DAY 2: EC2 FUNDAMENTALS & INSTANCE MANAGEMENT

Complete Step-by-Step Guide

Welcome to Day 2 of the AWS EC2 Series a hands-on journey to mastering EC2 for DevOps and Cloud Engineers.

---

🛠 HANDS-ON LAB SCENARIO

Lab: Multi-Instance Deployment & Management

Objective: Launch instances across AZs and practice lifecycle operations.

Step 1: Launch Instances

# Create key pair first
aws ec2 create-key-pair --key-name MyKeyPair --query 'KeyMaterial' --output text > MyKeyPair.pem
chmod 400 MyKeyPair.pem

# Launch instances in different AZs
aws ec2 run-instances \
    --image-id ami-0c02fb55956c7d316 \
    --count 3 \
    --instance-type t3.micro \
    --key-name MyKeyPair \
    --placement AvailabilityZone=us-east-1a

Step 2: Instance Operations

# Get instance IDs
INSTANCE_IDS=$(aws ec2 describe-instances --query 'Reservations[].Instances[].InstanceId' --output text)

# Stop one instance
aws ec2 stop-instances --instance-ids i-1234567890abcdef0

# Change instance type (must be stopped)
aws ec2 modify-instance-attribute --instance-id i-1234567890abcdef0 --instance-type "{\"Value\": \"t3.small\"}"

# Terminate instance
aws ec2 terminate-instances --instance-ids i-1234567890abcdef0

Step 3: Monitoring & Verification

# Check instance states
aws ec2 describe-instances --query 'Reservations[].Instances[].[InstanceId,InstanceType,State.Name]' --output table

# View pricing information
aws ec2 describe-spot-price-history --instance-types t3.micro --product-descriptions "Linux/UNIX" --start-time 2024-01-01T00:00:00Z

REAL-WORLD PROBLEM & SOLUTION

Problem Statement:
Company XYZ spends $5,000/month on EC2 with predictable 24/7 workload.
How to reduce costs by 40%?

Solution Implementation

Step 1: Analyze Current Usage

# Get current instance inventory
aws ec2 describe-instances \
  --query 'Reservations[].Instances[].[InstanceId,InstanceType,State.Name]' \
  --output table

# Check Cost Explorer (via AWS Console)

Step 2: Implement Reserved Instances

# Purchase Reserved Instances (conceptual)
aws ec2 purchase-reserved-instances-offering \
  --reserved-instances-offering-id <offering-id> \
  --instance-count 10 \
  --instance-type m5.large

###Step 3: Cost Monitoring Setup
# Create budget alert
aws budgets create-budget \
  --account-id 123456789012 \
  --budget file://budget.json \
  --notifications-with-subscribers file://notifications.json

budget.json

{
  "BudgetName": "ec2-monthly-budget",
  "BudgetLimit": {
    "Amount": "3000",
    "Unit": "USD"
  },
  "CostFilters": {
    "Service": "Amazon EC2"
  },
  "TimeUnit": "MONTHLY",
  "BudgetType": "COST"
}

🚀 CORE CONCEPTS Q&A

Q1: What's the difference between stopping vs terminating an instance?

Answer:

STOPPING:
- OS gracefully shuts down
- EBS root volume persists
- Instance can be restarted
- Keeps same private IP (usually)
- Charging stops for instance, continues for EBS

TERMINATING:
- Instance is permanently deleted
- EBS root volume deleted by default
- Cannot be recovered
- All data on instance store lost
- All charging stops

Q2: Explain EC2 purchasing options

Answer:

On-Demand:

• Pay by second/hour
• No long-term commitment
• Most flexible, highest cost
• Use case: unpredictable workloads

Reserved Instances:

• 1-3 year commitment
• 30-60% cost savings
• Types: Standard, Convertible, Scheduled
• Use case: predictable, steady-state

Spot Instances:

• Up to 90% savings
• Can be terminated with 2-minute warning
• Use case: fault-tolerant, flexible workloads

Savings Plans:

• 1-3 year commitment
• Flexible across instance family/region
• Use case: consistent usage patterns

🎯 COMMON INTERVIEW QUESTIONS

Technical Questions

🟡 "What happens to EBS volumes when you terminate an instance?"
By default, root EBS volume is deleted, additional volumes persist.

🟡 "How do you change instance types?"
Stop instance → Change instance type → Start instance.

🟡 "What's the difference between reboot and stop/start?"
Reboot = OS restart (same hardware).
Stop/Start = may move to new hardware, new public IP.

Scenario-Based Question

"A company has applications with different reliability requirements. How would you recommend instance types?"**

Critical production: On-Demand/Reserved
Testing/Dev: Spot Instances
Batch processing: Spot Fleets
Long-running services: Reserved Instances

Certification-Style Questions

Question 1

Your company needs to run a critical database server for 3 years. Which EC2 option provides the lowest total cost?

Options:

• A) On-Demand Instances
• B) Spot Instances
• C) Reserved Instances (All Upfront)
• D) Savings Plans

Answer: C - Reserved Instances with All Upfront payment for predictable long-term workloads.

---

Question 2

What happens to data on an instance store volume when you stop an instance?

Answer: Data is preserved on EBS volumes but lost on instance store volumes when instance is stopped/terminated.

---

Question 3

Which instance would be most cost-effective for a batch processing job that can handle interruptions?

Answer: Spot Instances — up to 90% savings for interruptible workloads.

---

Troubleshooting Common Issues

Issue: Instance failed to launch due to insufficient capacity

Solutions:

# Try different instance type
aws ec2 run-instances --instance-type t3.small ...

# Try different AZ
aws ec2 run-instances --placement AvailabilityZone=us-east-1b ...

# Use capacity-optimized allocation strategy for Spot

Issue: Cannot Connect to Instance via SSH

✅ Checklist

• ✅ Security group allows SSH (port 22)
• ✅ Correct key pair
• ✅ Instance is in running state
• ✅ Public IP address is correct

---

📊 Day2 Hands-On Checklist

• ✅ Launch 3 instances in different AZs
• ✅ Practice stop/start/terminate operations
• ✅ Change instance type on stopped instance
• ✅ Create and attach EBS volume
• ✅ Configure security groups for SSH access
• ✅ Set up basic CloudWatch monitoring
• ✅ Calculate potential Reserved Instance savings