Thank you for the suggestion. David Starks-Browning wrote:
On Wednesday 19 Feb 03, Jim Kleckner writes:
[snip]
If *all* shells source that line, then you have the problem that the first one to exit will kill ssh-agent, and maybe you don't want that. You could put something a bit more sophisticated in your .bash_profile (or whatever) to identify a "master" shell, depending on your requirements.
Unfortunately, this is exactly what keychain is for - to start the agent in one shell and then share access from many shells. I start up and shut down many rxvt/bash shells during the day without having to provide my passphrase. Stopping the agent with keychain is easy by just typing "keychain -k". I forget to do that just about every time. This is not a big problem. I just thought someone might know how to "strongly detach" the process while at the same time allowing stdin interaction for the passphrase. Jim -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/