v2.8.0 - 03/Mar/2026

Release Notes: SUSE® Observability Helm Chart v2.8.0

New Features & Enhancements

  • Helm 4 Support: The SUSE Observability Helm charts have been updated to support Helm 4.

  • Simplified Helm Installation: The installation process has been streamlined by deprecating the mandatory "extra values" generation step.

    • Backward Compatibility: While the original setup is still supported, it is now deprecated and support will be removed in a future release (at earliest after Q3 2026).

    • Migration: Users are encouraged to follow the Migration Guide to update their existing configurations to the new global mode.

  • Health History Visualization: The time-based health history chart now updates correctly when long time ranges are selected.

  • UI Enhancements: Improved the visual styling of bar charts for better data clarity.

  • Documentation Updates: The Ingress documentation now includes a comprehensive example for Traefik.

Bug Fixes

  • Timeseries Widget Settings: Re-organized the Y-axis settings within the timeseries widget for a more intuitive user experience.

  • Health Violation Navigation: Fixed a bug where related health violation links failed to open the correct component in the explorer.

  • UI Text Handling: Improved text truncation logic throughout the dashboard and application to prevent layout breaks.

  • StackPack Upgrade Performance: Resolved a timeout issue that occurred when upgrading StackPacks with a large number of instances.

  • Workload Observer Installation: Fixed a bug that caused the workload observer installation to fail in certain environments.

  • Backup CLI (sts-backup):

    • Corrected an issue where an incorrect storage class name was used during backup operations.

    • Resolved a lock contention issue during Kubernetes update operations.

    • Improved the readability of backup names when listing available backups via the CLI.

Agent Bug Fixes

  • HTTP Latency Noise Reduction: Added a known_high_latency label to HTTP metrics. This allows users to filter out connections pre-identified as inherently slow (such as the Rancher watch API), significantly reducing noise in latency-based alerts.

  • Splunk Integration:

    • Disabled saved searches in Splunk are now correctly ignored by the Splunk check.

    • Implemented a retry mechanism for HTTP 500 error responses received from the Splunk API to improve integration stability.