Intermittent workflow execution failed under Load

Questions
1

when execution concurrency increases, I start seeing random workflow failures with no deterministic node pattern. I’m using queue mode with Redis. Could this be worker memory pressure or stalled jobs due to event loop blocking?

Describe the problem/error/question

What is the error message (if any)?

Please share your workflow

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

This strongly suggests resource saturation at the worker level.

In queue mode:

  • Each execution is processed by a worker process.
  • If memory limits are too low, Node.js will trigger garbage collection aggressively, causing execution instability.

Recommended actions:

  • Increase worker memory limits.

  • Scale horizontally by adding more workers.

  • Set EXECUTIONS_PROCESS=main only for debugging, not for high concurrency.

  • Monitor Redi for stalled jobs, latency there can also delay job acknowledgement.

You should also inspect long-running nodes that block the event loop (large JSON transformations, heavy Code node logic).

2 Likes
3

Will work on this, gracias muchos

4

Yeah this is almost certainly stalled jobs from event loop blocking, what’s your per-worker concurrency set to? The default of 10 is way too high under load, drop it to like 3-5 and add more workers instead. Also make sure you’re on Postgres and not SQLite because SQLite will silently corrupt under concurrent writes in queue mode.

1 Like
5

well said , this will work

6

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.