apiVersion: "helm-expt.confighub.com/v1alpha1"
kind: "GitOpsRuntimeReview"
metadata:
  name: "prometheus-default-alertmanager-statefulset-sync-watch"
  chart: "prometheus-community/prometheus"
  version: "29.9.0"
spec:
  base: "default"
  controller: "Argo CD"
  targetShape: "kind proof rig with ConfigHub OCI source and local-path storage"
  receipt: "runs/live-helm-confighub-compare/prometheus-community-prometheus-default-29-9-0/receipt.yaml"
  observedResult: "watch"
  currentSummary:
    regularHelmRuntime: "pass"
    configHubDirectRuntime: "pass"
    configHubOciSync: "OutOfSync"
    configHubOciHealth: "Healthy"
    semanticParity: "pass"
    workloadRuntime: "pass"
  passed:
    - "regular Helm reached readiness"
    - "ConfigHub direct apply reached readiness"
    - "ConfigHub OCI/Argo workloads reached readiness"
    - "Prometheus server, Alertmanager, kube-state-metrics, node-exporter, and pushgateway became ready"
    - "semantic object parity passed for direct apply and OCI/Argo"
  watch:
    summary: "Argo stayed OutOfSync/Healthy even though all workloads reached readiness."
    observedResidue:
      - "StatefulSet/monitoring/prometheus-alertmanager: OutOfSync"
      - "Argo repeatedly server-side-applied prometheus-alertmanager and reported each sync attempt as succeeded"
    interpretation: "This is a GitOps StatefulSet drift/canonicalization watch row, not a render parity or runtime readiness defect."
  diagnosis:
    summary: "The default base includes Alertmanager state and persistent storage. The server-only-ephemeral base has passing live OCI evidence and is the simpler try-now path; this default base needs an explicit decision about Alertmanager StatefulSet drift under Argo server-side apply."
    productBoundary: "A production claim for this base needs target-scoped storage, retention, Alertmanager routing, backup, and StatefulSet drift decisions."
    nextInvestigation:
      - "Compare the live Alertmanager StatefulSet against the OCI desired object."
      - "Decide whether the k8s-zero-defaults canonicalization profile used for the older Prometheus default receipt applies to 29.9.0."
      - "Keep storage, retention, Alertmanager, and backup policy out of the production claim until they have target-scoped decisions."
  relatedPassingBase:
    base: "server-only-ephemeral"
    receipt: "runs/live-helm-confighub-compare/prometheus-community-prometheus-server-only-ephemeral-29-9-0/receipt.yaml"
  evidence:
    argoCoreChildJson: "runs/live-helm-confighub-compare/prometheus-community-prometheus-default-29-9-0/argocd-core-child.json"
    argoCoreChildTree: "runs/live-helm-confighub-compare/prometheus-community-prometheus-default-29-9-0/argocd-core-child-tree.txt"
    argoCoreRootJson: "runs/live-helm-confighub-compare/prometheus-community-prometheus-default-29-9-0/argocd-core-root.json"
    argoCoreRootTree: "runs/live-helm-confighub-compare/prometheus-community-prometheus-default-29-9-0/argocd-core-root-tree.txt"
  notClaimed:
    - "This row does not prove production Prometheus storage, retention, Alertmanager routing, backup, or HA."
    - "This row does not claim Argo OutOfSync is harmless in production."