Compliance: Manual synchronization of compliance data is now supported.
Filters: Updated default selection rules for the Company Proximity filter.
Editing Workflow: Introduced extra confirmation to ensure users can continue editing seamlessly.
Search:
Faster document and content loading across the platform, along with reduced web-worker requests to lower cloud costs..
Streamlined document processing for user changes and optimized search history.
Improved search reliability by refining analytics and reducing load on servers.
Simplified multi-section queries and optimized results prioritization.
AdminView Performance: Enhanced performance for AdminView company list and user metadata endpoints.
Expert Calls Access Restriction: Users without Expert Insights access can no longer view expert calls.
Infrastructure:
Enable precompiled NVIDIA driver for gpu-operator in Azure:
No outbound connection to archive.ubuntu.com needed
Gpu driver daemonset pods don’t need to compile NVIDIA kernel module at boot
Enable gpu slicing which will reduce gpu nodegroup count by half:
New labels needed for gpu nodegroups
Full gpu nodegroup rollout required to make gpu slicing effective
Observability:
Grafana: Add a comprehensive set of dashboards for Kubernetes observability
Grafana: Changed default Grafana user creation from admin to read-only privileges. Users can still view all Enterprise product dashboards and telemetry data, but with more appropriate access controls. Access for this user is updated here
Email notifications for notebook notes sharing is not working
Add to note button not available in Generative Search summary
Sentiment highlighting not working for prepared remarks & follow ups, speeches docs (GDS Docs)
Selected snippets are not navigating correctly within the document viewer.
Annotation viewer is shown as author, instead of real author of the annotation
Search doesn't return bookmarked docs or manually shared user docs with corresponding filters
Precompiled Nvidia driver is not enabled for gpu-operator in AWS and GCP, outbound connection to archive.ubuntu.com (download linux kernel headers) is required for AWS and GCP
The PROMETHEUS_TARGET_SCRAPE_DUPLICATE_TIMESTAMPS alert may transition to a Firing state as a result of metrics being received with identical timestamps from the same target. Please note that this is a benign issue and does not impact data collection or monitoring functionality.
The rms-bulk-activity-service may intermittently indicate liveness probe failures, even though the service is functioning as expected. This behavior is attributed to the current liveness probe configuration and does not reflect an actual service disruption.