New features in 4.19

Replication delay for RHACM applications

The application list page now shows the health status of Red Hat® Advanced Cluster Management (RHACM)-managed applications for improved monitoring of disaster recovery.

For more information, see Viewing health status of ApplicationSet-based and Subscription-based applications.

Additional disaster recovery recipe capabilities for CephFS-based applications

Disaster recovery recipes support additional applications, enabling automated disaster recovery for CephFS-based applications deployed using imperative models.

High availability option for MCG metadata database

MCG runs with high availability for metadata databases (DBs), eliminating single points of failure for MCG DB and reducing the risk of data loss during node failures.

Automatic scaling of storage for dynamic storage devices

Automatic capacity scaling can be enabled on clusters deployed using dynamic storage devices. When automatic scaling is enabled, additional raw capacity equivalent to the configured deployment size is automatically added to the cluster after the used capacity reaches 70%.

Prevention of unauthorized volume mode conversion

Volume mode conversion is prevented during the volume mode restore when the original volume mode of the persistent volume claim (PVC) for which the snapshot is taken does not match the volume mode of the newly created PVC that is created from the existing volume snapshot. This also helps to verify that conversions work properly.

Easy configuration of Ceph target size ratio

The target size ratio parameter can be set depending on the cluster usage and how the cluster is expected to fill among the three types of storage: block, shared filesystem, and object storage. The target size ratio is a relative value that influences the allocation of Ceph Placement Groups across storage pools. For more information, see Configuring Ceph target size ratios.

Reduce data transfer and improve performance using read affinity for RGW

In Local Storage deployments, the rados_replica_read_policy is set to localize for the RADOS Gateway (RGW) daemons. This helps to reduce the data transfer costs and improve performance by routing all the RGW read requests to the nearest OSD. For more information, see Performing localized reads in Administration Guide of Red Hat Caph Storage documentation.