Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 17 Apr 1996 23:30:24 -0500
From:      Jim Fleming <JimFleming@unety.net>
To:        Gary Palmer <gpalmer@FreeBSD.ORG>, "'Jim Binkley'" <jrb@cs.pdx.edu>
Cc:        "alf@bolzen.in-berlin.de" <alf@bolzen.in-berlin.de>, Dirk Froemberg <dirk@methan.chemie.fu-berlin.de>, "hackers@FreeBSD.ORG" <hackers@FreeBSD.ORG>, "xadmin@methan.chemie.fu-berlin.de" <xadmin@methan.chemie.fu-berlin.de>
Subject:   RE: ip-in-ip tunnel 
Message-ID:  <01BB2CB5.DB907200@webster.unety.net>

next in thread | raw e-mail | index | archive | help
On Wednesday, April 17, 1996 6:27 PM, Jim Binkley[SMTP:jrb@cs.pdx.edu] wrote:
@
@As another possibility,  I can give you some experimental
@kernel code that basically is integrated with the multicast
@tunnel code in the kernel and gives you a device called
@"mvif0" that allows you to put a route between two boxes
@and route all packets to a destination X between a tunnel system Y.
@Logically:
@	send all unicast packets to X via ip dst Y (ip dst in outer
@	ipip protocol packet)
@
@This is for support of mobile ip, what at this point I can
@at least do by statically inserting routes in the kernel. 
@I sent about 20000 packets through it yesterday so it isn't
@that bad, but it's still experimental.
@
@Depends on what you want to do.
@
@				regards,
@
@				Jim Binkley
@				jrb@cs.pdx.edu
@
@

It appears that many people are discovering that there is tremendous
potential in using the Internet to "re-invent the net".

This is a natural outgrowth of the success of the net. The Internet model
is being applied to the Internet itself to explore new ways to expand the
net. This keeps everything evolving away from centralized solutions which
the net has successfully challenged over the years.

Expanding on this theme...

Imagine that you have several networks scattered around the world. Imagine
that they run some strange protocol called IPv8 that is very similar to IPv4
and requires minimal changes to FreeBSD to support.

	see <http://comm.unety.net/US/IL/Naperville/Unir>; for IPv8 info

Imagine that IPv8 supports 2,048 different 32 bit address spaces and that
the goal is to tunnel between all of these networks in an efficient manner
to provide global connectivity for an OuterInternet that supports many more
sources and destinations than the current Internet.

Imagine that you decide to use the Legacy Internet as purely a transport
vehicle between large IPv8 islands or Galaxies. Imagine that the challenge
is to build a FreeBSD based system that can act as a gateway (or StarGate)
for these Galaxies and provide inter-Galaxy communication.

With a substantial increase in available "IP addresses", imagine that you
could easily obtain your own permanent /16 (formerly Class B) address to
allocate to friends, families, clients, customers, dogs, cats, etc. as you see fit.
These would be allocated to people that help with the system and express
a need and interest.

Imagine that there are 8 Galaxies and 256 Stargates per Galaxy (8x256=2,048)
Imagine that someone is willing to donate 8 fully loaded $8,000 FreeBSD-based
Pentium systems to be located around the world for testing this network. Imagine
that FreeBSD users could use these systems not only for testing various tunneling
strategies but also as FTP servers, mirror sites and other purposes.

The $64,000 question is...

Is the FreeBSD community cohesive enough to build an OuterInternet to
not only support its current development but also to explore the advantages
of re-inventing the net by using the Internet as a transport for a semi-private
FreeBSD net?

I would be curious if people think this has merit, is doable, is a worthwhile
project, would be interesting to students, would help the FreeBSD effort, etc.

I would also be curious whether people would be willing to host a StarGate
and where people think these StarGates should be placed in the Galaxies.

I am also curious how people would propose that these StarGates interchange
routing information and utilize the ip-in-ip tunneling which is where we came in...

...beyond this I am curious about people's ideas about distributed object
technology that can be transported on the FreeBSD network...we will save
that for later...if and when the net starts to grow out of the FreeBSD efforts...


--
Jim Fleming
UNETY Systems, Inc.
Naperville, IL 60563

e-mail: JimFleming@unety.net




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?01BB2CB5.DB907200>