You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
-Step 1 (Make sure aws-cli is installed through which terraform will access your keys)
terraform-code-deploy-jenkins-server
Project Name: Terraform to deploy a server and pre-loaded with Jenkins-server and Kubectl
Steps for the project
1. terraform init
2. terraform plan
3. terraform apply
4. Make sure google-cli is Installed on the server machine.
5. Then, run commands : $gcloud init
$gcloud auth application-default login {this is for authentication with google cloud}
$gcloud services enable compute.googleapis.com
$gcloud services enable container.googleapis.com
-Step 2 (Will make use of google-cli on the server installed in step 1)
jenkins-pipeline-deploy-to-gcp-kubernetes-cluster
Project Name: End to end deployment of Applications to Kubernetes using a Jenkins CI/CD Pipeline
Steps for the project
1. Create a Jenkins Server with all the dependencies, libraries and packagies needed.
2. Once completed, access the Jenkins server and Set it up
3. Run the jenkins-pipeline-deploy-to-gcp to create Kubernetes Cluster, create deployments and Services
4. Test that the application is running
5. Then, when all is running. In the root directory of this folder, use terraform to run final-terraform.tf to assign subdomain to their respective IP address(NB: you get the IP's when the deployments and services are up and running on the cluster).