ACD2022 Session Report:Enabling automated multi-region failover and failback using Route 53

Sathyajith Bhat

Session Overview:

Agenda:

What's DNS?

4 types of Server

Recursive resolver - acts as a middleman between client & nameserver.

Root nameserver - First step of hostname resolution and serves as a reference to another location fun fact: there are only 13 root name servers!

TLD Nameserver - Next step of host resolution and holds the TLD records Authoritative server - holds the actual IP address

DNS Terminology:

Zone File: this file contains a mapping to different IP

Record: this maps the Domain name to the Ip address

TTL: How Long a Result is cached

We Can query Domain names using Dig

For eg:

dig https://dev.classmethod.jp/

What's Route 53

AWS' Managed DNS service Route 53 lets you:

the simple boring stuff: register and manage DNS records in private & public hosted zones

Advanced routing (Latency/Geo/IP-based routing)

Deep Dive into Route 53 Record & Health Check

Route 53 Record Types:

1. Standard Routing no additional Routing Feature

2. Failover: this has 2 Records attached to it

2.1: failover Active/passive: only one has health check enabled if 1st Record fails it will shift to 2nd Record

2.2: failover Active Active: both of the Records have  health check enabled if either of 1 fails it will failover to another record

3. Weighted: the record has multiple records with weight associated with it. the request is distributed based on the weights

4. latency: Traffic is distributed based on latency (calculated by amazon)

5. IP based: let us map user IP to endpoint mapping

6. GeoLocation: In this Record is associated with geolocation, geo-targeting can be traffic from continent & country

Route 53  HealthCheck

Route 53 monitor the Health of Resource

3 types of Health Check

3.1. monitor an endpoint

3.2. monitor cloudwatch alarm

3.3 Calculate Health Check

HealthCheck- Monitoring Endpoint

We can monitor Specific endpoints by:

1.Protocol

2.Domain Name

3.IP

4.path

Advance Config Options :

1.Time Between Health Check

2.string Matching

3.failure thresholds

4.Inverted Health Checks

5.Customizable regions