Re: [ECOS] DHCP server xid

[Robin Farine <acnrf at dial dot eunet dot ch>]

On Tue, 2002-01-29 at 17:38, Martin Buck wrote:
> Robin Farine wrote:
> > Last year, we observed the case where two nodes booting simultaneously
> > were generating the same XID and then they both accepted the first
> > server reply which resulted in two nodes with the same IP address. I
> > think now that the DHCP client code didn't check the hardware address
> > from the reply.
> 
> Well, instead of "didn't" it might be "doesn't". :-(
> At least I couldn't find any comparison of Ethernet addresses when I
> looked at dhcp_prot.c in the eCos CVS a few minutes ago.

Yeah, I think so.

And about the XID? IMHO, always using the same XID may result in cases
where the client will mix requests and replies, otherwise the protocol
wouldn't define an XID after all. If the client would keep the same XID
until it gets a reply (while respecting the exponential increase in the
delays until it reaches the limit of 64 seconds) and increment the XID
for each reply, wouldn't that solve the problem you observed with the
DHCPDISCOVER packets on plc1g?

Robin