DHCPv6 and MAC Address inclusion

Simon Hobson dhcp1 at thehobsons.co.uk
Wed Jan 25 10:05:29 UTC 2012


Ted Lemon wrote:

>>I just don't understand why it was chosen that 
>>mac address (read: link layer identifier) was 
>>left out of the DHCPv6 packet.  What was the 
>>reasoning there?  What was the harm in 
>>including it?  Is it not better to have to much 
>>information rather than to little?
>
>If you can count on the MAC address being there, 
>you can use it as an identifier in a 
>non-conforming DHCPv6 implementation.   We 
>didn't want that, so we couldn't specify it in 
>such a way that that would be possible.

I find that a bit strange. Surely the best way to 
deal with that is to "name and shame" 
non-compliant implementations, rather than deny 
everyone something that could be useful "just in 
case". Lets face it, there are still plenty of 
ways to make a non-compliant implementation - 
such as splitting an LL or LLT identifier to 
extract the hardware address, something that is 
now encouraged by it not being present in the 
request packets in it's own field/option.


>I get the sense that you're experiencing a bit 
>of fear, uncertainty and doubt about the IPv6 
>transition.

I think everyone is !
It's easy to forget that IPv4 used to be as hard 
as people are finding IPv6 - at least for those 
of us with "a few years" under our belts. I 
recall my first contact with IP - this strange 
system, with funny numbers, something called a 
netmask, and seemingly (in the days before search 
engines) no documentation to be found. Back then, 
getting online meant a SLIP dialup from a single 
machine - putting a network online meant running 
a Unix box (or very expensive router) to do the 
routing.
The idea of being able to unpack a small plastic 
box, put in a username and password, and get a 
whole network online in minutes was just science 
fiction.

So here we are, in much the same position with 
IPv6. Some of the principals are the same (eg 
addressing and subnet masks), but a lot is so 
very different. A small number of people have 
been working on this for some time, the rest of 
us are having to start from scratch - and are at 
different stages of learning.

So I think it's OK to be confused by it all. I'm 
sure it will become clearer as it gets more 
widespread and the tools catch up. At the moment, 
I think we're still at the stage where people 
aren't quite clear what tools are needed, and how 
they should work - a bit of chicken and egg; 
until adoption is more widespread, it's not clear 
what's needed; and until the tools are there, 
adoption is a bit harder.

My ISP (Plusnet) is doing a trail of IPv6 to end 
users. "It works just fine", but I'm still 
struggling with some of the pieces. The ISC 
client doesn't support PPP interfaces, and 
Wide-DHCP client doesn't appear to support the 
scripting ability I *think* I want. So there's an 
element of experimenting and adjusting my 
requirements to see if I can close that gap.



José Queiroz wrote:

>Hi, sorry if this is a naive view. But, doesn't 
>auto-assignment with EUID-64 suffixes attends 
>your needs? This way you'll have a fixed IPv6 
>address, which haves a strong association with 
>the MAC Address, and better, is predictable if 
>you know the MAC address of the interfaces.

it will work for some, but it's not a good idea 
for public services. Simple example, what if you 
want to move a service to a different machine ? 
Using EUID-64 addresses means you either have to 
change the MAC on the new host to match, or 
change the DNS and wait while it propagates - 
with a day or 2 of parallel running while caches 
expire.
In practice, I suspect we'll start to see a 
return to using multiple addresses per 
machine/interface - which at one time was the 
only way to run multiple web sites on one host. 
So there may be some scope for using an EUID-64 
address for the machine, and additional addresses 
for the service(s) run on it - but that's a whole 
new level of management system to sort out. I'm 
not looking forward to applying IPv6 at work - I 
suffer badly enough with people that won't apply 
any logic to IPv4, I hate to think what they'll 
do when let loose with (say) 256 networks of 64k 
addresses !

-- 
Simon Hobson

Visit http://www.magpiesnestpublishing.co.uk/ for books by acclaimed
author Gladys Hobson. Novels - poetry - short stories - ideal as
Christmas stocking fillers. Some available as e-books.



More information about the dhcp-users mailing list