# 🚀 Day-7: High Availability with MIGs and Load Balancing (GCP UI)

## 🎯 Goal
Learn how to deploy a production-ready, highly available web application in GCP using **Managed Instance Groups (MIGs)**, **HTTP(S) Load Balancer**, and **health checks**, with observability through **Cloud Logging** — all using the **GCP Console (UI only)**.

---

## 📚 Key Concepts Covered

| Concept                        | Description |
|-------------------------------|-------------|
| **Managed Instance Group**     | Automatically manages identical VMs across zones |
| **Instance Template**          | Blueprint for creating VMs (OS, startup script, tags) |
| **HTTP(S) Load Balancer**      | Global load balancer with traffic distribution & health checks |
| **Health Check**               | Continuously checks backend health to route traffic |
| **Multi-zone Deployment**      | Ensures HA by distributing across zones |
| **Cloud Logging**              | Logs backend traffic and health status |

---

## 🧪 Hands-On Lab (Console UI Only)

### 🛠️ Step 1: Write Your Startup Script

    #!/bin/bash
    apt update
    apt install -y nginx
    echo "Welcome to MIG Demo - $(hostname)" > /var/www/html/index.html

> Save this script as `startup.sh` on your local machine.

---

### 🛠️ Step 2: Create an Instance Template

1. Go to **Compute Engine → Instance templates**
2. Click **Create instance template**
3. Set:
   - Name: `web-template`
   - Machine type: `e2-micro`
   - Boot disk: Debian or Ubuntu
4. Under **Advanced options → Automation → Startup script**:
   - Paste the contents of `startup.sh`
5. Add **network tag**: `http-server`
6. Click **Create**

---

### 🛠️ Step 3: Create a Managed Instance Group (MIG)

1. Go to **Compute Engine → Instance groups**
2. Click **Create instance group**
3. Set:
   - Name: `web-mig`
   - Location: **Multiple zones in `us-central1`**
   - Select zones: `us-central1-a`, `us-central1-b`, `us-central1-c`
   - Instance template: `web-template`
   - Autoscaling: Enable (optional)
   - Number of instances: 3 (1 per zone)
4. Click **Create**

---

### 🛠️ Step 4: Allow HTTP Traffic via Firewall

1. Go to **VPC network → Firewall**
2. Click **Create firewall rule**
3. Set:
   - Name: `allow-http`
   - Direction: Ingress
   - Targets: Tags
   - Target tag: `http-server`
   - Source IP ranges: `0.0.0.0/0`
   - Protocols: Allow TCP port 80
4. Click **Create**

---

### 🛠️ Step 5: Create a Health Check

1. Go to **Network services → Health checks**
2. Click **Create a health check**
3. Set:
   - Name: `web-health-check`
   - Protocol: HTTP
   - Port: 80
4. Click **Create**

---

### 🛠️ Step 6: Create Backend Service and Attach MIG

1. Go to **Network services → Load balancing**
2. Click **Create load balancer → Start configuration**
3. Choose:
   - HTTP(S) Load Balancer (From Internet to my VMs)
4. Set:
   - Name: `web-lb`
   - Backend configuration:
     - Create backend service
     - Add the MIG (`web-mig`)
     - Attach the **health check** (`web-health-check`)
5. Click **Next**

---

### 🛠️ Step 7: Configure URL Map and Frontend

1. In the same wizard:
   - Leave URL map as default
   - Frontend:
     - Protocol: HTTP
     - Port: 80
     - Create a new global IP (name it `web-ip`)
2. Click **Create**

---

### 🛠️ Step 8: Access the Load Balancer

1. Go to **VPC → External IP addresses**
2. Look for the **web-ip** you reserved earlier
3. Open it in a browser → You should see responses like:
    ```
    Welcome to MIG Demo - web-xxxxx
    ```

---

### 🛠️ Step 9: Enable Backend Logging

1. Go to **Network services → Load balancing**
2. Click into `web-lb`
3. Click **Edit backend service**
4. Enable:
   - **Cloud Logging**
   - Set sample rate: `1.0` (100%)
5. Save changes

To view logs:
- Go to **Logging → Logs Explorer**
- Query:
    ```
    resource.type="http_load_balancer"
    ```

---

## 📦 Real-World DevOps Use Case

> You need to deploy a frontend web app behind a highly available, globally accessible HTTP load balancer. It must:
> - Run across zones
> - Auto-heal unhealthy VMs
> - Log traffic for observability

✅ This setup mirrors how real-world production systems are deployed for uptime and scale.

---

## ✅ Key Takeaways

- Use **MIGs** for auto-healing and cross-zone scaling
- GCP’s **HTTP Load Balancer** is global, fast, and fully managed
- Health checks remove bad instances from serving traffic
- Logging adds visibility into real traffic and backend health
- This is a foundational DevOps skill for building resilient cloud infra

---