Re: [69ATTENDEES] DHCP

[Roland Bless <bless@tm.uka.de>]

Hi Jordi,

JORDI PALET MARTINEZ wrote:
> I was absolutely unable to get DHCP working during more than two hours
> yesterday afternoon, which is unacceptable in my opinion.

Luckily there weren't such heavy DHCP problems in the past
few IETF meetings. However, we had them some time ago
and are obviously common.

> It will be really nice to make sure that we can progress with more
> parameters available in the RA.

Yes, but host implementations must also be enable to make use
of such features...

> And by the way, unfortunately the DHCP issues are very very very very
> frequent in many networks, not sure if its a protocol/design issue or just
> not so good DHCP implementations, but it is the reality.

I can second that. I guess it is indeed a problem of broadcasting,
per client request processing and packet loss. Stateless autoconfig
seems to be much more robust and efficient in these cases.

Regards,
 Roland