8 months ago
I am trying to run multiple workers for my n8n instance and its deploying them all on the same port even though i specify different ports in env variables, causing crashes. I tried different templates and still its an issue.
1 Replies
7 months ago
Since n8n v1.2, the worker binary is no longer a web-server.
It just launches a task-runner and dials back to the “primary” n8n instance.
Internally it still listens on 5679 for its own broker/health-check, so it ignores the usual PORT value you inject.
Every n8n worker always binds to port 5679 inside its own container, that’s by design and causes no conflict as long as each worker runs in a separate container/VM.
If you must use a different port, set it on the primary with:N8N_RUNNERS_BROKER_PORT=<new-port>
and point each worker at it with:N8N_RUNNERS_TASK_BROKER_URI=http://<primary-host>:<new-port>