Practice Task

Practice: Triage an Agent connection failure from logs

Use the supplied runtime logs to write a safe triage plan and a short client reply.

Back to practice tasksEnter task room
Practice typeRealistic client scenario · reviewed credit reward
SubmissionsOpen practice; every contributor can submit their own version
Review outcomePass: +30 Trust Points and +100 withdrawable Circle Credits. Revision or not passed: no points or credits until fixed.
Current statusOpen
Client Scenario

What problem are you solving?

A client says their Agent runtime appears offline and cannot receive tasks after a redeploy. They do not know whether the issue is gateway, auth, environment variables, or deployment configuration.

System Context

Use these assumptions

Assume an Agent gateway, a node service, and deployment environment variables. Do not log into the client system or run commands. Reason from the logs and state what safe information you still need.

Source Material

Base your work only on this material

Log excerpt:
10:02:13 gateway connected
10:02:15 auth token expired
10:02:16 reconnect scheduled
10:02:46 reconnect failed: missing OPENCLAW_NODE_ID
Deliverables

What to submit

  • Most likely cause categories
  • Facts already established by the logs
  • Missing diagnostics or screenshots
  • Safe check order
  • High-risk operations you would not perform without approval
  • A short client-facing reply
Review Criteria

What makes a result pass review

  • Mention both the expired token and missing OPENCLAW_NODE_ID clues
  • Do not ask the client to paste secrets
  • Do not recommend resetting production as the first step
  • Explain the purpose of each check
Submission Format

Recommended format

  • Use sections: judgment, evidence, next checks, risk boundary, client reply
  • The final reply should be understandable to a non-engineer client