feat(infra-monitoring): v2 statefulsets integration tests

[nikhilmantri0902]

Pull Request

---

📄 Summary

First end-to-end integration tests for POST /api/v2/infra_monitoring/statefulsets.

Screenshots / Screen Recordings (if applicable)

N/A — backend test addition; no UI.

Issues closed by this PR

Part of https://github.com/SigNoz/engineering-pod/issues/4980

---

✅ Change Type

• 🧪 Test-only

---

🐛 Bug Context

N/A — not a bug fix.

---

🧪 Testing Strategy

53 tests across 21 functions in tests/integration/tests/inframonitoring/08_statefulsets.py + 9 JSONL datasets + 1 expected-values JSON (generated by tests/gen_statefulsets_datasets.py).

• test_statefulsets_happy_path — shape, 11 record fields (statefulSetName, 6 float metrics, desiredPods/currentPods as int, podCountsByPhase, meta); meta carries
  k8s.statefulset.name/k8s.namespace.name/k8s.cluster.name
• test_statefulsets_value_accuracy — exact 6 float metric values + replica counts + phase counts. Locks in Sum vs Avg split: cpu/memory = SpaceAggregationSum; cpu_request/limit +
  memory_request/limit = SpaceAggregationAvg; desired/current = latest-summed. v1/v2 parity confirmed by code reading (pkg/query-service/app/inframetrics/statefulsets.go:21-90 ↔
  pkg/modules/inframonitoring/implinframonitoring/statefulsets_constants.go:24-242)
• test_statefulsets_missing_metrics — seed only k8s.pod.cpu.usage, other 8 of 9 required metrics flagged → records=[], total=0
• test_statefulsets_filter_and / in / not_in / contains — each QB v5 operator
• test_statefulsets_filter_combos ×4 — AND+IN, AND+NOT IN, AND+CONTAINS, IN+CONTAINS
• test_statefulsets_filter_bad_attr_name, test_statefulsets_filter_bad_grammar ×2 — 400 paths
• test_statefulsets_pod_phase_aggregation — 4 Running + 1 Pending + 2 Failed
• test_statefulsets_desired_current_counts — desired=5, current=3, only 2 Running pods seeded. Locks in int typing + distinguishes replica counts from phase counts (separate code paths); models a
  scale-up in progress
• test_statefulsets_base_filter_drops_non_statefulset_pods — seed 1 SS pod + 1 standalone pod (no k8s.statefulset.name) + 1 Deployment pod (k8s.deployment.name only); only the real SS row
  appears. Documents statefulSetsBaseFilterExpr (statefulsets_constants.go:10, :63-69)
• test_statefulsets_groupby_namespace — groupBy=[k8s.namespace.name] → 2 records, statefulSetName="", meta surfaces k8s.namespace.name
• test_statefulsets_pagination_sync — K=7, limit=3, offsets {0,3,6}: page invariants, total invariant, full partition
• test_statefulsets_offset_beyond_total — empty records, total unchanged
• test_statefulsets_total_invariant_across_orderby — 8 metric columns × 2 directions = 16 calls, total==K
• test_statefulsets_orderby_correctness ×16 — 8 columns (cpu, cpu_request, cpu_limit, memory, memory_request, memory_limit, desired_pods, current_pods) × 2 directions. Exercises every
  entry in orderByToStatefulSetsQueryNames
• test_statefulsets_orderby_by_statefulset_name ×2 — metadata-name branch (PaginateMetadataByName)
• test_statefulsets_validation_errors ×12 — all Validate() rules including orderBy=k8s.statefulset.name + non-empty groupBy rejection
• test_statefulsets_auth ×2 — 401 no token, 200 admin

Manual verification:

• 53/53 statefulsets green on first run (~53 s)
• Full inframonitoring suite: 362/362 green (39 hosts + 45 pods + 43 nodes + 37 namespaces + 44 clusters + 48 volumes + 53 deployments + 53 statefulsets, ~5 min 50 s)

---

⚠️ Risk & Impact Assessment

• Blast radius: none — test-only, no product code changes
• Potential regressions: none; CI runtime grows by ~53 s for this file
• Rollback plan: revert — no migrations, no shared state

---

📝 Changelog

Field	Value
Deployment Type	N/A
Change Type	N/A
Description	N/A — internal test addition, not user-facing

---

📋 Checklist

• Tests added or explicitly not required
• Manually tested (53/53 green; full inframonitoring suite 362/362)
• Breaking changes documented (none — test-only)
• Backward compatibility considered (N/A — test-only)

---

👀 Notes for Reviewers