You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First of all, thanks for making this amazing tool! It's elegant in design and neat in implementation.
One of the intended use is
Access remote computers behind NATs and firewalls
but currently it is not the ideal tool for cases like accessing home servers from outside or troubleshooting deployed devices behind NAT because:
The host is assigned a different session ID each time it connects to the server, so an out-of-band channel is required to publish the session ID so people can log in. If ID clash is not a concern, it would be more convenient if one can simply specify the ID/name of the host to log in.
The session is deleted when the first logged in SSH session exits. What if we delete the session only when the host exits? I can imagine that to be useful in pairing or running a demo too - guests can join or leave without affecting the session. Or am I missing some important context here?
Person on the host side needs to press a key to start accepting connection. This is easy to change by introducing a flag.
So to support daemon mode, it would require flags on both host and server side for 1 and 3, but would also need to change the termination behaviour of 2. Before I get my hands dirty I'd like to ask 1) does the above make sense, and 2) is this daemon mode idea fits in the scope of upterm or I'd better assume it to be a "hard fork"?
The text was updated successfully, but these errors were encountered:
First of all, thanks for making this amazing tool! It's elegant in design and neat in implementation.
One of the intended use is
but currently it is not the ideal tool for cases like accessing home servers from outside or troubleshooting deployed devices behind NAT because:
So to support daemon mode, it would require flags on both host and server side for 1 and 3, but would also need to change the termination behaviour of 2. Before I get my hands dirty I'd like to ask 1) does the above make sense, and 2) is this daemon mode idea fits in the scope of
upterm
or I'd better assume it to be a "hard fork"?The text was updated successfully, but these errors were encountered: