Re: [v6ops] New Version Notification for draft-yourtchenko-ra-dhcpv6-comparison-00.txt (fwd)

[joel jaeggli <[e-mail]>]

On 1/8/14, 1:31 PM, Randy Bush wrote:
>>> the multi-million-euro enterprise can continue to do ipv4.  and that's
>>> what they are doing.  we can stick our heads <fill in> and pretend that
>>> this is not a festering and embarrassing problem or we can fix it.
>>
>> Okay, Randy.  You're a smart, articulate guy.  Can you tell us what is
>> missing that we need to add, and in what sense it's missing?  I have
>> yet to hear a clear statement of what the problem is.
> 
> then you are a newcomer to an eight year old conversation.  i will not
> bother with the technical issues which have been discussed endlessly,
> rinse repeat.  that's why we have list archives.
> 
> the bottom line is that the vast majority enterprises, campuses, ... are
> run by IT departments which pay the bills and do it their way.  they use
> dhcp, good, bad, indifferent.  we don't have to like what the people who
> pay us money want to do.  arguing with the customer is generally not a
> good strategy.
> 
> yes there are enterprises who drink the koolaid and follow the true
> religion.  so far, they are a teensie minority.  the vast majority of
> enterprises run ipv4, rfc1918, ...  when they look at ipv6 the first
> question, and often the only question is "how much will it cost?"  and
> the answer to that is bad enough without any speed bumps of changing
> what they are doing today.
> 
> if we want them to deploy ipv6, as opposed to more rfc1918 space, then
> we need to remove all speed bumps.  not discuss them.  not tell them
> that RA is the true religion.  frelling remove the speed bump.
> 
> there is no harm and there is trivial cost in giving the customer what
> she wants.

So not-withstanding my whatever personal feelings I have related to this
dicussion (which mostly involve digsust) there are real consequences for
keying up you radio and sending a dhcp request that may never be
answered every so often until the end of time... This applies in the
ipv4 case for ipv4 adhoc networks as well. but it certainly applies to
ipv4 only subnets with ipv6 speakers (not really an uncommon case), ipv6
adhoc networks, dual stack networks with no dhcpv6 (you should innore
RAs at your peril) and so on.

There's now way to get these things to shut up in ipv4 without managing
them. it would be nice if we could do better than that.

> that the ietf has refused to do so for years is characteristic of the
> arrogance and ignorance which leaves ipv6's survival still at risk.
> 
> randy
> 
> 
> 
> _______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops
>