IBM Fusion 2.10.1 Hotfix Notification

General Page

Fusion Hotfix Notification

Date: November 2025
Applies to: IBM Spectrum Fusion – Backup & Restore 2.10.1
Impact: High – Data protection and restore reliability

Abstract

Systems running Backup & Restore 2.10.1 may experience backup or restore failures, excessive pod restarts, or inconsistent job alerts due to defects in node-agent, data mover scheduling, and restore logic.
This hotfix provides critical stability and reliability improvements, addressing failures in Backup & Restore components and ensuring consistent protection of application data.

Affected Component

Backup & Restore service version 2.10.1

Recommended Action

For Online Environments

Apply the hotfix using the standard post-install patch script:

For Offline Environments

Mirror hotfix images to your local registry first using this script:

Then apply the post-install patch:

Issues Addressed

Backup Reliability

Fixed nil pointer exception in node-agent that caused pod crashes and backup failures.
Reduced data mover resource usage with improved loadAffinity and concurrency control.
Prevented SELinux relabeling failures on PVCs during backup.
Reduced frequency of node-agent restarts during backup and restore.

Restore Functionality

Fixed restore failures caused by old or previously failed jobs.
Corrected missing custom namespace labels during restore.
Allowed restores to alternate namespaces (altNamespace) when target namespace exists.
Resolved restore errors where application inventory info was not returned.
Restores from pre–2.9 backups now succeed on 2.10.x clusters.

System Stability & Alerts

Suppressed unintended restore failure alerts for stale jobs.
Prevented false “backup job completion” alerts for outdated jobs.
Stopped continuous PolicyAssignment CR updates from guardian-dp-operator.
Addressed Redis certificate expiration that caused job failures.
Reduced pod churn and improved scheduling stability in data transfer services.

Additional Improvements

Applied Kafka broker security patch.
Enhanced logging for backup error: "NoneType is not iterable".
Improved handling of long-running exechook operations (>1 hour).
Optimized job cancellation logic for CancelPending → Cancelled transitions.

Verification

After applying the hotfix:

Confirm all Backup & Restore pods are in Running state:
oc get pods -n ibm-spectrum-fusion-backuprestore
Run a test backup and restore to confirm functionality.
Verify the applied patch version:
oc get csv -n ibm-spectrum-fusion-backuprestore

[{"Type":"MASTER","Line of Business":{"code":"LOB69","label":"Storage TPS"},"Business Unit":{"code":"BU048","label":"IBM Software"},"Product":{"code":"SSFETU","label":"IBM Fusion"},"ARM Category":[{"code":"a8m3p0000000rXCAAY","label":"SW"}],"ARM Case Number":"","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"2.10.1"}]

Tips