Re: [radext] configuration auto-discovery [was: draft-cullen-radextra-status-realm-00]

[Bernard Aboba <[e-mail]>]

On Aug 13, 2023, at 13:37, Alan DeKok <[e-mail]> wrote:
> 
> This lack means that any fail-over mechanism is largely implementation-defined.  Which means there are often connection / destination flapping, similar to route flaps.

Fail-over and fail-back have been implemented widely, but I don’t recall much in the way of systematic studies. We built algorithms that worked “well enough” for particular deployments, and then left things alone.

And yes, hysteresis was needed to dampen flapping.

> 
>  There is no way for a proxy to say "I'm alive, but I can't route this particular packet.  Please redirect it somewhere else".  Better connection signalling would solve that issue directly.  A routing protocol would solve it indirectly.

Consider that transport giveup timers are set often based on route flap durations. So a “route unreachable” message could just be evidence of a flap that might resolve before the retransmission timer expires.

>  But both solutions would benefit from clearer and more standardized fail-over, load-balancing, and discovery mechanism.

If there’s enough data to validate an algorithm, this could make sense.