Comment on page
Syncing Kubernetes Clusters for CD
Create or link your clusters to Ozone for the purpose of CI/CD automation and workload observability
Once you've got the credentials setup in the form of a cloud provider, you can now use ozone to deploy container workloads, run CI/CD workflows and gain improved observability for container workloads.
Any cluster that is synced is scoped to the current project. To use this cluster in another project, an account admin must add it to another project.
- Once a cloud provider is created, clusters are automatically synced back to ozone
- You can also sync any clusters added to your account post the first sync by clicking on the "Sync Cluster" button
Creating a Cloud Cluster from Ozone
Any cluster that is attached is scoped to the current project. To use this cluster in another project, an account admin must add it to another project.
Use this option if the cluster is located on-premise from distributions such as PKS, TKG and Openshift.
This is the typical use case, but in theory you can attach any CNCF compliant Kubernetes cluster (including public clouds)
- Navigate to Resources -> Clusters
- Click on +ATTACH
- Enter the cluster name
- Select the type of the cluster (Distribution type)
- If the cluster has no exposed API service, do mark it as a private cluster
- In this case, ozone sets up a private software defined network to enable you to perform automated CI/CD for container workloads.
- Apply the YAML file on the cluster by using the commands shown on the screen
- The cluster should now be synced to the Resources -> Cluster section
- Once the cluster is linked successfully, it will be visible with an Active Status and a Green Check mark for Ozone's Agent status, indicating that it has successfully been created and setup.
Enable the following settings from either:
- The CI/CD Settings screen as the last step before the cluster is created
- Edit screen of the cluster once it has been synced to ozone (Resources -> Cluster -> Select Cluster -> Edit)
Supported CI/CD Settings for automation and observability: