On Wed, Jun 29, 2011 at 03:52:44PM -0700, Ethan Jackson wrote:
> > It sounds right. ?Have you tested it?
>
> I've done no testing on this patch. It seems fairly low risk to me,
> I'd be inclined to let the system testing we already have cover it.
OK, that's fine, I was curious more than anything.
> > The comment just above "if (bond->lacp_negotiated) {" should probably
> > be updated.
>
> Changed to the following:
>
> /* LACP bonds have very loose admissibility restrictions because we can
> * assume the remote switch is aware of the bond and will "do the right
> * thing". However, as a precaution we drop packets on disabled slaves
> * because no correctly implemented partner switch should be sending
> * packets to them. */
Great.
_______________________________________________
dev mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/dev
