Efforts have recently produced an OS/2 port of the OpenVPN client
but the reports of it working sound like test settings with fixed IPs
and the tunnel not crossing the public internet.
I'd like to use it to connect from home to the openVPN server at my
office, to access various internal http based tools, but haven't quite
been successful. The OS/2 home client fails when it tries to configure
its own ip and routes. I also have an XP client that does connect and
so gives a sort of model for how things should go, but I'm not clear on
how I should set up the OS/2 virtual adapter, whether with my real
adapter on DHCP the virtual one can self-configure, whether the tap-only
limitation I read about still exists, or matters, etc.
Before I try to wrestle with details, has anyone actually set up this
client to work with a "normal" remote system like mine?
I'm using last-free-version OS/2 FP15 and the old TCP/IP (inetver 6.2).
The OpenVPN client is 2.1.3 from netlabs 9/2010 but the virtual
adapter driver is from the hobbes archive from 2006.
Good day, Peter.