Operational Runbook

    This page provides the Enterprise-focused view of operational triage, including first-response checklists, router counters, and escalation paths for Enterprise deployments.

    Enterprise First-Pass Checklist

    1. Verify the alert is firing in the expected region and environment
    2. Identify which agent and deployment are involved
    3. Confirm the deployment's effective tuning matches expectations
    4. Check agent heartbeat status and last-known state
    5. Determine whether the issue is runtime (execution), governance (approval/policy), or infrastructure (host/transport)

    Enterprise-Specific Escalation

    SymptomCheckAction
    Deployment stuck in pendingApproval coverageVerify approval credential matches spec hash and environment
    Agent not reportingHeartbeat and connectivityCheck agent enrollment, API key, and network path to control plane
    Effective tuning degradeddoctor outputRun doctor on the target host, compare requested vs applied
    Incident bundle export failingArtifact uploadCheck artifact storage reachability and agent permissions
    Replay rejectedApproval and postureConfirm environment posture, spec hash resolution, and approval scope

    Canonical Published Surfaces

    Provenance
    Need the canonical source?
    Use the public hub to orient yourself, then jump to repo-owned docs or rustdoc when you need contract-level detail.
    Source-of-truth modelRepo docs index