Redis disconnect yields unexpected ServiceRunner termination

[TilBlechschmidt]

🐛 Bug description

When the Redis connection is interrupted, the ServiceRunner loop just terminates with an "Ok" status code.

🦶 Reproduction steps

Steps to reproduce the behavior:

1. Launch the collector locally and connect to a remote redis through kubectl port-forward
2. Terminate the redis in K8s
3. Watch as "unexpected EOF" errors are thrown
4. See error

🎯 Expected behaviour

It should log the error and either try reconnecting or be restarted by jatsl!

📺 Screenshots

 2021-08-17T13:34:57.877Z DEBUG webgrid::harness::redis > unexpected end of file
 2021-08-17T13:34:57.877Z DEBUG webgrid::harness::redis > unexpected end of file
 2021-08-17T13:34:57.885Z ERROR webgrid::library::communication::implementation::redis::queue_provider > Encountered error reading from redis stream unexpected end of file
 2021-08-17T13:34:57.885Z ERROR webgrid::library::communication::implementation::redis::queue_provider > Encountered error reading from redis stream unexpected end of file
 2021-08-17T13:34:57.893Z INFO  jatsl::scheduler                                                       > Finished         ServiceRunner(SchedulingWatcherService)
 2021-08-17T13:34:57.893Z INFO  jatsl::scheduler                                                       > Finished         ServiceRunner(CreationWatcherService)

Context

Version
On the architecture rework branch head 🙂