Chapter 5. Developer previews

Applications that manage resiliency at the application level can now use storage class with single replica without data resiliency and high availability.

This update allows the rook-ceph-operator-config environmental variables to change the defaults when OpenShift Data Foundation is upgraded from version 4.5 to another version. This was not possible in the earlier versions.

With this update, it is possible to change the target size ratio for any pool. In the previous versions, the pools deployed by rook in the Ceph cluster were assigned a target_ratio of 0.49 for both RBD and CephFS data and this could cause an under-allocation of PGs for the RBD pool and an over-allocation of PGs for the CephFS metadata pool. For more information, see Configuration of pool target size ratios.

Ephemeral volume support enable a user to specify ephemeral volumes in its pod specification and tie the lifecycle of the PVC with the pod.

This feature supports multisite configurations such as Zone, ZoneGroup, or Realm for internal or external OpenShift Data Foundation clusters. This setup helps to replicate data into different sites and recover the data incase of failure.

In this release, a lightweight object storage solutions is provided for single node OpenShift (SNO) clusters using MCG with backingstore layered on top of local storage. Previously, deployments running on SNO could only use block storage.

This feature provide in-transit encryption for Object Storage between OpenShift Data Foundation and Red Hat Ceph Storage when using external mode. It enables all data to be encrypted in transit and at rest. For more information, see knowledgebase article on how to use trusted certificates.