So Long Last /8 and Thanks For All the Allocations

Re:So Long Last /8 and Thanks For All the Allocations by jlgaddis on Tuesday April 24, 2018 @03:10PM (#16915637)

Well, see, compared to what most of us are used to nowadays, IP networking looked a little bit different back then.

In 1981, RFC791 [0] came about, describing a way of doing IP addressing on the ARPANET (based upon "classful networks" [1]). In accordance with this scheme, you would get either a /8 ("Class A"), a /16 ("Class B"), or a /24 ("Class C") -- depending on how many hosts you had (or thought you might reasonably have). This is the main reason why you see some organizations with a /8 today that you wouldn't expect.

Classless routing ("CIDR") [2] -- or, more specifically, variable length subnet masking (VLSM), what we're all familiar with today -- didn't officially exist until RFC1518 [3] and RFC1519 [4] (fall 1993). CIDR was introduced as a solution to a problem people were already starting to experience then: a shortage of IPv4 addresses!

(Yes, 25 years ago, they realized they were running out of IPv4 addresses and started working on solutions. IPv6 [5] first emerged as a "draft standard" in late 1998 -- but only officially became an "Internet standard" last summer!)

[0]: https://tools.ietf.org/html/rfc791

[1]: https://en.wikipedia.org/wiki/Classful_network

[2]: https://en.wikipedia.org/wiki/Classless_Inter-Domain_Routing

[3]: https://tools.ietf.org/html/rfc1518

[4]: https://tools.ietf.org/html/rfc1519

[5]: https://tools.ietf.org/html/rfc8200

[ Parent | on HN ]

Re:So Long Last /8 and Thanks For All the Allocations by pas on Tuesday April 24, 2018 @06:28PM (#16917212)

How come IPv6 wasn't standardized before 2017? I suspect there's some kind of story there. Do you happen to know the details? Any links/pointers? Thanks!

[ Parent | on HN ]

Re:So Long Last /8 and Thanks For All the Allocations by zAy0LfpBZLC8mAC on Tuesday April 24, 2018 @09:00PM (#16917964)

It's just nonsense. In the IETF's nomenclature, the label "Internet Standard" is used to refer to, and I quote:

   A specification for which significant implementation and successful
   operational experience has been obtained may be elevated to the
   Internet Standard level.  An Internet Standard (which may simply be
   referred to as a Standard) is characterized by a high degree of
   technical maturity and by a generally held belief that the specified
   protocol or service provides significant benefit to the Internet
   community.

   A specification that reaches the status of Standard is assigned a
   number in the STD series while retaining its RFC number.

Most internet standards are not "Internet Standard"s, as can be seen by the fact that IPv6 got assigned the standard number 86, and up until last year, IPv6 wasn't either, but it still was a perfectly fine and well-documented standard, and has been for a long time.

[ Parent | on HN ]

Re:So Long Last /8 and Thanks For All the Allocations by pas on Wednesday April 25, 2018 @12:06PM (#16922902)

I'm sort of familiar with how sacred these standards are. I'm interested in what was the deciding factor, when IETF and which WG decided now we have enough (significant) successful operational experience.

It's surprising, because IPv6 was set in stone at least around 2010 already, with a lot of experience (6bone ended in 2006, and so on), but on the hand general rollout is still very much ongoing.

[ Parent | on HN ]