On Thu, Nov 8, 2012 at 2:49 PM, Kyle Mestery (kmestery)
<[email protected]>wrote:

> On Nov 6, 2012, at 1:27 PM, Jesse Gross <[email protected]> wrote:
> >
> > Hi Kyle,
> >
> > Things are continuing to move along and I think that most of the major
> infrastructure pieces are either complete or coming along nicely.  One area
> that might make sense for you to work on is the VXLAN dataplane support.  I
> know that you've been maintaining Ben's patch out of tree so hopefully
> there isn't too much that needs to be done.  In the past I was reluctant to
> apply it on it's own but I think now that everything else is coming
> together it makes sense.
> >
> > A couple things that come to mind that you might want to look into:
> >  * Any updates to apply against master (it's looks pretty current but
> I'm guessing that the recent header caching changes will require some
> updates).
> >  * I think we probably want to make the destination port configurable as
> part of the port properties from the database.  Fixing it in the kernel
> seems a little restrictive.
>
> Hi Jesse:
>
> I spoke with our IETF folks, and they said that all shipping VXLAN
> implementations are using the OTV port (8472) as the destination port for
> VXLAN. Given this, do we still want to make this configurable? Please see
> section 3.1 of the L2 LISP draft here:
>
> http://tools.ietf.org/html/draft-smith-lisp-layer2-01


I think we should use the OTV port by default but still make it
configurable (which I know at least some other implementations enable as
well).  The situation seems fairly analogous to other upper layer protocols
like HTTP where there is a well defined port number but people still like
to change it.
_______________________________________________
dev mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/dev

Reply via email to