Preparing Kubernetes Engine (OKE) for Disaster Recovery

Perform the following prerequisites before adding Oracle Kubernetes Engine (OKE) to the disaster recovery topology.

Ensure that you have access to the OCI console and you meet all the requirements to create a Disaster Recovery Protection Group.

Create an OKE Cluster in the standby region if you have not already provisioned an OKE Cluster.
Create the Object storage buckets for backups in both the primary and standby regions. These buckets will be used to backup the configuration of the OKE cluster.
If the workload uses persistent volumes (PVs), then:
- Enable replication for each storage (block storage or file storage) used by the PVs.
- Ensure that the block volume groups or file storages are added to the DR Protection Group. See Preparing Block Storage for Full Stack Disaster Recovery and Preparing File System Mount or Unmount on Windows Instance for more information.
If your workloads use the OCI Native Ingress Controller, then, create a Load Balancer in the standby region that maps to each Load Balancer used by the primary cluster.
If your workloads use OCI Vault to store Kubernetes secrets, then, create a vault in the standby region that maps to each vault used by the primary cluster. Enable replication for each vault or manually copy secrets to the vault in the standby region.
If you are using jump host to access the OKE API service, then install the following on the jump host:
Note

The images already have wget and curl installed. However, if you want to use the other image, then ensure to install wget and curl.
- Python3
- Kubectl
- Helm
- Install the following Python modules:
  - PyYaml - pip install PyYAML
  - oci-cli - pip install oci-cli

Related Topics

Parent topic: Prerequisites for Full Stack Disaster Recovery