Everything sounds good to me. A few responses:
On Mon, Aug 08, 2011 at 08:03:53PM -0700, Sanjay Sane wrote:
> > I would consider adding a line to NEWS to mention this new feature.
> >> when LACP is enabled, even if bpdu-passthrough is enabled
> >> {it seems we always consume LACP frames, even if LACP is not enabled..}
> >
> > Is this a bug that we should plan to fix separately? ?I think I know
> > why it happens, if so.
>
> Based on Ethan and your subsequent comments, I'll remove this note
> from my patch, as this will be fixed soon.
It's fine to leave it that comment, if you like, since it is accurate at
the moment.
> > I'm not sure of the naming of "bpdu_passthru". ?I don't think that we
> > use the term "passthru" anywhere else in OVS. ?Usually in OVS we speak
> > of "admitting" or "accepting" or "forwarding" or (the opposite)
> > "dropping" packets. ?Nicer names might be "forward_bpdus" or (with the
> > opposite meaning) "drop_bpdus". ?Also I wonder whether "bpdus" is the
> > correct term. ?I believe that this setting covers all packets to
> > reserved MAC addresses. ?Are all of those properly called BPDUs?
>
> in STP/switching world, there's something called bpdufilter. Even in
> L2TP config, BPDU usually refers to these frames. So, the term "bpdu"
> is well-known for these control class frames
>
> I've changed the keyword to forward_bpdu (instead of "bpdus), does
> that sound ok ?
Thanks, I do like "forward_bpdu" better.
_______________________________________________
dev mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/dev
