I get these on the AI Agent sometimes currently so im wondering if this is intended. Basically I have it set to timeout after 30s and my intended behavior would be that it retries but this isn’t happening
Information on your n8n setup
n8n version: 1.73.1
Database (default: SQLite): Postgres
Running n8n via (Docker, npm, n8n cloud, desktop app): Docker
Update: This apparently applies to any error.
opened 04:02PM - 10 Sep 24 UTC
closed 02:33PM - 11 Sep 24 UTC
in linear
### Bug Description

The two options, 'Continue' or 'Continue (using error output),' always cause the node to return an error, even if there is a successful retry.
### To Reproduce
1. Go to tab Settings of any node that support retry
2. Turn on "Retry On Fail"
3. Select "Continue" or "Continue (using error output)" for "On Error"
Main-flow
[_DEV__Check_error.json](https://github.com/user-attachments/files/16947485/_DEV__Check_error.json)
Sub-flow to be trigger by Main-flow
[_DEV__Random_error.json](https://github.com/user-attachments/files/16947483/_DEV__Random_error.json)
### Expected behavior
The expected behavior should be the output of last success retry of the current node or the sub-flow trigger by the node (execute sub-flow).
### Operating System
Debian 11
### n8n Version
1.57.0
### Node.js Version
20.17.0
### Database
PostgreSQL
### Execution mode
queue