----- Original Message ----- From: "Tjapko Smits - iTS Consultancy" <itsc99@hotmail.com> Sent: Friday, August 22, 2003 11:19 PM Subject: Re: [Openh323gk-users] Strange test scenario > The part I don't understand is that same scenario works perfect when both > GK's are in public C. It does not work when the slave Gk is placed in the > routers DMZ with all ports allowed and no firewall activated for this test. > The slave GK does actually come up with a Location confirm but than it dies > immediately because parent site's router don't know where to put the packet. > > I would like you to verify the sequence of events. Do you have the same > scenario? I have exactly opposite scenario. I tried to dial from EP2 registered with child GK (slaveGK) to EP1 registered with parent GK. It also does not work (tpc connectio fails - I will try to use some packet sniffer to find the reason). Your problem is a bit different. Do you have properly configured your child GK (not as neighbour, bu as child GK - [Endpoint] section?). Normally, child GK behind NAT establishes tcp connection upon startup to be able to receive signalling messages (Setup,etc.), otherwise parent GK will not be able to setup signalling channel to child GK (destination unreachable). I think either: 1) your child GK is not running NAT thread to handle signalling channel (see trace files); 2) there is a bug in GK regarding handling NATed child-parant GK communication. --- Michal Zygmuntowicz ------------------------------------------------------- This SF.net email is sponsored by: VM Ware With VMware you can run multiple operating systems on a single machine. WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines at the same time. Free trial click here:http://www.vmware.com/wl/offer/358/0 _______________________________________________ List: Openh323gk-users@lists.sourceforge.net Archive: http://sourceforge.net/mailarchive/forum.php?forum_id=8549 Homepage: http://www.gnugk.org/
