Re: shim - transport/app communication

[Iljitsch van Beijnum <iljitsch@muada.com>]

On 15-mrt-05, at 0:19, Leif Johansson wrote:

> > Shim-aware transports/applications could also emit periodic 
> > keepalives when they're otherwise idle so the shim knows it shouldn't 
> > remove state yet. (In fact many applications and/or TCP already do 
> > so.)

> That is typically done as a hack to keep NAT-boxen from ditching you
> right?

No, I think it's done (at the tcp level) to flush out dead sessions. I 
rather like being able to put my laptop to sleep overnight and find my 
SSH sessions still alive, though.

> Some applications do that, others simply break.

Are you saying that NAT breaks certain applications? That surely can't 
be...

> I believe that keeping applications shim-unaware is a worthy goal.

Some applications may want to inform the shim of their special needs. 
But applications that don't have special needs shouldn't have to be 
shim-aware, of course.