pal-e-docs

Review: Alerting rules for CrashLoopBackOff (basketball-api#57)

review-70-2026-03-27 Review

review needs-refinement
Project

Verdict: NEEDS_REFINEMENT

Template Completeness

  • [ ] Type — MISSING. Should be ### Type\nFeature.
  • [x] Lineage — present
  • [x] Repo — forgejo_admin/basketball-api (BUT WRONG — see Repo Placement)
  • [x] User Story — present
  • [x] Context — present, strong motivation
  • [x] File Targets — present but vague
  • [x] Acceptance Criteria — 3 items
  • [x] Test Expectations — present
  • [x] Constraints — present
  • [x] Checklist — present
  • [x] Related — present

Traceability

  • [ ] story: label — missing.
  • [ ] arch: label — missing. Should be arch:monitoring.
  • [x] Forgejo issue — basketball-api#57, open
  • [ ] Board item has no labels.

File Targets

  • [ ] "k8s alerting rules" — VAGUE. Actual location: pal-e-platform/terraform/modules/monitoring/main.tf (PrometheusRule CRDs at lines 387+ and 684+).
  • [ ] "notification channel" — STALE. Telegram already configured via Alertmanager in pal-e-platform.

Repo Placement

WRONG REPO. Issue filed on basketball-api but all work is in pal-e-platform. PrometheusRule CRDs and Alertmanager config live in pal-e-platform/terraform/modules/monitoring/.

Dependencies

  • kube-prometheus-stack includes KubePodCrashLooping by default — may already be active. Need to verify before creating duplicate rules.
  • Existing pattern: blackbox_alerts and embedding_alerts PrometheusRules in monitoring module.

Acceptance Criteria

  • "Alert fires within 5 minutes" — may already exist via built-in rules. Verify first.
  • "/healthz non-200 for >2 minutes" — may overlap with existing EndpointDown blackbox alert.
  • "Notification reaches Lucas (channel TBD)" — channel is NOT TBD, Telegram is already configured.

Blast Radius

  • CrashLoopBackOff rule would be platform-wide, not basketball-api-specific. Scope should acknowledge this.
  • Potential duplicate alerts if built-in rules are already active.

Recommendation

  • CRITICAL: Wrong repo. Re-file on pal-e-platform or update repo field.
  • Add ### Type\nFeature.
  • Replace vague file targets with terraform/modules/monitoring/main.tf.
  • Check if KubePodCrashLooping built-in alert is already active and routing.
  • Add board labels: type:feature, arch:monitoring, repo:pal-e-platform.
  • Fix "channel TBD" — Telegram is already configured.