLAN IDs (internal VLAN <-> VXLAN id mapping is done by flowtables in br-tun, br-int bridges) to separate tenant networks ( ( ) , don’t you still end up with 4K VLAN restriction even with VXLAN support?

edit retag flag offensive close merge delete

1 answer

Sort by » oldest newest most voted

answered 2015-03-19 01:39:21 -0500

dbaxps gravatar image

updated 2015-03-19 01:46:20 -0500

Actually, links bellow ( regardless they connected GRE tunnels ) answer your question

In case of GRE or VXLAN tenants L2 networks the VLAN tags you see in the output of "ovs-vsctl show" and in output of "ovs-ofctl dump-flows br-tun" (mod_vlan_vid) are only locally significant. This VLAN tags are not really L2 tags added to the frames leaving on the physical interface.They are only used by openvswitch to separate traffic on the br-int, so the different  tap interfaces corresponding to different neutron subnets do not see each other's traffic

View also
Last sample was built by myself :

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2015-03-18 11:57:37 -0500

Seen: 861 times

Last updated: Mar 19 '15