DHCPv6 and MAC Address inclusion

[scott_stone at trendmicro.com]

Yes, I have the perfect use case for this.  I have many servers in my datacenter with multiple NICs, ie "front end" and "back end" networks.  We use a database-driven automatic DHCP configuration architecture for IP management (it's pretty cool, makes it super easy for the datacenter guys to deal with IP assignment without filing tickets for L2/L3... designed it myself) ... and as such I need different DHCP assignments for a single server per-NIC.  Currently I don't see that this can be done at all with DHCPv6 due to its inability to bind to a MAC address and instead binds to the DUID of the server, which ends up assigning the same IP to all the NICs.

 

So that's my huge blocking issue which this (below) would address, with DHCPv6, and I'm betting I'm not the only one.

 

Of course if I missed something here and there's a way to do this with current DHCPv6, I am listening :)

 

====================

Scott Stone <scott_stone at trendmicro.com>

Manager, DCS-RD

Trend Micro, Inc. http://www.trendmicro.com

 

From: dhcp-users-bounces+scott_stone=trendmicro.com at lists.isc.org [mailto:dhcp-users-bounces+scott_stone=trendmicro.com at lists.isc.org] On Behalf Of Ted Lemon
Sent: Monday, January 23, 2012 11:56 AM
To: Users of ISC DHCP
Cc: Users of ISC DHCP
Subject: Re: DHCPv6 and MAC Address inclusion

 

On Jan 23, 2012, at 8:26 PM, "perl-list" <perl-list at network1.net> wrote:

	Do we know if the draft document submission to the IETF regarding the inclusion of the link layer address in the DHCPv6 packets that Frank Sweetser mentions was ever done?  If so, how was the document received at the IETF?

 

There's certainly interest in this, although it's somewhat sad that DHCPv4 is driving features in DHCPv6. 





	Are they planning to amend DHCPv6 specs with the requirement of the link layer address?

 

It could never be a requirement, first because it's too late, and second because doing so would create massive interoperability problems.   It might be added as an option, though.





I don't see how we, at our organization, can proceed with IPv6 rollout using DHCPv6 without being able to somehow correlate both a DHCPv4 and DHCPv6 assignment to a single client device.  

 

Why not?   Can you write up a use case that explains the problem that this creates for you?   It's important that you say what it is that you can't do without this, not simply say "because I need it," which is unfortunately what we usually hear.   The most convincing argument I've heard for this is that the MAC address is what's printed on the box the device comes in, and you can read it and enter it into your inventory with a bar code scanner.   But from the way you said this, it sounds like your use case is different.

 


TREND MICRO EMAIL NOTICE
The information contained in this email and any attachments is confidential and may be subject to copyright or other intellectual property protection. If you are not the intended recipient, you are not authorized to use or disclose this information, and we request that you notify us by reply mail or telephone and delete the original message from your mail system.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20120123/8b98fb14/attachment.html>