RE: issue 34 proposal

[email protected] writes:
> there is one situation where "dynamic updates" has clear
> advantage: voice. Dynamic updates survives much faster 
> than IKE approach and call would stay up.

NAT rebooting will cause probably tens of seconds or even minutes of
network lossage anyways, so that will normally cause people to break
the connection and try to reconnect before that. I think most of my
friends will usually allow about 2-10 seconds of bad connection in gsm
phone, before they will hang up and redial trying to fix the problem.
Most of the NATs boot much slower than that.

> And yes, I know VoIP is out of the scope, but still IPsec is used to
> protect voice traffic as well, and if possible that should be taken
> into account.

I do not think VoIP makes any difference here. If the NAT is really
rebootted, it does not really matter if the connection is restored in
120 seconds (when the first ESP packet reaches the other end after NAT
if finally up), or if it takes little longer before the DPD kicks in.

Another case of causing NAT mapping disappearing is the too long
keepalive, but that should not really matter in VoIP traffic, as there
is traffic going on fast enough to keep the NAT mappings up.
-- 
[email protected]