Route 53
Last updated
Last updated
Route-53 viblo.asia | DNS's FAQ | FAQs | Routing policy |
Can be used to manage both public and private domain.
user can access app using these domain name.
provice health check for resources -> route traffic away from unhealthy resouces.
Route53 responses to DNS queries.
Performance
Low query latency for end-user
Low update latency for DNS record management
Scalable
Can handle large query volume.
Flexibility
Route policy
CNAME:
point host name to another hostname. ex: app.mydomain.com -> blabla.anything.com
ONLY FOR NON ROOT domain. ex: aka.sth.mydomain.com
Alias (A or AAAA)
free
point a host name to ONLY ONE AWS resource.
ex: app.mydomain.com -> blabla.amazonaws.com
work for both ROOT and NON-ROOT domain.
Always of type A or AAAA
A: IPv4
AAAA: IPv6
NS: Name Server for Hosted Zone
MX: mail exchange record
Target group
point to the host
point to AWS resources
Target group value
DNS or Hostname
DNS
Response
return DNS/Hostname. Need to query Alias to know the IP
A or AAAA
$
Yes
No
TTL
Yes
No
Root domain
No
Yes
Public: how traffic is routed in the internet.
Private: determine how traffic is routed within VPC
Weighted: eg 5% traffic go to production environment.
Sum of total weighted % is not neccessary 100%
DNS record need to be same type, same domain or subdomain.
If you set weight = 0 for all records, the traffic will be balanced for all.
Latency: evaluate the latency between your users & AWS regions, to minimize the latency.
Use cases:
Load balancing
Testing new software version.
Letting Route 53 serve user requests from the AWS Region that provides the lowest latency.
Use case: improve performance / reduce latency.
Lets you choose the instances that will serve traffic based on the location of your users
For example: US -> instance-1, France -> instance-2
Lets Amazon Route 53 route traffic to your resources based on the geographic location.
You can also optionally choose to route more traffic or less to a given resource by specifying a value, known as a bias. A bias expands or shrinks the size of the geographic region from which traffic is routed to a resource.
HTTP Health check are only for PUBLIC resouces.
Health check -> Automated DNS Failover.
Route53 can monitor (the health) of the end-point
Support methods: HTTP, HTTPS, TCP
PASS only when the response code is 2xx or 3xx.
Health check can work with other healthchecks.
TTL is required setting value for all record types, except A
Check TTL by nslookup -q=SOA google.com or dig google.com SOA
Each Amazon Route 53 account is limited to a maximum of 500 hosted zones and 10,000 resource record sets per hosted zone.
DNS (Domain Name Service): map domain name to IP addresses, or resources (EC2, ELB, S3 buckets, CloudFront...). For example:
Domain Registra: Godday, お名前.com, Route53...
Name server: server that converts name -> ip
Top Level Domain: .com, .vn
Second Level Domain: google.com, aws.com
TTL (Time to live): the client will cache the result (return from DNS server) for TTL minutes. To save cost, set the TTL value higher for lesser requests to DNS server.
If you want to check the health of PRIVATE resources, then use CloudWatch Metric. This means healthcheck just monitor the Metric, not that PRIVATE resources
Health check in Route53 are only for PUBLIC resources.
