On-going work on DHCP project : RFC4361 support
thomas.pegeot at sogeti.com
Mon Mar 15 15:55:33 UTC 2010
Actually, the changes I brought to the DHCP source code dont force DHCPv4
clients to use DUID based client-identifiers in order to keep backward
compatibility. Forcing every user to migrate to this new kind of
client-identifiers would be a mess. Thats why Ive chosen to enable the use
of a DUID based client-identifier only if the required argument is passed to
the dhclient executable.
Regarding the DHCID, I intended to add a new compilation option (Ive
already started to implement it) that would enable the support of real DHCID
records. This option would be by default disabled: in this case, the old TXT
records would be used.
Guys, what do you think about this?
De : dhcp-workers-bounces+thomas.pegeot=sogeti.com at lists.isc.org
[mailto:dhcp-workers-bounces+thomas.pegeot=sogeti.com at lists.isc.org] De la
part de Rene Garcia
Envoyé : lundi 15 mars 2010 16:26
À : dhcp-workers at lists.isc.org
Objet : Re: On-going work on DHCP project : RFC4361 support
Here is some additional information on Thomas'work : the stored DHCID in
DNS servers will change and therefore upgrading to this new behaviour will
need a lot of work to migrate old emulated TXT DHCID to new ones. That's why
I aked him to add support for real DHCID RRs in DNS as ISC BIND 9.5+ seems
to handle them.
Including Thomas'code change into the CVS tree is a major change and creates
an incompatibility with existing DNS records. How can we handle this ? Maybe
using an optional compilation using a new --enable-feature option in
configure script ?
Therefore I hope his work will be usefull for the DHCP project.
Quoting Thomas PEGEOT <thomas.pegeot at sogeti.com>:
> As part of my final year project (French equivalent of a Master degree in
> computing science), Im working on an ipv4 to ipv6 migration research
> First, I was asked to work on ISC DHCP in order to make dual stack clients
> be able to identity correctly for DDNS updates. This consists in solving
> problem described in the Dual IPv4/IPv6 Client Considerations part of
> RFC 4703 and therefore implementing the RFC 4361. It aims at allowing
> clients to be identified by an IAID and a DUID, and in that way having
> DHCPv4 and DHCPv6 clients compute the same DHCID from a common DUID.
> The modifications I wrote are currently reviewed by my tutor (René Garcia,
> who already reported bugs here). Afterwards, Im going to send these
> to this mailing list in hopes of having them merged with the CVS code.
> As the DHCID support is not implemented yet in ISC DHCP, Im going now to
> study this problem and try to write down patches in order to solve this. I
> will keep you informed about this.
> I hope that you will be interested in my work.
> Thomas PEGEOT
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the dhcp-workers