[109993] in North American Network Operators' Group
Re: What is the most standard subnet length on internet
daemon@ATHENA.MIT.EDU (bmanning@vacation.karoshi.com)
Fri Dec 19 01:10:16 2008
Date: Fri, 19 Dec 2008 06:08:44 +0000
From: bmanning@vacation.karoshi.com
To: =?utf-8?B?7KCV7LmY7JiB?= <lionair@samsung.com>
In-Reply-To: <11527796.20751229654447763.JavaMail.weblogic@epml12>
Cc: nanog@nanog.org
Errors-To: nanog-bounces@nanog.org
On Fri, Dec 19, 2008 at 02:40:47AM +0000, l l9l wrote:
> However, what I am really wondering is what is the most standard subnet length that always can be guaranteed through Internet. less than /24 bit ?
>
while one can get away w/ /24s (if that is all one has) for many places,
I suspect that there will be increasing pressure to drop more specific
/24s as folks routing tables grow.
your question, "...length that can be guaranteed through the Internet."
argues for fairly short netmasks, e.g. a /16 is likley to be accepted
by most folks while very short masks, e.g. /8 or smaller are likly to
be seen with some level of consideration since so very few prefixes of
that size are likely to be origin-sourced (often proxy aggregates from
transit parties)...
as others have pointed out - this "acceptable" value is fluid, changing
over time and variable between ISPs. Creating a static policy is likely
to be flawed.
--bill (crawling out from under his rock, blinking in the bright lights)
