Node dies if connection to caster fails

[tiesus]

On my machine the node dies, raising an exception if connection to caster fails. From the parameter section, i thought the node should do retries.

[tonynajjar]

Any update there? We also have the issue. This is probably the culprit

[pondersome]

For me (on ros2) it does do retries. But it also has a retry limit: reconnect_attempt_max

If you see this line in your output:
Exception: Reconnect was attempted 10 times, but never succeeded

Then you've probably run into the intended functionality. You can change the number of attempts and/or the wait (reconnect_attempt_wait_seconds) to allow servers or the network more time to recover from outages. It doesn't make much sense to retry forever. The exception and node disappearing can signal that something needs to be attended to.

[pondersome]

To be clear, the retry behavior I described above takes place once you have successfully connected to a server at least once. Then it will retry on intermittent communications outages. It is true that if the connection fails on the first attempt, the exception raises immediately. I don't know if that's a feature or a bug.