openclaw - 💡(How to fix) Fix agentTurn cron jobs silently fail with FailoverError: LLM request timed out on cron lane [1 comments, 2 participants]

lane wait exceeded: lane=cron waitedMs=119951 queueAhead=0 lane task error: lane=cron durationMs=270 error="FailoverError: LLM request timed out." lane task error: lane=session:agent:main:cron:<job-id> durationMs=120226 error="FailoverError: LLM request timed out."

Fix Plan

To resolve the issue with cron jobs failing silently, we need to modify the cron job configuration and the gateway settings. Here are the steps:

• Update the cron job configuration to use sessionTarget: "main" and payload.kind: "systemEvent" to route the reminder into the main session instead of spawning an isolated one.
• Increase the timeout for LLM requests in the gateway settings to prevent timeouts.

Example code snippet to update the cron job configuration:

const cronJob = {
  sessionTarget: 'main',
  payload: {
    kind: 'systemEvent'
  }
};

Verification

To verify that the fix worked, follow these steps:

• Create a new cron job with the updated configuration.
• Trigger the cron job manually with cron run.
• Check the cron runs API to see if the job was executed successfully.
• Check the gateway log to see if there are any errors related to LLM request timeouts.

Extra Tips

• Make sure to update the OpenClaw version to the latest version to ensure you have the latest bug fixes and features.
• Consider implementing retry logic for LLM requests to handle temporary failures.
• Monitor the gateway log for any errors related to LLM request timeouts and adjust the timeout settings as needed.