[85678] in North American Network Operators' Group
Re: IPv6 news
daemon@ATHENA.MIT.EDU (Jared Mauch)
Sat Oct 15 12:11:14 2005
Date: Sat, 15 Oct 2005 12:08:40 -0400
From: Jared Mauch <jared@puck.nether.net>
To: Mark Prior <mrp@mrp.net>
Cc: Peter Lothberg <roll@Stupi.SE>, nanog@nanog.org
In-Reply-To: <0AA76399-4890-4212-A742-C9C2A3E79619@mrp.net>
Errors-To: owner-nanog@merit.edu
On Sat, Oct 15, 2005 at 08:36:29PM +0930, Mark Prior wrote:
> It might be "closer" if we turned up IPv6 with Sprint but are they
> native yet?
Nope.
http://www.nanog.org/mtg-0405/augmentation.html and
http://www.nanog.org/mtg-0405/pdf/rockell.pdf
Although it's dated, I don't believe it's changed (much) at all.
the added complexity is:
1) software testing/validation
. need to test v6 transport to/from box
. need to test v6 igp (ospfv3, isis, rip, w00t)
. need to test v6 ibgp
. need to test v6 ebgp
2) added link configuration
. need to assign a /31 to link and a /126
3) added network config complexity
. need to add v6 ibgp and ebgp peers
. need to validate nobody forgot step #2 in the path
4) some devices do IPv6 in software, this means using limited
rotuer cpu resources for something that is done in hw for
ipv4.
5) internal policies to make 1-4 happen, including increased
budgets for IT tools, etc.. require serious commitment
at large corporations.
- jared
--
Jared Mauch | pgp key available via finger from jared@puck.nether.net
clue++; | http://puck.nether.net/~jared/ My statements are only mine.
