It seems odd to have an in-process SSH library that is doing our connecting, only to have it spawn out to a required "ssh" executable.
I thought the point of using gocrypto's ssh was only when an actual "ssh" wasn't available, which means ProxyCommand doesn't seem very useful.
I realize there could be other use cases for ProxyCommand, though I haven't seen them myself.
Is there a different use case that I'm missing?
Is this intended so that we can proxy via the API server?
It seems odd to have an in-process SSH library that is doing our connecting, only to have it spawn out to a required "ssh" executable.
I thought the point of using gocrypto's ssh was only when an actual "ssh" wasn't available, which means ProxyCommand doesn't seem very useful.
I realize there could be other use cases for ProxyCommand, though I haven't seen them myself.
Is there a different use case that I'm missing?
Is this intended so that we can proxy via the API server?