This document details the operational mechanics, disaster recovery protocols, and deployment primitives for the highly available infrastructure provisioned for the production environment. The architecture leverages a containerized microservices approach orchestrated via Kubernetes.

All ingress traffic is routed through a resilient load balancing tier with automated TLS termination and strict WAF policies applied.

Environment bootstrapping is entirely codified. Operators should not execute manual shell commands for cluster creation.

To prevent supply chain attacks, all container images deployed must match their cryptographic SHA-256 manifests declared in the Git repository.

Sensitive configuration is injected dynamically at runtime via HashiCorp Vault. No `.env` files contain plain-text credentials in the repository.

Microservices implement strict `/health` (Liveness) and `/ready` (Readiness) endpoints. The Kubernetes control plane utilizes these to govern traffic routing and perform automated pod cycling upon deadlock detection.

For bare-metal backing services (e.g., node-level logging agents), `systemd` is configured for auto-recovery on node reboot with exponential backoff.

In the event of complete region failure, follow Runbook DR-001. State backups are continuously replicated to cold storage in an isolated AWS account.

Refer to the standard telemetry dashboards in Grafana. If Alert ID ERR_OOM_550 triggers, verify the JVM heap limits configured in the Helm values overlay.