monkeysphere ssh-proxycommand should be willing to exec the remaining arguments instead of connecting
monkeysphere ssh-proxycommand optionally has a --no-connect argument, which lets it terminate.
this lets it be incorporated into a larger and more complex proxycommand, but it means usually that the user has to have an external shell script.
I'm proposing adding a new approach where @monkeysphere ssh-proxycommand@ would exec its remaining arguments, something like:
monkeysphere ssh-proxycommand --chain %h %p netcat %h %p
It might be even more elegant to not use the --chain argument as well and just exec the remaining arguments if there are more than the first 2. if there are only the first two, then we would carry on with our internal netcat or socat as usual. And just don't do anything if --no-connect is present.
what do folks think?
(from redmine: created on 2013-02-06)