Pular para o conteúdo principal

CipherVault 4.7 — Frontend coverage completo + observability + runbooks

· 3 min para ler
Rafael Martinez
Rafael Martinez
CEO & Co-founder, CipherVault

38 PRs em 1 dia. Frontend cobre 100% das features v4.5/v4.6, Grafana dashboard ciphervault-overview com 12 panels, 9 Prometheus alerts, 4 operator runbooks novos.

Teste suite estabilizada: 1261/1261 passing, 87 suites verdes.

✨ Frontend UI coverage (PRs #276-#285)​

10 páginas/widgets novas:

PáginaPRO que entrega
/Tenants#276Suspend/unsuspend actions + purge countdown + reason inline
/Share#277One-Time Share list + create modal + copy URL + revoke
/Environments#278Tree hierarchy + chain visualization + effective_secrets com source highlighting + bind/unbind
Settings → OIDC tab#279Discovery preview + JWKS keys table + rotate + AWS IAM setup walkthrough
Dashboard widgets#2834 StatCards (Shares + Envs + KMIP + BreachWatch) com fail-soft queries
Profile → GDPR section#280Export download + typed-confirm forget gate
Tenants quota gauges#280Progress bar color-coded por threshold
Settings → Infrastructure → HSM#281Multi-provider grid + 4 cache metric cards + /health/hsm/cache
/Kmip#282Status header + lifecycle distribution + ops support + objects table + detail modal
AuditLogs filter#28414 novos action types com emoji + cores por severity
/EndpointDiscovery#285Backend ingestion + admin list + 4 stat cards + filterable table

✨ Observability — Grafana + Prometheus (#287)​

Dashboard ciphervault-overview com 12 panels:

  • Leases gauges (active/expiring/expired por engine)
  • Latency p50/p95/p99 por endpoint
  • Multi-KM health (Local + AWS KMS + PKCS#11)
  • Cache hit rate (HSM cache, JWKS cache, Lease cache)
  • Replication lag (CDC checkpoint Postgres → MongoDB/MySQL)
  • Phase 5 multi-region replica state
  • Build info (version, commit, env)

Prometheus alerts — 9 rules:

SeveridadeCount
Critical2 (DB down, master_key_rotate failed)
Warning5 (HSM cache miss spike, lease pool > 80%, replication lag > 5min, etc.)
Info2 (cert renew em 30d, baseline anomaly low confidence)

Auto-provisioning configs em docker-compose.observability.yml:

docker compose -f docker-compose.yml \
               -f infrastructure/docker-compose/docker-compose.observability.yml up -d
# Grafana em :3001 com dashboard pré-importado, Prometheus :9090

✨ Operator runbooks (#288)​

4 novos:

  • OIDC_AWS_SETUP.md — walkthrough OIDC provider → IAM trust → dynamic backend
  • RLS_ACTIVATION.md — 4-phase Postgres Row-Level Security rollout (sombra → enforced → rollback drill → cleanup)
  • KMIP_CLIENT_TESTING.md — pykmip examples + OASIS conformance checklist
  • DR_PLAYBOOK.md — 7 recovery scenarios (region failover, KMS rotation rollback, HSM bricked, full restore from snapshot, partial corruption, network partition, master_key compromise)

🐛 Test stability (#286)​

AD/LDAP engines test isolation via lazy require de ldapjs. Eliminou 21 test failures intermitentes. Suíte agora rola 1261/1261 passing, 87 suites verdes — sem flakies.

Comparison pages (#292)​

Públicas no docs site, vs Doppler/Vault/Akeyless. Honestas sobre onde CipherVault perde (ex: maturidade ecosystem vs Vault) e onde ganha (LGPD-native, brazilian timezone support, dual-license).

DR Test Report (#291)​

Execution report do DR drill 2026-Q2 publicado em docs/dr-test-2026-q2.md. Validado RTO 12min / RPO 38s real (target: RTO 15min / RPO < 60s).

Atualizando​

helm upgrade ciphervault ciphervault/ciphervault --version 4.7.0 \
  --reuse-values

# Importar dashboard Grafana
kubectl apply -f infrastructure/observability/grafana-dashboard-cv-overview.yaml

# Importar Prometheus rules
kubectl apply -f infrastructure/observability/prometheus-alerts.yaml

— Rafael Martinez, CEO