[Openswan dev] Error building klips-ipv6 (missing include?)

[Harald Jenny]

On Mon, Oct 11, 2010 at 03:58:19PM -0400, D. Hugh Redelmeier wrote:
> | From: Harald Jenny <harald at a-little-linux-box.at>
> 
> | On Mon, Oct 11, 2010 at 02:42:28PM -0400, D. Hugh Redelmeier wrote:
> 
> | > Off the top of my head, I can vaguely recollect two issues:
> 
> 
> 
> | > - we want control of which interfaces are used.  The only way was to
> | >   control the IP addresses
> | 
> | Hmmmm you mean for NETKEY? Because the interfaces= lines for KLIPS do work?
> 
> No, my comment was all about IKE messages: strictly Pluto.

Ok thanks for making this clear.

> 
> That does not mean that there are no Pluto/kernel issues, just that I
> didn't mention them (and don't remember any).

I see

> 
> | > - we want to know the inbound IP address on each packet.  recvfrom
> | >   doesn't give you that, so we nail down which IP by only accepting
> | >   a specified IP address on each socket.  So we have to know the IP
> | >   address.
> | 
> | Well if we specify an IP which is not bound yet wouldn't we be able to check
> | in the main loop of pluto if/when this IP "arrives"?
> 
> I don't know what that means.

That it wouldn't be necessary to listen on ANY IP but rather detect if a new IP
gets assigned to an interface?

> 
> If a UDP packet comes in on an ANY, we don't know its source address.
> That would seem to be a problem.

Yes for sure.

> 
> Another arcane fact: because Pluto only listens to what it is told to
> listen to, it is actually possible to run more than one Pluto on a
> system.  I used to do that for regression testing.

Sounds like changing this might break some setups...

> _______________________________________________
> Dev mailing list
> Dev at openswan.org
> http://lists.openswan.org/mailman/listinfo/dev