Skip to main content

Secret Management

Master plan for secure handling of secrets, credentials, and sensitive configuration. Covers secret storage solutions, rotation strategies, least-privilege access patterns, and integration with CI/CD pipelines and application runtime environments.

Key Topics

  • Secret Storage Solutions: HashiCorp Vault, AWS Secrets Manager, Azure Key Vault, GCP Secret Manager
  • Secret Rotation: Automated rotation strategies, zero-downtime updates, and grace periods
  • Access Control: RBAC, policy-based access, service identity management, and audit logging
  • Encryption at Rest: Key encryption keys (KEK), data encryption keys (DEK), and envelope encryption
  • Encryption in Transit: TLS/mTLS, certificate management, and secure key exchange
  • Secret Injection: Environment variables, mounted volumes, sidecar containers, and init containers
  • Development Workflows: Local development secrets, .env files, and git-secret/git-crypt
  • CI/CD Integration: Pipeline secret handling, ephemeral credentials, and OIDC token exchange
  • Secret Scanning: Pre-commit hooks, GitHub secret scanning, and leaked credential detection
  • Compliance: SOC 2, PCI-DSS, HIPAA requirements for secret handling and audit trails

Primary Tools & Technologies

Secret Management Platforms:

  • HashiCorp Vault (self-hosted, dynamic secrets)
  • AWS Secrets Manager (managed, AWS-native)
  • Azure Key Vault (managed, Azure-native)
  • Google Secret Manager (managed, GCP-native)
  • Doppler, 1Password Secrets Automation (developer-focused)

Kubernetes Secret Management:

  • External Secrets Operator (sync from external sources)
  • Sealed Secrets (encrypted secrets in Git)
  • SOPS (encrypted secrets with age/PGP)
  • Vault Agent Injector (Vault integration)

CI/CD Integration:

  • GitHub Actions secrets, GitLab CI/CD variables
  • OIDC federation (GitHub → AWS, Azure, GCP)
  • Vault agent for ephemeral credentials

Secret Scanning:

  • TruffleHog, GitLeaks (pre-commit hooks)
  • GitHub secret scanning (automated detection)
  • SpectralOps (real-time secret detection)

Development Tools:

  • direnv (environment variable management)
  • git-secret, git-crypt (encrypted secrets in repos)
  • .env files with validation (dotenv, python-decouple)

Integration Points

Upstream Dependencies:

  • Infrastructure as Code: Secrets referenced in Terraform, CloudFormation, Pulumi
  • Container Orchestration: Kubernetes Secret objects and CSI drivers

Downstream Consumers:

  • Data Architecture: Database connection strings and credentials
  • Streaming Data: Kafka broker credentials, schema registry authentication
  • Data Transformation: Pipeline secrets for source/destination connections
  • SQL Optimization: Secure credential passing for query execution

Cross-Functional:

  • Application Security: API keys, OAuth tokens, JWT signing keys
  • Monitoring & Alerting: Secret access logging and anomaly detection
  • Compliance & Audit: Secret access trails and rotation evidence

Status

Master Plan Available - Comprehensive guidance for secure secret management, covering Vault, cloud-native secret stores, rotation strategies, and CI/CD integration.


Part of the Data Engineering skill collection focused on securing sensitive credentials and configuration data.