Docker deployment does not break if output plugin is broken

[D1sD3s]

Describe the bug

The issue I'm encountering involves the Docker deployment. Specifically, when the output plugin experiences errors and fails to function as expected, the Docker deployment does not terminate. This occurs because exceptions are being caught and disregarded by the honeypot. In my perspective, it would be more desirable for the Docker deployment to exit when such issues with the plugin arise, as the plugin's proper operation is crucial.

To Reproduce

Steps to reproduce the behavior:

start the following docker-compose.yml with docker-compose up

--- 
version: '3' 
services: 
  cowrie: 
    image: cowrie/cowrie:3568988b 
    ports: 
      - "22:2222" 
    volumes: 
      - ./cowrie.cfg:/cowrie/cowrie-git/etc/cowrie.cfg 
    restart: always 

Use this section in cowrie.cfg to configure HPFeeds:

# HPFeeds3 
# Python3 implementation of HPFeeds 
[output_hpfeeds3] 
enabled = true 
server =_hpfeeds-broker_ 
port = 10000 
identifier = my_ident 
secret = my_secret 
debug = false 
channel = my_channel 

Expected behavior

In the event of a malfunctioning output plugin in the Docker deployment, the deployment should gracefully exit or stop. This would ensure that any issues with the plugin are addressed and don't compromise the overall functionality of the honeypot.

Additional context

Related issue, this approach would prevent overseeing problems like mentioned here: issue #1969

[D1sD3s]

I added a PR i hope its okay.