When we know where we want to connect, we may want to retry a number of times until we succeed. This would allow the server and client to be started in parallel. ie: https://github.com/mviereck/x11docker/issues/167#issuecomment-506864026
There should probable a configurable limit of retries.
We now retry up to XPRA_CONNECT_TIMEOUT (defaults to 20 seconds):
Not doing VSOCK or named-pipes.
Improvements:
this ticket has been moved to: https://github.com/Xpra-org/xpra/issues/2346