This topic describes how to back up and restore Kubernetes clusters provisioned by VMware Tanzu Kubernetes Grid Integrated Edition (TKGI).
Overview
TKGI uses BOSH Backup and Restore (BBR) to back up and restore Kubernetes clusters provisioned by TKGI. For more information about BBR, see BOSH Backup and Restore.
BBR orchestrates triggering the back up or restore process on the BOSH deployment, and transfers the backup artifacts to and from the BOSH deployment.
In context of Kubernetes clusters provisioned by TKGI, BBR backs up and restores the following components:
- TKGI cluster control plane nodes
- TKGI cluster worker nodes
BBR can also be used to back up and restore the TKGI Management Plane. See Back Up and Restore TKGI Components.
In context of TKGI, BBR does not back up and restore:
- Kubernetes workloads, see Backing up and restoring Kubernetes workloads.
- Ops Manager VM (including the BOSH Director and TKGI tiles), see Backing up and restoring Ops Manager.
- NSX-T objects and resources, such as load balancers, see Backing up and restoring TKGI Infrastructure.
- Harbor VM
To use BBR to back up and restore Kubernetes clusters provisioned by TKGI, see the following topics:
- Install and Configure BOSH Backup and Restore
- Back Up Kubernetes Clusters Provisioned by Tanzu Kubernetes Grid Integrated Edition
- Restore Kubernetes Clusters Provisioned by
Testing Considerations
As part of your TKGI back up and restore planning and testing, consider the following test scenario.
- Back up a TKGI-provisioned Kubernetes cluster using BBR.
- Delete the cluster control plane and worker node VMs. Delete the disks and references.
- Restore cluster nodes using BBR. Delete old worker nodes using kubectl and restart kubelet.
Confirm the cluster nodes have been restored, and the cluster is operational. If an application was deployed before cluster back up, the VIP of the load balancer for the application changes. To recover without application redeployment, create the service using a static IP. How to do this is described in the Workload Back Up and Restore documentation. If an application was deployed after cluster back up, the application is no longer available and the NSX-T objects created for the application are automatically deleted. In this case you will need to restore the application using Velero and the NSX-T objects using NSX Manager.
