GregLab | Exam Prep

Reliability and Business Continuity

Health Checks, Multi-AZ Capacity, and Fault-Tolerant Operation

Core

Build and troubleshoot layered availability using healthy capacity, load-balancer target checks, Route 53 endpoint health, replacement, and managed Multi-AZ failover.

Aligned to AWS Certified CloudOps Engineer - Associate (SOA-C03) Version 1.1, verified August 24, 2026.

Why this matters

A load balancer, health check, Auto Scaling group, or DNS record cannot create healthy capacity by itself. Availability depends on correctly connected layers and independent failure zones.

Must Know

  • ELB target health depends on the configured protocol, port, path, expected response, thresholds, target listener, and allowed network path.
  • Place healthy service capacity across Availability Zones and preserve enough per-zone capacity; an ELB cannot distribute to capacity that was never deployed.
  • Route 53 health checks produce endpoint health evidence. Failover records must associate and evaluate the appropriate health state.
  • Stateless compute resilience combines multi-AZ placement, health-based routing, and replacement of failed capacity.
  • RDS Multi-AZ maintains a managed availability topology and failover path; a read replica primarily serves read scaling and has a different promotion model.
  • Elasticity handles demand variation; fault tolerance handles component or zone failure. Some workloads require both.

Compare and Distinguish

  • ELB target health governs targets behind the load balancer; Route 53 health and routing govern DNS answers.
  • Multi-AZ addresses an Availability Zone or host failure; Multi-Region DR addresses a wider recovery boundary.
  • Auto Scaling replaces capacity; load balancing distributes requests; neither alone proves the application is healthy.
  • RDS Multi-AZ availability and read-replica scaling solve different operational needs.

Scenario examples

  • Scenario: A target answers on the application URL but fails the configured health path. Correct the health path or target response rather than lowering DNS TTL.
  • Scenario: An AZ fails and all registered targets disappear. Restore multi-AZ capacity and health before changing outer DNS routing.
  • Scenario: A database must fail over after primary host loss. Use the appropriate RDS Multi-AZ deployment and test client reconnection.

Exam traps

  • Do not infer target health from one manual request to a different path.
  • Do not expect a routing policy to generate health evidence automatically.
  • Do not call one-AZ Auto Scaling fault tolerant.
  • Do not treat a read replica as identical to a synchronous standby.

Key takeaways

  • Healthy capacity is the foundation of availability.
  • Measure health at the layer that routes traffic.
  • Distribute capacity before relying on failover.
  • Test client and dependency behavior during failover.
How it works
  • Confirm that sufficient service capacity exists in each intended Availability Zone.
  • Test the configured health endpoint and trace its response through target networking and load-balancer evaluation.
  • Validate DNS or database failover from a client, including reconnection, cached names, and dependency health.
When to use it
  • Use Elastic Load Balancing target health to control which registered targets receive traffic; use Route 53 health evaluation and routing to control DNS answers.
  • Design Multi-AZ capacity for host or Availability Zone failure; select a Multi-Region disaster recovery pattern when the recovery boundary extends beyond one Region.
  • Use Amazon EC2 Auto Scaling to replace or add capacity and load balancing to distribute requests, then verify application health separately.
  • Choose Amazon RDS Multi-AZ for managed availability and failover; use read replicas when the requirement is read scaling.
Security and governance implications
  • Expose only the minimum health-check path and network access needed for the component performing the check.
  • Keep failover and replacement permissions scoped while retaining health and change evidence for review.
Failure signals and diagnosis
  • Start at the innermost unhealthy target, then move outward to load balancing and DNS.
  • Verify both health-check reachability and the application response.
  • During failover testing, inspect connection reuse, DNS caching, replacement capacity, and dependency health.
More detail
  • Know how target protocol, port, path, response, thresholds, listener, and network access determine ELB target health.
  • Relate deployed capacity in each Availability Zone to load-balancer distribution and Auto Scaling replacement behavior.
  • Use Route 53 health for DNS routing decisions and RDS Multi-AZ for the supported managed database failover boundary.

Ready for the quiz?

  • Targets are unhealthy behind a load balancer even though the application responds on another URL. Which health-check settings and network evidence should you compare?
  • Why does an RDS read replica not satisfy the same requirement as an RDS Multi-AZ deployment?
  • What evidence distinguishes an unhealthy ELB target from a healthy target that Route 53 has stopped returning?

Related objectives

  • D2.2 — Task 2.2: Implement highly available and resilient environments.
  • 2.2.1 — Configure and troubleshoot Elastic Load Balancing (ELB) and Amazon Route 53 health checks.
  • 2.2.2 — Configure fault-tolerant systems (for example, Multi-AZ deployments).

Learn more

Free AWS Certified CloudOps Engineer - Associate prep

Build focused SOA-C03 quizzes from skill areas, topics, and product references.

Practice with exam-style multiple-choice and multiple-response questions, score breakdowns, explanations, and a compact reference for this lane's official exam domains.

Read Topics Build a quiz

Exam Weights

Exam snapshot

SOA-C03 at a glance

Category
Associate
Duration
130 minutes
Questions
65 total; 50 scored and 15 unidentified unscored
Formats
Multiple choice and multiple response
Scoring
100–1,000 scaled score; 720 minimum passing score

Quiz builder

Choose your practice set

Mode

Exam fidelity: AWS documents 65 questions in 130 minutes: 50 scored and 15 unidentified unscored, using multiple-choice and multiple-response formats. This site's practice accuracy and readiness do not reproduce AWS's 100–1,000 scaled scoring or identify unscored items. Difficulty labels describe this site's Associate-level scenario complexity, not an AWS-published question rating.

Reference

SOA-C03 topics and reference map

Study links

SOA-C03 resources