When you are running kubernetes on GKS, AKS, etc, you use your cloud provider specific storage bucket such as Amazon S3, google cloud storage, azure blob storage, to store your raw binary content in a super fast, easy to use manner.

But when you are deploying Kubernetes behind the firewall, you do not have a storage bucket SDK from your cloud provider.

What is the alternative to this when you are using, for example, OpenShift? Do we have to just use persistent volumes?

1 Answer 1


You have many, many options. The following all provide an S3-compatible object storage API:

  • If you are using OpenShift, you can install OpenShift Data Foundation, which includes the Noobaa object storage service. This provides ObjectBucket and ObjectBucketClaim resources that are analgous to PersistentVolume and PersistentVolumeClaim resources.

  • You could run MinIO, which can be deployed on Kubernetes via an operator or a helm chart.

  • You can run your own Ceph cluster, which includes the object gateway.

  • You could run OpenStack Swift.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.