Execution marked as error after running successfully

Questions
, ,
1

Describe the problem/error/question

Hello :slight_smile:

We have some executions that run for a long time and get marked as Error after a while. When looking at the execution, all the nodes ran successfully and the execution completed. We are running a self-hosted instance of n8n on kubernetes in queue mode. We use Redis Cluster for the queue.

We also don’t see any error logs from the main or worker instances. The worker logs show that the job for the execution completed successfully. In the execution_entity table the execution is marked with a status of crashed. The main and worker pods have enough memory allocated to them (No OOM errors). Below are the logs we see in the main and worker instances:

n8n-worker - Worker finished execution 20382 (job 1016)
n8n-main - Execution 20382 (job 1016) finished successfully
n8n-main - Enqueued execution 20382 (job 1016)
n8n-worker - Worker started execution 20382 (job 1016)

We are unsure of how to investigate this further. Any help is appreciated!

What is the error message (if any)?

We don’t see any error message.

Please share your workflow

It happens on several different workflows and they do run successfully, just are marked as in error.

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Information on your n8n setup

  • n8n version: 1.116.2
  • Database (default: SQLite): PostgreSQL
  • n8n EXECUTIONS_PROCESS setting (default: own, main): default
  • Running n8n via (Docker, npm, n8n cloud, desktop app): Docker
  • Operating system: Ubuntu
2

Am I correct in understanding that when you open a run in the Executions Overview, you see “Success” everywhere in the log tab?

Do you see very long runtimes for a workflow step despite Success?

3

Hi Tombo!

Thanks for taking a look. Do we see “Success” everywhere in the log tab? => Unfortunately not, we expect to see a success status but in the executions overview the execution shows Error in Xh Xm. Most of our executions run in a couple of seconds but the ones where we see this issue can be running for a few hours. When we look at the execution we can see all the nodes are green. In the UI it is marked as in Error and in the DB it is marked as crashed.

Do you see very long runtimes for a workflow step despite Success? => I checked the time it took to run each workflow step and I did not see any long runtimes. We also have subworflows and we initially thought the issue was coming from there but their runtimes are within our expected runtimes.

Are there other things we should check? Thanks for your help!

4

I’m facing the same issue. Some executions, after running all nodes successfully, aren’t getting marked as finished and just end up marked as crashed when we restart n8n.

Running self hosted queue mode with 2 workers.

@scyb did you find a fix for this?