Try or ByeTry or Bye

n8n Executions Stuck in Queued / Starting Soon

Ongoing n8n Cloud, self-hosted queue mode Updated October 21, 2026

Quick answer

Executions stuck in 'Queued / Starting soon' mean no worker is picking them up. On self-hosted setups, fix mismatched queue-mode settings or restart workers; on n8n Cloud, restart the workspace or contact support.

Most likely fix: Use one execution mode consistently (self-hosted).

Symptoms

Executions list shows runs as 'Queued' or 'Starting soon' for hours, Code nodes spin indefinitely, and the UI may show no active executions.

Why it happens

  • Main instance and workers running in different execution modes
  • Worker crashed or stopped consuming the queue
  • Execution spike on n8n Cloud exceeding concurrency
  • Long-running or waiting executions blocking concurrency slots
  • Main and worker on different n8n versions

How to fix it

Try these in order — the most common fix is first.

  1. 1. Use one execution mode consistently (self-hosted)

    1. Either run a single container in regular mode, or
    2. Set EXECUTIONS_MODE=queue on both the main instance and every worker
    3. Make sure main and workers run the same n8n version

  2. 2. Restart the workers

    1. Restart the worker service(s)
    2. Check worker logs for errors or memory warnings
    3. Confirm Redis is reachable from main and workers

  3. 3. Free up concurrency

    1. Open Executions and stop runs stuck in Running or Waiting
    2. Reduce trigger frequency if executions spike

  4. 4. Restart the n8n Cloud workspace

    1. Open the n8n Cloud admin dashboard as the workspace owner
    2. Restart the workspace
    3. If executions still don't start, contact n8n support with the workspace name

This guide also covers

  • n8n executions stuck queued
  • n8n starting soon
  • n8n workflow stuck running
  • n8n queue mode not working

Still not working?

If n8n keeps failing, you're not alone. See every n8n problem users report or the alternatives people switch to.

Other n8n errors