Every 5 seconds or so, another 10 or 20 canceled executions appear in the execution history on a workflow. I’ve turned off the workflow, restarted the container… as soon as I turn the workflow back on… they keep appearing.
Weird stuff. We do need more information on how you setup n8n to have a guess of what is going on.
Assuming this is selfhosted.
It’s a Docker self hosted? Docker logs?
The same is happening for me. I have updated to version 1.84.3. The installation is the queue version with Redis and Postgres.
Hi, I have been having this issue for multiple months (at least october, and I update n8n every 2 weeks or so. I self host n8n using Docker on debian 12 up to date, in queue mode with Redis, with postgres (version 16 and 17).
Still happens today with 1.86.0, a workflow got stuck in this “canceled” loop where
I get dozens of canceled executions per second and the workflow never executes despite having a schedule trigger.
Manual run still works, disable/re-enable trigger won’t do anything. Only workaround I found is to re-create the workflow.
My guess is that it happens when restarting docker compose, or changing configuration even slightly, with workflow that have very frequent triggers (here every 3 minutes). I remeber it happening almost every time with a RSS trigger on Raindrop.
I just copy everything, create a new workflow, paste, delete old workflow.
1 Like
Sure it works, but the error can still happen again
Oh, absolutely. But I guess I was saying “that’s all I can work out to ‘solve’ the problem when it happens.”
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.