Identify EC2 Instances used in ElasticSearch Cluster

Question

I have setup an ElasticSearch cluster in AWS with couple of m4.large.elasticsearch as master nodes and 4 i3.xlarge.elasticsearch as data nodes. I am however unable to find any of these instance in the EC2 console. I can get directed to Cloudwatch metrics for the node based on its instance-id from ES console. This instance-id is in different format than the ones assigned to regular Ec2 instance.

Aren't these also types of ec2 instances? How do I figure out the additional information about these nodes used in the ElasticSearch cluster? I am trying to tag these instances, if possible.

Answer 1

AWS-elasticsearch is a fully managed service, which means you can just specify the types of nodes(master and data nodes) and AWS behind the scene provision and manages them, also as its pay as you use model, AWS won't allow you to control these instances.

In-fact in fully managed service, AWS automatically scale-up and scale-down these servers based on your requirements and that is the advantage of it, the downside is now you can't have SSH access or control the infrastructure of fully-managed service.

Note: This is not just with AWS-elasticsearch, but for all fully managed services like DyanmoDB, Elastic-cache, etc, hence you can't tag these instances.

Also same is mentioned in Faq's of fully managed services