Fix OpenClaw Cron Jobs Failing Silently in 2026.3.8

What's Happening

Users upgrading to OpenClaw 2026.3.8 have reported cron jobs that completely stop executing. You'll see {ok: true, enqueued: true} when you manually run openclaw cron run <job-id>, but the jobs never actually start. No sessions are created, and there are no logs or errors indicating what went wrong. The gateway logs might show "cron: started" but nothing further.

The Fix

The most common cause appears to be that the jobs are getting disabled in your jobs.json configuration file. Run this command for each job that's failing:

openclaw cron enable <job-id>

Replace <job-id> with the actual ID of your cron job. You can find job IDs by running openclaw cron status.

Alternatively, you can manually edit ~/.openclaw/cron/jobs.json. Find the job in question and ensure its enabled property is set to true:

{
  "id": "your-job-id",
  "name": "Your Job Name",
  "enabled": true, 
  "schedule": "* * * * *",
  "command": "..."
}

After enabling, test by running openclaw cron run <job-id> again. Jobs should now execute correctly.

Why This Occurs

This behavior seems to be a regression introduced in version 2026.3.8. While the exact root cause in the scheduler hasn't been fully detailed, the symptom is that the enabled flag in the jobs.json file is being reset or incorrectly interpreted. Even though the API calls report success, the scheduler isn't picking up jobs that are marked as disabled internally.

Prevention

Always test your cron job functionality after upgrading OpenClaw. Keep an eye on openclaw cron status and ~/.openclaw/cron/jobs.json to ensure jobs remain enabled. Consider adding a post-upgrade check script that verifies cron job execution for critical workflows.

If you encounter this issue, double-check the enabled status in jobs.json before assuming a deeper scheduler problem.

Last Updated: March 2026