Difference between revisions of "Network/Meetings"
Markmcclain (talk | contribs) (→Other team reports (e.g. Open Source Plugins)) |
Markmcclain (talk | contribs) (→ML2) |
||
Line 56: | Line 56: | ||
=== ML2 === | === ML2 === | ||
+ | ** https://blueprints.launchpad.net/quantum/+spec/modular-l2 - WIP patch https://review.openstack.org/#/c/20105/ in review. Need to add linuxbridge support and tunnel support for openvswitch, add unit tests, get tested with hyperv. Initial merge on track for H-1. | ||
+ | ** https://blueprints.launchpad.net/quantum/+spec/portbinding-ex-db - Quantum patch https://review.openstack.org/#/c/24969/ merged. Still need Satish's nova patch https://review.openstack.org/#/c/21141/ to be merged. Quantum side ready for H-1, but not getting exercised until nova side merged. | ||
+ | ** https://blueprints.launchpad.net/quantum/+spec/ovsplugin-hardware-devices - Review https://review.openstack.org/#/c/20754/ raises issues regarding relationship to ml2 plugin effort. Assuming ml2 does replace openvswitch and linuxbridge plugins, is ml2's (to-be-defined) MechanismDriver API sufficient, or is a separate HardwareDriver API still valuable? | ||
+ | ** https://blueprints.launchpad.net/quantum/+spec/vxlan-linuxbridge - Review https://review.openstack.org/#/c/26516/ making good progress. Changes to linuxbridge plugin/agent RPC also needed for ml2 plugin, but RPC version needs to match openvswitch and hyperv. Assuming ml2 is replacing linuxbridge and a VxlanTypeDriver for ml2 can be developed for H-1, its not clear if server-side plugin changes for VXLAN should be merged yet. Should linuxbridge RPC changes be merged separately first before VXLAN added? | ||
=== Python Client (gongysh) === | === Python Client (gongysh) === |
Revision as of 20:47, 20 May 2013