Resiliency

In places where the services access external resources (e.g. RabbitMQ or SQL Server), I have added retry logic to cope with stuff not being up & running when accessing it. When the solution starts, you will therefore probably see a couple of Unable to connect to RabbitMQ/SQL Server, retrying in 5 sec. messages in there. This is expected and not a problem. This is Polly doing its work to make sure that failures that occur when calling a component that is still starting up are handled gracefully.

Sometimes though when you run the application for the first time, starting and initializing SQL Server takes a lot longer. So you could run into the situation that starting and initializing the SQL Server takes longer than the configured timeout and the services are unable to access the database. You will see errors in the logging and the application will not run correctly. In that case, just stop the containers restart them.