[openstack-dev] [Neutron] Different behavior of admin_state_up attribute between trunk and port

Le, Huifeng huifeng.le at intel.com
Wed Sep 26 02:04:15 UTC 2018


When using neutron, it is found that the behavior of admin_state_up attribute is different between port and trunks which may impact user experience. e.g. for port, Setting admin_state_up = FALSE will set port status to "Down", then disable the port to block send/receive packages (e.g. set port's tap device state to down or move the port to DEAD_VLAN etc.); and for trunks, Setting the admin_state_up = FALSE locks the trunk in that it prevents operations such as adding/removing subports, but the trunk can still operational (e,g. the trunk sub-port can still send/receive packages), and.

So does it make sense to make a change (e.g. submit a bug or RFE) which aligns the behavior between trunk and port (e.g. setting admin_state_up= FALSE on a trunk will disable that trunk on the vswitch - stop processing VLAN packets arriving from that VM instance) to avoid potential confusing of usage?

Thanks much!

Best Regards,
Le, Huifeng
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20180926/f068a2c5/attachment.html>

More information about the OpenStack-dev mailing list