DHCPDISCOVER coming in on server's native interface

[David W. Hankins]

On Thu, Oct 29, 2009 at 03:47:10PM -0300, Bruce Hudson wrote:
>     As far as I can see, the server logs "via eth0" is there the GIADDR
> field in the packet is not set. While this normally means the packet
> was broadcast on the local network (assuming well-behaved clients) it
> seems a unicast DISCOVER from elsewhere on the network would trigger 
> the same behaviour. 

This is true.  The only exception is DHCPREQUEST packets when the
client is RENEWING.  In this case, the client unicasts their
DHCPREQUEST directly to the DHCP server; it does not go via a relay
agent, so giaddr is not set.  These messages will also say "via eth0",
but of course the client is not actually directly attached on eth0.

Obviously the OP is looking at DHCPDISCOVER packets, which are always
broadcast and never unicast, so he probably really does have local
clients on that wire for some reason (or some really weird broken
client or tool that is unicast a DHCPDISCOVER!), but I wanted to say
the exceptional case for the benefit of googlers hitting the archives.
:)

-- 
David W. Hankins	BIND 10 needs more DHCP voices.
Software Engineer		There just aren't enough in our heads.
Internet Systems Consortium, Inc.		http://bind10.isc.org/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20091119/d4d569c1/attachment.bin>