Real-World AWS DevOps

Define measurable service-level indicators and service-level objectives for an AWS-hosted application.

Derive recovery time and recovery point objectives from business impact statements.

Select EC2, ECS/Fargate, or EKS for a workload and justify the decision against stated requirements.

Draw a multi-AZ ECS/Fargate reference architecture using appropriate AWS networking, load-balancing, security, and observability services.

Trace an application request through DNS, TLS termination, an Application Load Balancer, private subnets, and ECS tasks.

Structure Terraform root configurations and reusable modules for separate AWS environments.

Configure encrypted remote Terraform state with state locking and restricted access.

Provision a VPC with public and private subnets, route tables, internet access, and controlled outbound connectivity using Terraform.

Translate an application traffic flow into least-privilege security group rules.

Create distinct IAM trust and permissions policies for a delivery pipeline, ECS task execution role, and application task role.

Evaluate a Terraform plan for destructive changes, security exposure, configuration drift, and rollback risk.

Build a minimal, non-root container image with a meaningful application health check.

Configure an ECR repository with immutable image tagging, vulnerability scanning, and lifecycle controls.

Create an ECS task definition with explicit resource limits, port mappings, health settings, and CloudWatch log configuration.

Deploy a multi-AZ Fargate service behind an Application Load Balancer and validate target health.

Inject configuration and secrets into ECS tasks from Systems Manager Parameter Store or Secrets Manager without exposing values in code.

Configure ECS service autoscaling and failure recovery from measurable workload signals.

Configure GitHub Actions to access AWS through short-lived credentials using OpenID Connect.

Build, test, scan, tag, and push a container image to ECR in a repeatable pipeline.

Add Terraform validation, plan review, environment approval, and controlled apply stages to a delivery workflow.

Promote one immutable image artifact across environments without rebuilding it.

Select and execute a rolling or blue-green ECS deployment with health gates and a tested rollback path.

Query structured ECS application logs with CloudWatch Logs Insights to isolate a production fault.

Create CloudWatch metrics, dashboards, and actionable alarms for traffic, errors, latency, saturation, and ECS service health.

Diagnose ECS task startup and deployment failures involving images, IAM, secrets, resources, or health checks.

Diagnose request failures across Route 53, an Application Load Balancer, security groups, routes, and ECS targets.

Use CloudTrail, AWS Config, image scan results, and AWS security findings to identify and remediate a risky production change.

Map the Fargate architecture to an EC2 Auto Scaling design and identify the added responsibilities for bootstrapping, patching, scaling, and recovery.

Map the Fargate architecture to EKS resources and identify the roles of pods, services, ingress, workload identity, and cluster autoscaling.

Estimate major AWS cost drivers in the reference architecture and recommend requirement-aware optimizations.

Write an incident runbook for a failed deployment or degraded ECS service, including triage, containment, rollback, recovery, and follow-up actions.

Process a realistic company change request by documenting architecture impact, security risk, implementation steps, validation, rollout, rollback, monitoring, and operational handoff.