On Mon, Sep 17, 2012 at 9:23 AM, Ryan Rustong <[email protected]>wrote:
> I suggest the following configuration: >> >> hosts/server1: >> >> Address = 1.1.1.1 >> Subnet = 10.18.73.0/24 >> >> server1's tinc-up: >> >> #!/bin/sh >> ifconfig $INTERFACE 10.18.73.23 netmask 255.0.0.0 >> >> hosts/client1: >> >> Subnet = 10.0.2.0/24 >> >> client1's tinc-up: >> >> #!/bin/sh >> ifconfig $INTERFACE 10.0.2.15 netmask 255.0.0.0 >> >> That way, you should be able to ping 10.0.2.15 directly from the server. >> >> I did not realize that the same IP address can be used on multiple > interfaces. Unfortunately this does not work either. > > Ok, this actually did work, I didn't have the subnet mask set quite that large in the tinc-up script. After making that correction I am now able to get to the IP of the tunnel interface (which is the same as the IP on the eth0 interface). I can also get to other IPs in the same subnet by setting a route on them to use the tinc "server" as the gateway. However, I am not able to get to any other subnets by adding additional subnets. Do all subnets that I wish to access need to fall within the same subnet as is defined in my tinc-up script? The problem is that I have some 10.x.x.x addresses and some 192.168.x.x addresses that I would like to be able to access. Also, I am not always going to have the clients using the same IP space on their eth0 interface, which was why I originally defined a separate subnet specifically to be used on the tun0 interface. I have some clients with 10.x.x.x addresses on eth0 and some with 192.168.x.x on eth0. Unfortunately, even after adding those subnets to the host files as you mentioned, I still cannot get it to work unless I use switch mode. Is this a limitation of how tinc works, or am I possibly still doing something wrong? Thanks for your help, Ryan
_______________________________________________ tinc mailing list [email protected] http://www.tinc-vpn.org/cgi-bin/mailman/listinfo/tinc
