On 10/17/06, Stefan Tunsch <[EMAIL PROTECTED]> wrote:
I'm not sure I get your point.
You always need to provide your IP address to the Dynamic DNS provider. If
you don't, it will cease to be dynamic. The ways to do that vary. You can
for example simply do a http request to some special page of your Dynamic
DNS provider passing along your credentials and the page will automatically
know your IP address, compare it with the stored one and eventually update
it if there has been any change.
This would be the method I described that would solve your issue.
DynDNS does indeed provide with a mechanism to manually introduce an IP, but
the problem here is the ability of pfSense to update this entry if the
public IP address changes.
This is how we do it today.
Regarding the routing, I'm not sure what you mean. Holger has clearly stated
that indeed the dynamic DNS service of pfSense only checks and updates the
public IP address on the WAN interface and that in the future pfSense will
get the functionality to choose the interface you want to update with this
service. (Multiple dyndns clients updating different interfaces would be
nice...)
Maybe I confused you. Policy routing using pf rules (as you note
further down). The pfSense box itself uses the routing table on the
box and not the policy rules. So, just because your WAN interface has
your default route, it doesn't mean that the WAN interface is the way
to reach your destination. pfSense connections will source from the
closest IP to the destination (usually this is the WAN IP).
On the other hand, policy based routing of some traffic through one or the
other interface can be controlled by adding a simple rule to the firewall.
--Bill
