[26113] in resnet
Re: IPv6
daemon@ATHENA.MIT.EDU (Ingen Schenau, Jeroen van (ICTS))
Fri Mar 25 10:20:57 2011
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 8bit
X-UTwente-MailScanner-From: j.vaningenschenau@utwente.nl
Message-ID: <1301062813.27533.88.camel@icts-sp-039>
Date: Fri, 25 Mar 2011 15:20:13 +0100
Reply-To: Resnet Forum <RESNET-L@listserv.nd.edu>
From: "Ingen Schenau, Jeroen van (ICTS)" <j.vaningenschenau@utwente.nl>
To: RESNET-L@listserv.nd.edu
In-Reply-To: <532BD1A910E6AD47A472B02C33F5D26005CB7D@Marlin.stonehill.edu>
Hi,
> I searched the achieves but did not see this post. Given all the
> resent interest in IPv6 since the last blocks of IPv4 addresses were
> handed out I thought it would be interesting to find out the
> following.
>
> - Have you Migrated to IPv6?
We have been offering native IPv6 on all wired segments for a couple of
years now, so all clients can get services over IPv6. However, most of
our internal services (mail, website etc) aren't v6 enabled yet.
> - Do you have plans to migrate in the next year or two?
IMHO, migration is a long process. If you haven't started with IPv6 yet,
the sooner you start reading up (and planning implementation) the
better.
There's no "big bang" to switch over; running IPv4 and IPv6 alongside is
the way to go for the next couple of years. Just make sure your clients
can use both protocols and slowly start making services available over
both; you'll see that traffic levels on v4 will slowly decrease in
favour of IPv6.
> - Have you thought about it but hoping not to migrate for a
> long time?
The longer you wait, the harder it's going to hit you. The rest of the
world is moving forward, don't stay behind... Our first experiments with
v6 were in 2000; for clients it's been in production for 2 years now.
> - To the questions above can you please say why or why not?
Simply because IPv4 is a dead end. Its life span has been stretched over
and over again, but nothing can be stretched indefinitely: it will
eventually tear and rip apart.
I'm not saying that IPv6 is perfect, but it has been designed to scale
much further than IPv4.
> We currently have no plans to migrate but I feel as if it’s something
> I need to further investigate. I’m most interested in the question of
> when should we migrate.
You should start today (even better "yesterday") :)
Start with a very limited deployment. Suggested first steps:
1) Get external IPv6 connectivity (Internet2 and/or commodity provider).
2) Read up on numbering schemes and security best practices for IPv6.
3) Join IPv6 related mailing lists.
4) Enable native IPv6 on the segment used by your IT department.
5) Exchange experiences with other institutions.
6) Enable native IPv6 for power users, eg CS department.
7) Start enabling IPv6 on selected servers and publish the relevant DNS
records (AAAA, if you're running AD it will happen automatically).
8) Keep adding client segments and services until everything is dual
stack.
9) You're done until IPv4 is legacy enough to start turning off IPv4
support...
As with everything "new", just start by dipping your toes and eventually
you'll learn to swim. It's easier that way than staying on the dry until
you can't wait any longer and suddenly you have to dive...
Regards,
Jeroen van Ingen
ICT Service Centre
University of Twente, P.O.Box 217, 7500 AE Enschede, The Netherlands
___________________________________________________
You are subscribed to the ResNet-L mailing list.
To subscribe, unsubscribe or search the archives,
go to http://LISTSERV.ND.EDU/archives/resnet-l.html
___________________________________________________
