[RRG] reachability

[RJ Atkinson <rja@extremenetworks.com>]

Earlier, Brian Carpenter wrote:
% Is it OK if only 95% of telephones can be called, too?
Brian,

  My guess is that at present my own telephones can reach
less than 90% of the total PSTN-connected telephones deployed
around the globe.

  There are several *countries* worth of telephones that
simply are not reachable from my telephones, whether for
policy reasons (e.g. governmental/political policies)
or because there are no inter-connection agreements
(equivalent to "peering agreements"), or for other reasons.

  Similarly, the global Internet has not been fully
connected for at least a decade now.  I would guess that
one would have to go back to EGP days (i.e. before BGP
deployment) to have a fully connected Internet, and that
was a wildly smaller Internet than we have now.

  A common reason for lack of complete connectivity is that
many IP routing prefixes are only advertised regionally
(example: they might be advertised within Asia/Pacific,
but are not reachable from or advertised within North
America).  Other reasons might include political policy
(e.g. government regulations on which other countries are
authorised to be reachable) in various places, security policy
(e.g. firewalls implementing organisation-specific policies),
or the deployment of NATs (which can sometimes break
reachability on a relatively fine-grained basis).

  Most people *perceive* the Internet to be fully-connected.
I suspect they have this perception for multiple reasons.
Two likely reasons are (1) some reachability might temporarily
be gone; users can't readily distinguish "never reachable"
from "not reachable just now" and (2) the most popular
destinations tend to be very highly reachable (although some
countries have national policy blocks on their users reaching
certain highly popular destination web sites -- example:
www.CNN.com is usually not reachable from inside the PRC).

CONCLUSIONS:

1)  I think that it is beyond the scope of the RG to insist
that universal reachability exist in future, given that it
does not exist at present.

2)  Further, the details of reachability are very likely
to be *engineering* details, whilst the RG is only chartered
to make an *architectural* recommendation to the IETF
(i.e. not a protocol recommendation).  Ultimately the IETF
will either act upon, ignore partly, or ignore completely
whatever recommendation this RG might make.

Yours,

Ran
rja@extremenetworks.com



--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg