Routing over Demand Circuits

[Gerry Meyer <[e-mail]>]

TCP and Stale information - an explanation.
------------------------------------------

Take an example:

Suppose I have a router with a LAN interface and a WAN interface configured
to exchange routing information with N other WAN sites.

Let there be a temporary glitch on the WAN - insufficient time for routes to
time out.   Say the cleaner knocks the cable out for 60 seconds.

This happens to coincide with a series of events on the LAN  which require
an update to be sent on the WAN.  Say this results in M different updates
being sent to the WAN.

The following has happened:

* I have clogged up my slow WAN link with useless information that I would
  *prefer* had done the honourable thing and died.

* I have used up memory corresponding to M * N * Size of update message(s).
  If I cannot get memory for the latest update I've blown it.

Tony Li used a spurious argument about BGP (in his last posting):

>   ?  Processing the stale information and then overwriting it with
>   current information is not a concern.  If this was a serious problem,
>   BGP would have already stopped working a long, long time ago.

The *key* difference is BGP does incremental updates:

* so all things being equal (similar overall number of routes) requires
  to transmit less 'routes' than RIP in the event of an update being
  required.

* but more importantly the 'stale' information MUST be transmitted because
  it is *required* for routing table consistency.

    Gerry