Hi, a new patch set from Heiko is coming up, which is going to change the "IFCONFIG_BEFORE_TUN_OPEN" to "what everyone else does" (because what we currently have is older than anyone remembers, and makes life much more complicated than needed).
Now, we are not currently properly testing Windows - on the unix systems,
I'd build a t_client.rc test set with like 5 extra tests for "--route-method
adaptive/ipapi/exe", "IPv4 only", "IPv6 and IPv6", interactive service
or not, etc. - but I'm not sure what the best way is to do this on
windows.
Would "use mingw/msys bash" work for stuff like "put regular openvpn.exe
into background with 'openvpn.exe & pid=$! ' and kill with "kill $opid"?
I have been toying with the idea of replacing t_client.sh with a perl
script that uses the management interface to control openvpn (at least
the "has our startup finished? have we exploded? shutdown orderly now!"
bits...) - but that's also not trivially implemented as one would have
to redo the whole "look at ifconfig/route output and parse in a system
dependent way" - and test on a gazillion of platforms :-)
So... other ideas? Selva, Valdikss, how do you test windows builds?
Jens, since you have dived into t_client improvements - any nice ideas
how to make this happen with "reasonable effort"?
gert
--
USENET is *not* the non-clickable part of WWW!
//www.muc.de/~gert/
Gert Doering - Munich, Germany g...@greenie.muc.de
fax: +49-89-35655025 g...@net.informatik.tu-muenchen.de
signature.asc
Description: PGP signature
