Operations

Upgrading - Pull latest sources or images. - Rebuild and restart: docker compose up -d --build - Apply database migrations: docker compose exec api alembic upgrade head

Backups - Database: dump PostgreSQL regularly (e.g., pg_dump inside the DB container). - Elasticsearch: use the snapshot API to back up to object storage or a mounted path.

Monitoring & Health - API Health: GET /api/portal/health/ (requires global admin). Returns a grouped status report with components.infrastructure (DB / Redis / Elasticsearch / LLM gateway connectivity) and components.workload (queue-age checks: transactional mail, BAS phishing mail, PII classification, attachment sandbox scans). Top-level status rolls up to ok, degraded, or critical. Always returns HTTP 200 - consumers parse the body's status field. For simple HTTP-status-driven uptime checks, target POST /api/plugin/health instead (returns 503 on infrastructure outages). - Logs: docker compose logs -f api and proxy logs.

Scaling - For larger tenants, allocate more RAM/CPU to Elasticsearch and the API containers.

Security Maintenance - Keep Docker base images and host OS updated. - Rotate secrets and certificates.