RabbitMQ default user+pass are not taken in account causing the openCTI not to start up

[vladovarga]

The current docker-compose.yml configuration causes the rabbitMQ not to start up properly and hence, the openCTI container thinks the RabbitMQ seems down (and then the workers dependent on the openCTI API are not started up neither)

The problem is in the RabbitMQ configuration, specifically the RABBITMQ_NODENAME env variable causes the RabbitMQ to start up in a way where in the logs this sentence appears: Will not seed default virtual host and user: have definitions to load...

This means that the:

1. RABBITMQ_DEFAULT_USER and RABBITMQ_DEFAULT_PASS will be ignored and the default configuration is taken in account (guest/guest)
2. openCTI container is not able to log in to the RabbitMQ and this log message appears: {"category":"APP","error":{"context":{"category":"technical","http_status":500,"reason":"RabbitMQ seems down"},"message":"A database error has occurred","name":"DatabaseError","stack":"DatabaseError: A database error has occurred\n at error (/opt/opencti/build/src/config/errors.js:8:10)\n at DatabaseError (/opt/opencti/build/src/config/errors.js:58:48)\n at /opt/opencti/build/src/database/rabbitmq.js:205:13\n at processTicksAndRejections (node:internal/process/task_queues:95:5)\n at checkSystemDependencies (/opt/opencti/build/src/initialization.js:143:3)\n at platformStart (/opt/opencti/build/src/boot.js:215:5)"},"level":"error","message":"[OPENCTI] Platform start fail","timestamp":"2023-10-30T11:04:26.468Z","version":"5.11.12"}
3. Workers are not able to connect to the openCTI container as the container keeps restarting

Possible solution - commenting out the RABBITMQ_NODENAME env variable. But that's just my hot-fix.