[2093] in SIPB IPv6
Re: IPv6 at MIT
daemon@ATHENA.MIT.EDU (Ken Raeburn)
Fri Apr 14 16:34:36 2006
In-Reply-To: <20060413230240.GF23883@gaston.mit.edu>
Cc: jis@mit.edu, sipbv6@mit.edu, Gregory D Troxel <gdt@mit.edu>
From: Ken Raeburn <raeburn@MIT.EDU>
Date: Fri, 14 Apr 2006 16:34:23 -0400
To: Daniel Jared Dominguez <danjared@mit.edu>
On Apr 13, 2006, at 19:02, Daniel Jared Dominguez wrote:
> (I'm cc'ing Jeff on this in case he has any more input.)
>
> What technical obstacles to campus-wide IPv6 deployment exist? It
> seems
> that--given the recent upgrades to the campus backbone a few weeks ago
> and the measures needed to protect MITnet from the IPv6 exploit on
> Cisco
> routers last year--that MITnet's backbone is IPv6 capable.
>
> Are there many non-backbone routers on MITnet that aren't IPv6-
> ready? Is
> IPv6 already carried on the NoX? Are there layer 9 issues?
Well, Jeff could give you the official answers, if any, but my
understanding/impression/guess is (IS&T hat off, random SIPB-alum
geek hat on):
Getting upgrades is all nice and fine, but doesn't substitute for
actual testing of new functionality in a production environment, and
the campus network is critical enough that it wouldn't surprise me if
they want to put off any such testing while classes are in session.
The network group, like most, is understaffed for the amount of work
they'd like to be doing, and IPv6 is not at the top of the priority
list. In fact, aside from Jeff getting me to stop bugging him, I'm
not sure if it's on anyone's priority list. :-) Seriously, how many
students or professors are saying to the network group, "I want to
get IPv6 service in building N"? A non-trivial portion of our IPv6
traffic seems to involve the downstream off-campus tunnels, rather
than on-campus networks.
Harvard's got web pages about the NoX address blocks allocated from
Internet2 and Harvard's IPv6 testbed, but some stuff I've heard and
read (including on the Harvard web pages, describing the addresses as
not "valid" and needing to be changed someday) leave me unsure of the
current status and future of those address assignments.
> Also, by tunnels, I'm assuming you mean only IPv6 tunnels and not
> those
> in the 18.101/16 address space as well.
The IPv6 tunnels are the ones I'm referring to. I don't know
anything new about the status of the 18.101 tunnels.
