Validate orchestrator restart by inspecting logs and confirming job reconciliation.
npx skills add https://github.com/tinkermonkey/switchyard --skill check-restart
Please help me install this Agent Skill. Skill: check-restart Source: https://github.com/tinkermonkey/switchyard/tree/main/.claude/skills/check-restart Command: npx skills add https://github.com/tinkermonkey/switchyard --skill check-restart
This skill helps ensure that a restart of the orchestrator completes cleanly by verifying logs, reconciliation, and the status of active jobs.
Run a restart verification to confirm the orchestrator is healthy and all active jobs have resumed.
High-intent search queries and answers about installing and using this skill.
To verify a clean orchestrator restart, you need to inspect logs for errors, confirm reconciliation completes, and check that all active in-flight jobs resume correctly without being left in limbo.
Orchestrator reconciliation is the process of validating that in-flight tasks and job states persist correctly across a restart. You need to check it during post-deploy validation, incident recovery, or routine health verifications.
Checking job status after a restart involves scanning orchestrator logs for errors, verifying reconciliation checkpoints, and enforcing that active jobs meet defined success criteria to ensure tasks resumed properly.
Yes, you can run a health-check to validate pipeline recovery during maintenance windows by confirming job state persistence, inspecting logs for errors, and ensuring all active jobs have resumed correctly.
Tasks are left in limbo after a restart when reconciliation fails or job state persistence does not meet defined success criteria, preventing in-flight tasks from resuming correctly across the pipeline.
Verifying orchestrator restarts via log inspection is limited to confirming the absence of errors and validating reconciliation status; it cannot fix underlying job state persistence failures or missing checkpoints.
TienchingIterate tasks persistently with retries, parallel execution, and architect verification.
qa-amanCreate operational runbooks with health checks, escalation paths, and recovery steps.
aaronbrethorstMonitor Render services health, metrics, logs, and resource usage.
maedmattOrchestrate Railway deployments, logs, status checks, and environment variables via CLI.
