Contents

Tools and software

Manipulating clusters with kubectl

Contents

Overview

For advanced users and troubleshooting purposes only!

Most users never need to interact directly with the underlying Kubernetes clusters of Scalable Pixel Streaming. Manually modifying the details of the Kubernetes objects used by Scalable Pixel Streaming can cause unexpected behaviour or break the system entirely, necessitating a full reinstall that would delete all of your deployed Pixel Streaming applications, so proceed with caution.

Configuring kubectl access for AWS

To access any Amazon Elastic Kubernetes Service (EKS) Kubernetes cluster with the kubectl command, you will need to first install and configure the relevant tools:

  1. Follow the steps in the AWS CLI getting started guide to first install the AWS CLI tool for your operating system and then configure it to access your AWS account.

  2. Follow the guide from AWS to install the kubectl tool.

Once the AWS CLI tool is configured and the kubectl tool is installed, you can configure access to your Scalable Pixel Streaming Kubernetes cluster. If you have installed Scalable Pixel Streaming in multiple Kubernetes clusters across different geographic regions, then you will need to repeat these steps for each cluster:

  1. Identify the EKS cluster that was automatically created by AWS CloudFormation when Scalable Pixel Streaming was installed by navigating to the list of clusters for the selected geographic region. The EKS cluster will have the name you specified during the installation process, which defaults to “scalable-pixel-streaming”.

  2. Follow the guide from AWS to create a kubeconfig file to access the EKS cluster.

Configuring kubectl access for CoreWeave

Follow the official CoreWeave guide to obtain access credentials for your account and install the kubectl tool. Since there is only ever one global Kubernetes cluster per account, there is no need to distinguish between multiple clusters in different geographic regions.