Normally, the options ClientAliveInterval and ClientAliveCountMax are set
on the server side, to unspoofably watch for the client not responding.
What I would like to do is be able to make those settings through the
client itself. I am setting up some forwarding where -R or RemoteForward
will be done, which will start listening on the client side. The problem
is that if the network goes down for a while, the client can detect this
via ServerAliveInterval and ServerAliveCountMax settings, but the server
does not (because it is not a host-wide default and I do not want it to
be that for other sessions). This results in ghost server side processes
holding the listen for the RemoteForward port without a way to kill them
off (because the users doing this won't be grantded shell access). So I
would like for them to be able to make sure the server side will clear up
any ghosts listening processes, in a reasonable time (several minutes).

--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
| first name lower case at ipal.net / spamtrap-2008-01-30-1608@ipal.net |
|------------------------------------/-------------------------------------|