8 May
2011
8 May
'11
8:57 p.m.
On gio, mag 05, 2011 at 03:46:49 +0200, Andrew Lunn wrote:
Thanks!
--
Antonio Quartulli
..each of us alone is worth nothing..
Ernesto "Che" Guevara
On Thu, May 05, 2011 at 09:13:09AM +0200, Antonio
Quartulli wrote:
You are right (dhcpdecline is unicast too).
I have to inspect the dhcp header and look at the message type option.
+ /* If old_gw != NULL then this packet is
unicast.
+ * So, at this point it can only be a renewal packet (because it is the
+ * only dhcp client message sent as unicast) and we have to decide
+ * whether to drop it or not */
What about the release message? Is that not also unicast?
+ if (old_gw
&& curr_gw->orig_node != old_gw) {
+ /* If the dhcp packet has been sent to a different gw, we have
+ * to evaluate whether the old gw is still enough reliable */
still reliable enough.