On Wed, Oct 23, 2013 at 11:04:07AM -0700, Jarno Rajahalme wrote: > > On Oct 23, 2013, at 10:29 AM, Ben Pfaff <[email protected]> wrote: > > > On Tue, Oct 22, 2013 at 05:20:44PM -0700, Jarno Rajahalme wrote: > >> Signed-off-by: Jarno Rajahalme <[email protected]> > > > > Thanks, I applied this to master. > > > > I didn't know what you meant by this, perhaps we should clarify it? > >> [...] Switch behavior when setting > >> +\fIttl\fR to zero is not well specified, though. > > Currently, zero is a valid value for a set action, but it is up to > interpretation weather the switch should send the packet to the > controller when the ttl is set to zero (as is the case for dec ttl > action. It might make sense to change the behavior to return an > invalid value error for the set TTL actions, as there should be no > reason for a controller to set a TTL to zero?
Ah. I think I'd rather maintain the status quo, by letting the controller set TTL to zero. Plausibly, one could use that to test an IP stack's behavior. _______________________________________________ dev mailing list [email protected] http://openvswitch.org/mailman/listinfo/dev
