Re: IPv6 addresses really are scarce after all

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




I am quite startled by this thread, both in its emotion
and in the apparent oversight of multiple approaches
to the issue of having LOTS of connected user devices
at a house/site/office when an IPv6 /64 prefix has been
provided by one's upstream network provider.

First, giving each end user a /64 means that the user can
have up to 2^^64 devices at their site/home/office.  That
is a whole lot of devices -- more than any one house/site/office
is going to have.  Just as an example, all the IEEE 802
MAC addresses that ever will be issued (now or future)
only would amount to 2^^48 (much smaller than the 2^^64 limit)
since IEEE 802 uses 48-bit MAC addresses -- and a number
of those MAC addresses are reserved for special purposes
(e.g. link multicast MACs, local-scope MACs).

Second, Ethernet bridging is a well understood technology
and it works just fine even with very large numbers of devices.
Some years back I was at a research lab (~2500 people on site)
where virtually everyone had at least one IP connected
computing system (some had more than one).  About 1/4 of that
site was connected via a big yellow Ethernet cable running
classic CSMA/CD Ethernet at 10 Mbps (half-duplex).  We weren't
alone in having very large bridged networks.  A typical house
works fine just with very-low-cost Ethernet switches or hubs
internally and perhaps a very-low-cost IP router between the
house/site/office and the uplink device (e.g. cable modem,
DSL modem, FIOS ONT).  Most wireless access points are not
merely radio repeaters, but also include bridge capability,
so that traffic not needing to go over there air won't.
One nice thing about using bridging is that it doesn't really
care whether the upper layer protocol is IPv4, IPv6, IPX, or
(gasp) CLNP.  "Ethernet just works." (TM)

Third, DHCP is a well understood technology that is deployed
at millions of sites world-wide and generally works quite well.
Those very-low-cost Ethernet routers one can purchase generally
include both DHCP Client (for the uplink interface) and DHCP
Server capabilities (for user devices).  These are normally
configurable.  So one ought to be able to use DHCP to provision
IPv6 addresses (out of that overall block of ~2^^64 IPv6 addresses
mentioned above) using say 16 bits (bits 64-79) for IPv6 subnetting
and the remaining 48 bits for naming IPv6 interfaces.
[NB: While I personally remain a fan of the 8+8 architecture,
so far the IETF has declined to formally adopt that,
so that ought not be an issue for this kind of deployment today.]

Fourth, lots of folks (me included) happen to find it convenient
to use NAT between my site/house/office and my upstream provider.
I'm a bit sceptical of the extravagent security claims sometimes
made for NAT, but I do find it convenient.  Mind, with the other
options above, no one *needs* to use NAT with IPv6 even if their
site/office/house gets a /64 routing prefix.

Now I do think that folks here have overlooked one lurking issue.
So far, few of the very-low-cost Ethernet switch/IP router boxen,
or the very-low-cost wireless bridges/router boxen, or the
very-low-cost security-gateway boxen, support IPv6 at all yet.
One imagines that this is because the vendors of such devices
don't perceive a current large market demand for IPv6 capability.
So the IPv6 advocates might want to go work on making such
very-low-cost boxen available.  If IPv6 really takes off, there
will be money to be made in that market segment.

In turn, the paragraph just above means that most home users
(a price sensitive lot if ever I've seen one) today will be using
some sort of old home-brew BSD/Linux PC-hardware as their
router/bridge.  This means that one has even MORE flexibility
than using the very-low-cost widgets, since shipping BSD/Linux
has a pretty full set of IPv6 capabilities, IPv4 capabilities,
and not a bad set of Ethernet bridging capabilities.  Mind,
such boxes are not easy for novice users to setup and deploy.

So folks, not to worry, there are a number of off-the-shelf
network deployment approaches that are practical and affordable
today to enable significantly sized IPv6 networks to be built
using a single /64, including options that support IPv6 routing
and sub-netting.

:-)

Cheers,

Ran Atkinson
rja@xxxxxxxxxxxxxxxxxxx



_______________________________________________

Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf

[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Fedora Users]