aws-xray-kubernetes

Code examples showing how to run AWS X-Ray on a Kubernetes cluster for deep application insights. Please also see the acompanying blog post for background information.

Changelog

• 04/05/2020 Clean up permissions for X-Ray daemon. Expose tcp and udp port.
• 03/31/2020 Update deployment specs to latest k8s version. Use official AWS X-Ray Docker image.
• 03/17/2020 Update sample app dependencies

Run AWS X-Ray on Kubernetes

The xray-daemon folder contains the code required to build and deploy an AWS X-Ray daemon Docker image and deploy this to an existing EKS or Kubernetes cluster.

Images have been built and pushed to Docker Hub for easier consumption of this project.

Deploying

Set up the correct permissions in AWS IAM so your pod can utilize. Utilize the EKS feature for IAM for Service Accounts. See the file xray-k8s-daemonset.yaml or EKS Userguide for setup instructions. Then run kubectl apply -f xray-k8s-daemonset.yaml to install the AWS X-Ray daemons on your Kubernetes cluster.

Fallback option is to attache the IAM policy named arn:aws:iam::aws:policy/AWSXRayDaemonWriteAccess to the worker nodes in your cluster.

Demo apps

Two simple demo applications are provided to showcase how AWS X-Ray enables deep application insights into a microservices architecture.

Simply run kubectl apply -f k8s-deploy.yml to install both services. Look up the endpoint for service-a and send a couple of requests against this endpoint. Switch to the AWS X-Ray console and see how the traces are showing up in the console.

License

This project is licensed under the Apache 2.0 License.