INFRASTRUCTURE LEARNINGINFRASTRUCTURE LEARNING
インフラ大全インフラ大全IT INFRASTRUCTURE / LEARNING
Infrastructure learning guide
Find a lesson or lab/

Esc to close · Search terms stay on your device

LEARNING日本語

Topic lessonTopic: Audit coverage and recovery design

Route 53 health checks and DNS failover: from lookup to business recovery

Check the DNS answer, the connection and the outcome separately. Trace health checks, cached answers and an actual submission in one fictional incident.

What you will learn

Relate health checks, failover records, recursive resolvers and a business validation to distinct observations.

Prerequisite

This example assumes a fictional public HTTP service. No actual DNS or application changes are involved.

Trace the decision from the health check to the client

Route 53 can use health checks for specified resources, other health checks or CloudWatch alarms. In a DNS failover setup, health status can influence record selection.

Health decisions and observed client paths
Health decisions and observed client paths1: East health — Select a record — Authoritative DNS. 2: Authoritative DNS — Observe the resolver — Recursive resolver. 3: Recursive resolver — Test the transaction — Business check123East healthAuthoritative DNSRecursive resolverBusiness checkHealth decisions and observed client paths1: East health — Select a record — Authoritative DNS. 2: Authoritative DNS — Observe the resolver — Recursive resolver. 3: Recursive resolver — Test the transaction — Business check123East healthAuthoritativeDNSRecursiveresolverBusiness check
  1. East health → Authoritative DNSSelect a record
  2. Authoritative DNS → Recursive resolverObserve the resolver
  3. Recursive resolver → Business checkTest the transaction
East health
Monitoring URL response
Authoritative DNS
Answers change to West
Recursive resolver
A prior answer may persist
Business check
Connect to West and submit

Seeing a West DNS answer and completing a West write

Authoritative DNS answers West at 10:02; an East response remains cached
MeasurementResultWhat it establishes
10:02 Authoritative DNSWestAuthoritative response changed
10:02 Site A resolverEastSite A still has the old response
10:03 Lookup from site BWestSite B obtained a West address
10:04 Submit and read back in WestFailedBusiness service remains unavailable

Resolvers can cache answers based on TTL. Earlier answers and open connections may persist after the authoritative answering service changes its response.

Find where the chain stops

Check what the monitor measures. Compare the configured path and timestamps with the transaction test. Record which dependencies the monitoring path actually exercises.

Measure answers separately. Record authoritative response, recursive response and the destination of an actual connection, per site and time.

Define when to reopen. Pair a representative submission with read-back. Use a valid transaction, in addition to the monitoring URL, as evidence of recovery.

Check your decision

Choose an option to see the result and explanation.

Authoritative DNS now answers West. What does this establish?
Read the answer and reason

Answer: The authoritative response changed · The measured authoritative server returns a different answer.

Only site A still resolves East. What do you check next?
Read the answer and reason

Answer: Site A recursive response and TTL · Compare the site A resolver response and remaining TTL with the authoritative response.

Clients connect to West but the write fails. What is the outcome?
Read the answer and reason

Answer: West fails the business check · Business recovery requires the write and read-back.

Check, then explain in your own words

  1. Read what the health check measures.
  2. Observe authoritative, recursive and actual destinations.
  3. Verify the required transaction before reopening.
Check your understanding: Does an authoritative West response prove recovery?

Check where users connect and whether the business transaction completes.

DNS is only one step in the failover.

Sources and conditions checked

The official references used in this lesson are collected here with their scope, conditions and corresponding sections. Independent expert review and live AWS comparison have not been performed.