Installation of Production Grade Kubernetes Cluster with KOPS

Installing a production-grade Kubernetes cluster using Kubernetes Operations (kops) on Amazon Web Services (AWS) involves several steps. Below is a structured guide to help you through the process.

Prerequisites

AWS Account: Ensure you have an active AWS account.

AWS CLI: Install the AWS Command Line Interface (CLI) and configure it with your credentials.

Installation instructions can be found here.

Configure it using: aws configure

root@DevOps:—# aws configure 
AWS Access Key ID [****************W4M5]: 
AWS Secret Access Key [****************jtII]:

kubectl: Install kubectl, the Kubernetes command-line tool.

Installation instructions can be found here.

kops: Install kops, the Kubernetes Operations tool.

Installation instructions can be found here.

Resources Required

IAM Permissions: Ensure your AWS user has permissions to create and manage EC2 instances, VPCs, IAM roles, and Route 53 records.

S3 Bucket: Kops uses an S3 bucket to store the cluster state. You will need to create one.

Step-by-Step Instructions

Step 1: Create an S3 Bucket

• 1. Log in to the AWS Management Console.

     2. Navigate to the S3 service.

     3. Click on "Create bucket".

     4. Enter a unique bucket name (e.g.,my-k8s-kops).

     5. Enable public access for this bucket and enable the checkmark on “I acknowledge”.

        

     6. Enable Bucket Versioning.

        

     7. Choose a region and configure other settings as needed.

     8. Click "Create bucket".

Step 2: Set Up Your Domain

• 9. If you use “xxx.xx.local”, we don’t need to use Route 53. If using Route 53, we need a domain name and we need to create a hosted zone for your domain.

     10. Note the nameservers provided by Route 53.

Step 3: Create a Kubernetes Cluster Configuration

• 11. Set environment variables for your cluster name and S3 bucket:

      12. export KOPS_CLUSTER_NAME=mycluster.example.com (or eg. K8s.kops.local)

      13. export KOPS_STATE_STORE=s3://my-k8s-kops

Step 4: Create the Cluster

• 14. Use kops to create the cluster configuration:

      15. kops create cluster --name ${KOPS_CLUSTER_NAME} --state ${KOPS_STATE_STORE} --zones us-east-1a --node-count 2 --node-size t2.medium --master-size t2.medium --dns-zone example.com

Replace us-east-1a and example.com with your desired availability zone and domain.

Step 5: Edit the Cluster Configuration (Optional)

• 16. You can edit the cluster configuration if needed:

      17. kops edit cluster --name ${KOPS_CLUSTER_NAME} --state ${KOPS_STATE_STORE}

Step 6: Validate the Cluster Configuration

• 18. Validate the configuration to ensure everything is set up correctly:

      19. kops validate cluster --name ${KOPS_CLUSTER_NAME} --state ${KOPS_STATE_STORE}

Step 7: Create the Cluster

• 20. Once validated, create the cluster:

      21. kops update cluster --name ${KOPS_CLUSTER_NAME} --state ${KOPS_STATE_STORE} --yes

          root@DevOps:—# kops update cluster --name kops.k8s.local --yes 
          *******************************************************************************
          A new kops version is available: 1.30.3 
          Upgrading is recommended More information: https://github.com/kubernetes/kops/blob/master/permalinks/upgrade kops.md#1.30.3
          ************************************************
          5804 apply cluster.go:545] Gossip DNS: skipping DNS validation 
          5804 executor.go:103] Tasks: 0 done / 92 total; 42 can run
          5804 vfs castore.go:590] Issuing new certificate: "etcd-clients-ca"
          5804 vfs castore.go:590] Issuing new certificate: "etcd-manager-ca-events"
          5804 vfs castore.go:590] Issuing new certificate: "etcd-manager-ca-main"
          5804 vfs castore.go:590] Issuing new certificate: "etcd-peers-ca-events"
          5804 vfs castore.go:590] Issuing new certificate: "ca"
          5804 vfs castore.go:590] Issuing new certificate: "etcd-peers-ca-main"
          5804 vfs castore.go:590] Issuing new certificate: "apiserver-aggregator-ca"
          

Step 8: Configure kubectl

• 22. After the cluster is created, configure kubectl to use the new cluster:

      23. kops export kubecfg --name ${KOPS_CLUSTER_NAME} --state ${KOPS_STATE_STORE}

Step 9: Verify the Cluster

• 24. Check the status of your nodes:

      25. kubectl get nodes

Conclusion

You have successfully installed a production-grade Kubernetes cluster using kops on AWS. For further management and scaling of your cluster, refer to the official kops documentation.