Describe the problem/error/question
Guys, I’m building a system that works with multiple workflows. I’ve separated their responsibilities. Basically, there’s one that receives messages, another that orchestrates, and another that sends the response. Between the second and third, there can be others. Currently, there are 3 more. The orchestrator sends to one of these sub-workflows, and after processing, this one sends it to the one that responds.
It looks something like this:
- Receives Messages
- Orchestrator:
- Specific Action 1;
- Specific Action 2;
- Specific Action 3;
- n actions;
- Sends Response.
The path goes through 1 and 2, and then can go to a specific action and then to 3.
When an error occurs in a sub-workflow of a specific action, both that flow and 1 and 2 also throw an error.
So I have two questions:
-
How to separate these execution dependencies? I thought that when I executed step 1 and then moved on to step 2, step 1 would finish its work, but it seems they are still connected.
-
How do I inform users that there was an error and that they should try again?
I’ll put an example of what the configuration looks like here.
I put an error message that occurred in one of the sub-workflows (it has already been adjusted), but it appeared in workflow 1. The error came back from workflow to workflow until it reached the first one.
The “Wait For Sub-Workflow Completion” is set to false.
Please share your workflow
Share the output returned by the last node
No prompt specified
Expected to find the prompt in an input field called ‘guardrailsInput’ (this is what the guardrails node node outputs). To use something else, change the ‘Prompt’ parameter
Information on your n8n setup
- n8n version: [email protected]
- Database (default: SQLite): Supabase
- n8n EXECUTIONS_PROCESS setting (default: own, main): main
- Running n8n via (Docker, npm, n8n cloud, desktop app): n8n cloud
- Operating system: Windows