On Thu, Mar 05, 2020 at 05:16:04PM -0500, Ihar Hrachyshka wrote: > On Thu, Mar 5, 2020, 15:15 Ben Pfaff <[email protected]> wrote: > > > On Thu, Mar 05, 2020 at 02:56:17PM -0500, Ihar Hrachyshka wrote: > > > As per RFC2131, section 4.1: > > > A server or relay agent sending or relaying a DHCP message directly > > > to a DHCP client (i.e., not to a relay agent specified in the > > > 'giaddr' field) SHOULD examine the BROADCAST bit in the 'flags' > > > field. If this bit is set to 1, the DHCP message SHOULD be sent as > > > an IP broadcast using an IP broadcast address (preferably 0xffffffff) > > > as the IP destination address and the link-layer broadcast address as > > > the link-layer destination address. > > > > > > This patch changes destination IP address to 255.255.255.255 when client > > > set BROADCAST flag in their DHCPREQUEST. Note: the offered IP address is > > > still part of the DHCP payload. > > > > > > Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=1801006 > > > > > > Signed-off-by: Ihar Hrachyshka <[email protected]> > > > > Does the reply actually get broadcast to all the VMs in the logical > > network? Or is it just the IP destination that gets changed and it's > > actually unicast just to the requester? The latter is what I'm inclined > > to prefer. > > > > DHCP responder flow sets outport=inport and then runs 'output' so I assume > it's delivered to the original port only, just with broadcast ip address. > (I believe all DHCP responses are looped back.)
Great. (It'd be nice to add that to the commit message.) _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
