[radvd-devel-l] poor man's routing daemon

[Marc Singer]

On Sat, Aug 20, 2011 at 2:43 PM, Lukasz Stelmach <stlman at poczta.fm> wrote:

> W dniu 20.08.2011 19:37, Sergei Zhirikov pisze:
>
> In the simplest setting (one uplink, one router, one LAN)
> radvd+hosts act exactly the same way as routed broadcasting RIP
> messages to hosts running RIP client listening to them (nb. Win98 or
> even 95 had a RIP client). Routing daemons and protocols provide
> enough information to build a complex routing table but for a LAN
> with a single gateway information radvd provides is just enough.
>
>
I wonder if this is a matter of framing the responsibilities of each
component
to illuminate the assumptions in contemporary network routing.  If I may.

In most networks, hosts only know a couple of things.  They are assigned
or they discover their IP address.  The have a default gateway.  They have
access to a name server.  And in many cases they optimize local traffic
due to the fact that they know the network address of their directly
connected
ethernet segment.  This isn't mandatory, but it is typical.

Routers act as gateways for hosts and for other routers.  They know the
network prefixes that are accessible on each interface.  They also have
a default gateway, a route of last resort, which is the uplink
for leaf networks.  In a multiple uplink setup there could be several
gateway routes though one may be used exclusively as is
probably the case with an IPv6 tunnel.

Radvd is concerned with the self-configuration of host IP addresses, with
selection of the default gateway for hosts, and discovery of the name
server.
DHCP, too, delivers this information to hosts via a different kind
of mechanism.

It is a coincidence, that radvd appears to be able to advertise
inter-network routes
as one might expect a router to do.  Routers will probably take on the
duty of advertising IPv6 prefixes just as routers serve DHCP.  But that is
not
an avenue for blending routing with IP address configuration/assignment.

So, here's the crux.  Switches and routers have merged such that companies
commonly deploy a backbone switch with routing capabilities.  In these
networks,
there is no advantage for hosts to know anything about routes.  The switch
is the
gateway and it has a fast electronics for switching and routing packets.

Home networks tend to use less capable equipment. There may be an access
point
with some routing capabilities, but it is unlikely to be as capable as a
backbone switch.
It probably doesn't terminate the IPv6 tunnel and it probably cannot perform
route
advertisement vis a vis radvd.  Thus, the IPv4 gateway would be different
from the
IPv6 gateway.

It's tempting to think that we could optimize routing in a small network
such that each host knows the best route for traffic.  But there may be no
real
advantage to this.  Consider that uplinks are probably less than 10Mib and
switches
and hosts have 100Mib interfaces.  An extra hop will probably introduce a
small
latency in the traffic, but the router won't be a throughput bottleneck.

So, the apparent simplicity of the corporate network isn't going to be
available to
home networks for some time.  I've resorted to building my routing tables
by hand.  Fragile?  Somewhat.  Effective?  Sure.

Cheers
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.litech.org/pipermail/radvd-devel-l/attachments/20110821/1077a53f/attachment.html>