As a VRO developer, I want my other containers to stay alive when rabbitmq goes down so that I can have a resilient system.\
Description
When the system is up but dormant for a few hours, killing the rabbitmq-service-1 will instantly kill the svc-pdf-generator-1, svc-assessor-dc7101-1 and svc-assessor-dc6602 containers. Restarting rabbitmq or app-1 will not bring them back up- they must all be manually started at this point. Operation appears normal thereafter.
Log output from these containers at this point are the same:
2023-01-18 16:10:59 2023-01-19 00:10:59 INFO Aborting transport connection: state=1; <socket.socket fd=6, family=AddressFamily.AF_INET, type=SocketKind.SOCK_STREAM, proto=6, laddr=('192.168.32.7', 51368)>
2023-01-18 16:10:59 2023-01-19 00:10:59 INFO _AsyncTransportBase._initate_abort(): Initiating abrupt asynchronous transport shutdown: state=1; error=None; <socket.socket fd=6, family=AddressFamily.AF_INET, type=SocketKind.SOCK_STREAM, proto=6, laddr=('192.168.32.7', 51368)>
2023-01-18 16:10:59 2023-01-19 00:10:59 INFO Deactivating transport: state=1; <socket.socket fd=6, family=AddressFamily.AF_INET, type=SocketKind.SOCK_STREAM, proto=6, laddr=('192.168.32.7', 51368)>
2023-01-18 16:10:59 2023-01-19 00:10:59 INFO AMQP stack terminated, failed to connect, or aborted: opened=True, error-arg=None; pending-error=ConnectionClosedByBroker: (320) "CONNECTION_FORCED - broker forced connection closure with reason 'shutdown'"
2023-01-18 16:10:59 2023-01-19 00:10:59 INFO Stack terminated due to ConnectionClosedByBroker: (320) "CONNECTION_FORCED - broker forced connection closure with reason 'shutdown'"
2023-01-18 16:10:59 2023-01-19 00:10:59 INFO Closing transport socket and unlinking: state=3; <socket.socket fd=6, family=AddressFamily.AF_INET, type=SocketKind.SOCK_STREAM, proto=6, laddr=('192.168.32.7', 51368)>
2023-01-18 16:10:59 2023-01-19 00:10:59 ERROR Unexpected connection close detected: ConnectionClosedByBroker: (320) "CONNECTION_FORCED - broker forced connection closure with reason 'shutdown'"
2023-01-18 16:10:59 2023-01-19 00:10:59 ERROR BlockingConnection.close(200, 'Normal shutdown') called on closed connection.
2023-01-18 16:10:59 Exception ignored in: <function RabbitMQConsumer.del at 0x7f853665ef80>
2023-01-18 16:10:59 Traceback (most recent call last):
2023-01-18 16:10:59 File "/home/docker/main_consumer.py", line 32, in del
2023-01-18 16:10:59 self.connection.close()
2023-01-18 16:10:59 File "/usr/local/lib/python3.10/site-packages/pika/adapters/blocking_connection.py", line 800, in close
2023-01-18 16:10:59 raise exceptions.ConnectionWrongStateError(msg)
2023-01-18 16:10:59 pika.exceptions.ConnectionWrongStateError: BlockingConnection.close(200, 'Normal shutdown') called on closed connection.
Steps to Reproduce
With all containers up and running in a healthy state, walk away from your computer for a few hours. Go on, you've earned it.
Okay that's enough, now get back to work and kill the rabbitmq-service-1 container from Docker Desktop
Observe that svc-pdf-generator, svc-assessor-dc7101-1 and svc-assessor-dc6602-1 all go down with it instantly
Acceptance Criteria
Bringing down rabbitmq-service-1 after a few dormant hours should not bring down any other containers
User Story
As a VRO developer, I want my other containers to stay alive when rabbitmq goes down so that I can have a resilient system.\
Description
When the system is up but dormant for a few hours, killing the rabbitmq-service-1 will instantly kill the svc-pdf-generator-1, svc-assessor-dc7101-1 and svc-assessor-dc6602 containers. Restarting rabbitmq or app-1 will not bring them back up- they must all be manually started at this point. Operation appears normal thereafter.
Log output from these containers at this point are the same:
Steps to Reproduce
Acceptance Criteria
Related to #527