Get hands-on experience with 20+ free Google Cloud products and $300 in free credit for new customers.

Creating a GKE Autopilot clone from a GKE standard cluster - TF + helm or Backup for GKE???

We are seeing a few performance issues with our large scale GKE Standard Production deployment .I want to be able to quickly standup an isolated sandbox GKE Autopilot cluster to troubleshoot the production issues we are seeing. What is the fastest way to standup a new sandbox GKE Autopilot cluster? Remember that PROD is GKE Standard and sandbox is GKE Autopilot. Can i use the Backup for GKE option in this case will it work? Another straightforward option which is pretty obvious is to provision a new GKE Autopilot cluster and deploy the workloads from scratch.

Solved Solved
0 2 341
1 ACCEPTED SOLUTION

Hi @dheerajpanyam,

Welcome to Google Cloud Community!

Yes, it may work and could still be chosen as an option. However, Backup for GKE has its limitations. To summarize, Backup for GKE does not back up the following:

  • GKE cluster configuration information such as node configuration, node pools, initial cluster size, or enabled features.
  • Only the Kubernetes resources that describe the workload and refer to the container images are backed up. If an image referenced by a workload manifest in a backup is removed from its image repository, then a subsequent restore of that configuration won't successfully restore the workload.
  • Configuration information or state of services outside the cluster, such as Cloud SQL or external load balancers.
  • Only Persistent Disk type volumes are backed up. Other volume types, such as Filestore NFS or Google Cloud NetApp Volumes, aren't backed up. 

Note: Backup for GKE is designed for disaster recovery, backup, and restore purposes. 

For recommendation, the most straightforward and fastest way is to provision a new GKE cluster - autopilot(Sandbox) and deploy all the workloads from scratch. This allows you to configure the cluster with the exact configurations and replication of workloads needed. 

For best practices in Managing Google Kubernetes Engine, you can refer to this documentation.

I hope the above information is helpful.

View solution in original post

2 REPLIES 2

Hi @dheerajpanyam,

Welcome to Google Cloud Community!

Yes, it may work and could still be chosen as an option. However, Backup for GKE has its limitations. To summarize, Backup for GKE does not back up the following:

  • GKE cluster configuration information such as node configuration, node pools, initial cluster size, or enabled features.
  • Only the Kubernetes resources that describe the workload and refer to the container images are backed up. If an image referenced by a workload manifest in a backup is removed from its image repository, then a subsequent restore of that configuration won't successfully restore the workload.
  • Configuration information or state of services outside the cluster, such as Cloud SQL or external load balancers.
  • Only Persistent Disk type volumes are backed up. Other volume types, such as Filestore NFS or Google Cloud NetApp Volumes, aren't backed up. 

Note: Backup for GKE is designed for disaster recovery, backup, and restore purposes. 

For recommendation, the most straightforward and fastest way is to provision a new GKE cluster - autopilot(Sandbox) and deploy all the workloads from scratch. This allows you to configure the cluster with the exact configurations and replication of workloads needed. 

For best practices in Managing Google Kubernetes Engine, you can refer to this documentation.

I hope the above information is helpful.

Thanks @francislouie You summed it up very well from your note, makes sense thanks much.