N8n Workflows Not Respecting Timeout Settings – Executions Stuck for 100+ Hours

Questions
1

We are experiencing an issue with multiple workflows in n8n where executions are not terminating despite configured timeout settings. Several workflow executions have been running continuously for extended durations (100 to 200+ hours), indicating that the timeout configuration is not being enforced.

We have already configured the workflow timeout setting to 30 minutes; however, this does not appear to have any effect on long-running or stuck executions. As a result, workflows remain in a running state indefinitely, impacting system performance and reliability.

This issue is affecting multiple workflows and is causing resource utilization concerns.

Problem Statement:

  • Workflow executions are not stopping after the configured timeout (30 minutes).

  • Executions remain active for excessively long durations (100–200+ hours).

  • Timeout configuration in n8n settings is not being honored.

  • Potential impact on system performance due to stuck or hanging workflows.

3

@Kishan_Ghetiya the workflow-level timeout only applies to executions that actually reach a checkpoint between nodes — if a single node is hung on an HTTP request or DB query it never gets checked, that’s why your 30min setting does nothing. set EXECUTIONS_TIMEOUT=1800 and EXECUTIONS_TIMEOUT_MAX=3600 as env vars and restart, that enforces it globally at the worker level

4

Hey @achamm Thanks for quick reply! What is the difference between EXECUTIONS_TIMEOUT and EXECUTIONS_TIMEOUT_MAX?

5

EXECUTIONS_TIMEOUT is the default applied to every execution, EXECUTIONS_TIMEOUT_MAX is the upper cap a user can set per-workflow — so even if someone bumps a workflow timeout to 10h in the UI, max forces it back down to 1h. set max equal to or higher than the default, otherwise n8n logs a warning on boot.

6

Got it, thanks