Hi,
I have a multi-runner deployment on my infrastructure that works well and does what I want. I’ve worked with Python orchestrators like Prefect and Airflow before, and when deploying a worker (runner in n8n parlance), you tag it in several ways so that you’re able to attach it to a group/queue so that it can be targeted specifically when executing a workflow. Is this possible within n8n currently?
I’d like to be able to deploy a runner inside a specific network, to allow it to execute workflows from within it, that are reliant on local conditions/services that aren’t available from the main n8n instance.
This does in effect make n8n more of an orchestrator (which I can do with Prefect), but I’d like to know if it’s possible within the n8n ecosystem because, well, I really like n8n.
Cheers,
1 Like
Hey @notScottSmith !
Maybe if you use multi-main n8j instances (to “de-couple” internal/external) and use directly task-runners with the main instance(but then here is only JS allowed in internal).
I don’t think you can assign directly a specific job to a runner since workers get jobs from Redis, and if they encounter Js/Python Code, the available task-runner gets that job(not aware to do with specific one), if is “external” mode.
In internal mode(no Python, only JS), the main can have it’s own task-runner(no queue mode )…
So it s more a combination on how you can mix this proccess, but exactly what you are looking for, curios as well if there is an “approach”(or who knows, maybe is getting added if there s a feature resquest).
Hope i didn’t understood you wrong…
Cheers!
P.S and welcome back to community! 
1 Like