If an Amazon Elasticsearch cluster is in a virtual private cloud (VPC) Kibana is only accessible inside this VPC. There are several way´s to access Kibana from outside that VPC as described here

This sample templates can be used as a starting point to provision a NGINX Proxy to access Kibna following the guidance from this post

With this Option only a single instance with an NGINX Proxy is deployed in the given Subnet.

Deploy to your account:

aws cloudformation deploy \
    --template-file es-nginx-single.yaml \
    --stack-name <STACKNAME> \
    --parameter-overrides ESHost=<ES-HOST> \
        CognitoHost=<COGNITO-HOST> \
        VPCID=<VPC> \
        SubnetID=<SUBNET> \
    --region <REGION>

With this Option, the Proxy is deployed in an Autoscaling Group with an Application Load Balancer in Front. This is the recommended deployment for High Availability.

You must specify at least 2 public Subnets. Every Subnet must be in another Availability Zone.

You must also provide a Certificate for the HTTPS Listener of the Application Load Balancer.

aws cloudformation deploy
    --template-file es-nginx.yaml \
    --stack-name <STACKNAME> \
    --parameter-overrides ESHost=<ES-HOST> \
        CognitoHost=<COGNITO-HOST> \
        VPCID=<VPC-ID> \
        ALBSubnets=<SUBNET1-ID>,<SUBNET2-ID> \
        SSLCertificate=<CERTIFICATE-ARN>
    --region <REGION>

After the stack is created the Link to the dashboard is available in the Stack Outputs.

Or via console.

aws cloudformation describe-stacks  \
        --stack-name <STACKNAME>  \
        --query 'Stacks[0].Outputs[?OutputKey==`KibanaEndpoint`].OutputValue' \
        --region <REGION>   

To avoid incurring charges, delete the AWS CloudFormation stack when you are finished experimenting:

• Sign in to the AWS CloudFormation console and choose your stack.
• Choose Delete to delete all resources.

Standard AWS charges apply to the resources you deploy with this template.

Amazon EC2 and Elastic Load Balancing provides customers in the AWS Free Tier free usage of up to 750 hours per month for each compute and load balancing. This template will use t2.micro instances by default which are eligible for the Free Tier.

If you found yourself wishing this set of frequently asked questions had an answer for a particular problem, please submit a pull request. The chances are good that others will also benefit from having the answer listed here.

See the Contributing Guidelines for details.

This library is licensed under the MIT-0 License. See the LICENSE file.