On Thu, Oct 16, 2014 at 02:14:11PM -0700, Alex Wang wrote:
> On Thu, Oct 16, 2014 at 1:47 PM, Ben Pfaff <b...@nicira.com> wrote:
>
> > On Thu, Oct 16, 2014 at 12:53:50PM -0700, Alex Wang wrote:
> > > >
> > > > >
> > > > > Signed-off-by: Niels van Adrichem <n.l.m.vanadric...@tudelft.nl>
> > > >
> > > > Thanks. I applied this. It is definitely an improvement.
> > > >
> > > > There are two ways that we should continue to improve it, though. One
> > > > is incremental: CFM, as well as BFD, should control forwarding.
> > >
> > > Hey Ben, could you explain more on this 'incremental'?
> > >
> > > I think with this commit, the bfd does control the forwarding of group
> > > action
> > > in the following way:
> > > 1. when bfd forwarding flag flaps, it changes the global sequence number.
> > > 2. as reaction to global sequence number change, port_run() is executed.
> > > 3. if the port_run() execution finds that the 'ofport->may_enable' flaps
> > and
> > > sets the 'backer->need_revalidate'.
> > > 4. the set of 'backer->need_revalidate' causes the dp flow revalidation
> > and
> > > changes the forwarding of group action.
> > >
> > > Does this make sense?
> >
> > I think that that is correct. I mean that we should also support CFM,
> > not just BFD, here.
> >
>
> Sounds like we should just use the ofport->may_enable,?
I did not think of that. It is better, I think.
I posted a patch:
http://openvswitch.org/pipermail/dev/2014-October/047455.html
It passes the unit tests. Niels, will you check that it works for your
use case?
Thanks,
Ben.
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
