Re: IPv6 Deployment for the LAN

[Joe Maimon <jmaimon () ttec com>]

Iljitsch van Beijnum wrote:
> On 22 okt 2009, at 22:52, Mark Smith wrote:
>
> > > > Seriously, we're all adults.  So treating us like children and
> > > > taking away the power tools is not appreciated.
>
> > > Stop trying to break the internet and I'll treat you like an adult.
>
>
> > Great way to shoot down your own credibility. Just because you don't
> > have or don't understand problems other people have doesn't mean they
> > don't have them or they're invalid.
>
> When people want something which is clearly a bad idea (doing default 
> gateways in DHCPv6 the same way as in DHCPv4) and they ignore it when a 
> better solution is suggested (having a DHCP option that allows a DHCPv6 
> server to tell hosts which of the available routers to use) then the 
> discussion tends to take a turn for the worse.


Thats your opinion. However ICMP router advertisment and before that RIP 
have always been available to provide default router in IPv4 and the 
userbase has already decided which they prefer. History is not on your 
side on this one.

That doesnt mean that DHCPv6 could not do a better job, such as being 
able to configure hosts with multiple specific routes, including a 
default one, or being able to tell hosts to use RA or which potential RA 
learnt gateways should be used and in which preference order.

But requiring default gateway information be learnt from RA and that RA 
be operational for DHCPv6 operation is as clearly to me a bad idea as is 
allowing people to use DHCP with ipv6 in a manner they have come to rely 
on it for IPv4 is to you.

A DHCP server that requires a working router RA is like having a 3 
legged race.

> But wouldn't hardcoding a prefix length also be a bad idea? We now 
> pretty much always have /64 but there are just enough exceptions to make 
> it dangerous to just assume /64.

There is no reason to assume we will be stuck with /64. Once upon a time 
nobody thought subnet masks would ever be anything longer than /24 /16 
or /8 depending on the first few bits in the address. I dont think that 
lasted very long and was completely erased by CIDR adoption.

The one lesson we should be taking from that is not to assign magic and 
sacred powers to bit boundaries.

> However, I firmly believe that whether is done with DHCPv6 it will 
> continue to have problems. Even if DHCPv6 itself would operate well and 
> consistently in all cases, then there are still the permuations between 
> hosts that support stateless autoconfig and not DHCP, those that support 
> both, those that are configured to only do DHCP, those that listen for 
> the M/O bits and those that don't...


So in effect, you are saying that now that this mess has been created 
over peoples strenuous objections that they are forced to live with it?

Thats not going to win any arguments.

And in any effect, you are probably making the point that using non /64 
subnet lengths with a properly operational DHCPv6 is actually a good 
idea for those who wish to completely skip all RA.

Joe