Dieser Inhalt ist in der von Ihnen ausgewählten Sprache nicht verfügbar.
Chapter 3. Enhancements
This section describes the major enhancements introduced in Red Hat OpenShift Data foundation 4.20.
3.1. Support forceful deployment of ODF Link kopierenLink in die Zwischenablage kopiert!
A new flag has been added to enable forceful deployment of the storage cluster. This helps redeployment with automation.
For more information, see the knowledgebase article Install Red Hat OpenShift Data Foundation 4.X in internal-attached mode using command line interface..
3.2. Automated key rotation and support for missing KMS in encrypted StorageClasses Link kopierenLink in die Zwischenablage kopiert!
With this release, annotations required for encrypted StorageClasses are now automatically added when Key Management Services (KMS) are missing. This streamlines key rotation and improves security configuration.
3.3. Pool level health status Link kopierenLink in die Zwischenablage kopiert!
Pool level alerts for near full and full status of the pool are now available with actionable messages.
For more information, see Resolving cluster alerts.
3.4. Multicloud Object Gateway Link kopierenLink in die Zwischenablage kopiert!
3.4.1. Unified CLI: mcg-cli capabilities integrated into odf-cli Link kopierenLink in die Zwischenablage kopiert!
Multicloud Object Gateway (MCG) commands are now available through the odf-cli utility. This enhancement consolidates ODF, Ceph, and MCG operations into a single command-line interface, eliminating the need to download and manage multiple binaries.
3.4.2. Public access limit option for S3 resources in MCG object browser Link kopierenLink in die Zwischenablage kopiert!
A new option is available in the MCG object browser to configure public access limits for S3 resources. This enhancement improves control over data exposure and strengthens security for object storage.
For more information about this procedure, see Setting up public access limit to S3 resources using MCG object browser.
3.4.3. Option to disable external access routes to MCG Link kopierenLink in die Zwischenablage kopiert!
A new configuration option to disable all routes that enable external access to the Multicloud Object Gateway (MCG). This feature helps ensure that MCG services are only accessible within the OpenShift environment.
For more information, see Securing Multicloud Object Gateway.
3.4.4. Bucket-Level metrics for replication state Link kopierenLink in die Zwischenablage kopiert!
New metrics have been introduced to provide detailed visibility into the replication progress of the buckets. These metrics help determine data safety and availability on the secondary site. The following metrics are now available per bucket, per replication cycle:
- Total number of objects scanned
- Number of objects successfully replicated
- Number of objects that failed to replicate
For more information, see Obtaining metrics to reflect bucket replication state.
3.4.5. MCG introduces Metrics and AlertRule for detecting failures relating to noobaa-db Link kopierenLink in die Zwischenablage kopiert!
There is a new alert to detect failures relating to noobaa-db. This helps to improve the health visibility of NooBaa DB.
This helps to identify internal failures when there is a disconnect between the components while a component by itself is in ready state.
For more information, see Resolving alerts and errors.
3.5. Alert triggered when CSI clones near soft limit Link kopierenLink in die Zwischenablage kopiert!
An alert is triggered when CSI clones or snapshots approach the soft limit of 200. This notification recommends switching to volume snapshot cloning for better performance, helping to avoid delays caused by excessive clone operations.
3.6. Improvements to disaster recovery uninstall workflow Link kopierenLink in die Zwischenablage kopiert!
This release introduces improvements to the DR uninstall workflow, streamlining the removal of resources created during various stages of DR deployment. Previously, uninstalling DR components required manual cleanup across multiple layers. With this enhancement, the uninstall process is more intuitive and automated.
3.7. Smarter resource merging for component configuration Link kopierenLink in die Zwischenablage kopiert!
Resource requirements for OpenShift Data Foundation components are now merged with default values instead of being fully replaced. If a user specifies only one resource type such as memory, OpenShift Data Foundation will automatically apply default values for the other type, like CPU. Previously, partial specifications caused missing fields to be dropped, leading to components running without complete resource settings and resulting in unpredictable performance. This enhancement ensures safer and more balanced configurations with minimal user input.
3.8. Selective merging of placement configurations Link kopierenLink in die Zwischenablage kopiert!
Placement configuration for components has been improved to support selective merging. Previously, specifying any placement section such as, Tolerations would override the entire default placement, leaving other sections like Node Affinity or Topology Spread Constraints empty. This caused incomplete or suboptimal placement configurations.
With this enhancement, OpenShift Data Foundation now merges user-defined placement values with the default configuration. Users can specify only the sections they want to customize, while defaults for other placement types are preserved, ensuring consistent and expected placement behavior for components.
3.9. Support custom SCCs for VolSync DataMover pods Link kopierenLink in die Zwischenablage kopiert!
VolSync DataMover pods could not access data when using custom Security Context Constraints (SCCs), leading to sync failures in clusters with custom configurations. The DRPC spec now includes a VolSyncSpec field that allows users to configure the following:
-
MoverSecurityContext: Defines the PodSecurityContext for DataMover pods -
MoverServiceAccount: Specifies a custom Kubernetes ServiceAccount for fine-grained RBAC control
These configurations are propagated to the VolumeReplicationGroup (VRG), ensuring VolSync components inherit the required privileges for successful operation.
3.10. Configurable memory and CPU for kube-rbac-proxy in ocs-metrics-exporter Link kopierenLink in die Zwischenablage kopiert!
Users can now configure memory and CPU for kube-rbac-proxy pods through the custom resource (CR). This enhancement addresses out-of-memory (OOM) issues encountered during ocs-metrics-exporter operations. By allowing resource adjustments, users can prevent running into these issues.