hoangsonww

Advanced Deployment & DevOps Infrastructure

Name: Advanced Deployment & DevOps Infrastructure
Author: hoangsonww

DevOps community intermediate

SOURCE https://github.com/hoangsonww/Agentic-AI-Pipeline

Description

This document outlines the advanced deployment and DevOps infrastructure implemented for the Research Outreach Agentic AI project. It details the new deployment strategies, automation scripts, GitOps

Installation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open the source below and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

Repository README

This is the README for hoangsonww/Agentic-AI-Pipeline, shared by 3 entries in this directory. It describes the repository, not this entry specifically.

Advanced Deployment & DevOps Infrastructure

This document outlines the advanced deployment and DevOps infrastructure implemented for the Research Outreach Agentic AI project. It details the new deployment strategies, automation scripts, GitOps configurations, and monitoring setups added to enhance the existing Kubernetes and AWS ECS environments.

Deployment Strategies

🔵🟢 Blue/Green Deployments

**Kubernetes:**

Dual deployment configurations (blue/green)
Service-based traffic switching
Persistent volume claims for shared state
Zero-downtime deployment script

**AWS ECS:**

Terraform module with CodeDeploy integration
Dual ALB target groups
Automated deployment script
Test listener for pre-production validation

**Location:**

k8s/blue-green/ - Kubernetes manifests
hashicorp/terraform/modules/ecs_blue_green/ - ECS Terraform module
scripts/blue_green_deploy.sh - K8s automation
scripts/ecs_blue_green_deploy.sh - ECS automation

🕯️ Canary Deployments

**Native Kubernetes:**

Stable and canary deployments
NGINX Ingress-based traffic splitting
Progressive rollout automation
Manual control script

**Flagger Integration:**

Automated progressive delivery
Prometheus metrics analysis
Load testing integration
Auto-rollback on failures

**AWS ECS:**

Terraform module with weighted target groups
Separate stable/canary services
Configurable traffic distribution

**Location:**

k8s/canary/ - Kubernetes manifests
hashicorp/terraform/modules/ecs_canary/ - ECS Terraform module
scripts/canary_deploy.sh - Canary automation
k8s/canary/flagger-canary.yaml - Flagger config
k8s/canary/hpa.yaml - Horizontal Pod Autoscaler

📊 GitOps Configurations

**ArgoCD:**

Application manifests for standard, blue/green, and canary
AppProject with RBAC and sync windows
Argo Rollouts progressive delivery
Analysis templates for metrics-based promotion

**Flux v2:**

GitRepository sources
Kustomization configs for all strategies
Image automation and scanning
Slack notifications
Multi-environment support

**Location:**

gitops/argocd/ - All ArgoCD configurations
gitops/flux/ - All Flux configurations

🛠️ Deployment Automation Scripts

**Blue/Green Deployment** (`scripts/blue_green_deploy.sh`)
- Automated image updates
- Health check validation
- Interactive traffic switching
- Rollback instructions
**Canary Deployment** (`scripts/canary_deploy.sh`)
- Manual progressive rollout
- Flagger automated mode
- Smoke testing
- Metrics monitoring prompts
**ECS Blue/Green** (`scripts/ecs_blue_green_deploy.sh`)
- Task definition updates
- CodeDeploy integration
- Deployment monitoring
- Status reporting
**Deployment Monitor** (`scripts/deployment_monitor.sh`)
- Continuous health checks
- Pod, service, and log monitoring
- Resource usage tracking
- Auto-rollback capability
**Quick Rollback** (`scripts/rollback.sh`)