On Thu, Feb 11, 2016 at 12:58 PM, Russell Bryant <russ...@ovn.org> wrote: > > On 02/05/2016 05:53 PM, Han Zhou wrote: > > Before this patch, inter-chassis communication between VIFs of same > > lswitch will always go through tunnel, which end up of modeling a > > single physical network with many lswitches and pairs of lports, and > > complexity in CMS like OpenStack neutron to manage the lswitches and > > lports. > > > > With this patch, inter-chassis communication can go through physical > > networks via localnet port with a 1:1 mapping between lswitches and > > physical networks. The pipeline becomes: > > > > Ingress -> Egress (local) -> Ingress (remote) -> Egress > > > > The original tunneling mechanism will still be used if there is no > > localnet port configured on the lswitch. > > > > Signed-off-by: Han Zhou <zhou...@gmail.com> > > --- > > > > Notes: > > v1->v2: rebase on master, and more updates on documents > > v2->v3: updated based on Russell's comments > > I'm sorry for being slow to review this. I've been missing some emails > lately and didn't receive this one. > > The patch seems to fail to apply: > > Applying: ovn: Connect to remote lports through localnet port. > error: patch failed: ovn/controller/binding.c:122 > > can you rebase it?
Sure, will rebase with v4 by today. > > Otherwise, this is looking good to me, so I'll ack the rebase once I can > apply and run tests one last time. > > > @@ -393,6 +392,14 @@ > > Note that you can not create an ACL matching on a port with > > type=router. > > </p> > > + > > + <p> > > + Note that when <code>localnet</code> port exists in a lswitch, for > > + <code>to-lport</code> direction, the <code>inport</code> works only if > > + the <code>to-lport</code> is located on the same chassis as the > > + <code>inport</code>. However, specifying a <code>localnet</code> port > > + as <code>inport</code> would not have this restriction. > > Specifying the localport in an inport match may have unexpected > behavior, though. It's going to match on the remote hypervisor, even > when the logical source was another lport on the same network. I'm not > sure if it's worth trying to explain that, or just not mentioning > localnet ports at all here. I think I will just remove this sentence in v4, to avoid more confusion. -- Best regards, Han _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
