March
New Features
PostgreSQL Tracing – Kerno now automatically traces PostgreSQL queries, giving you detailed performance metrics and full visibility into request and response payloads, including headers and bodies, with zero code changes.
GKE Support – You can now deploy Kerno to your Google Kubernetes Engine (GKE) cluster using a single Docker command.
AI-Powered Issue Investigation [Beta] – Kerno's AI agent can now assist with initial issue triage. When an issue is detected, it analyzes relevant signals and provides a summary along with suggested starting points for further investigation.
Kubernetes Alerts – Kerno now supports alerting for Kubernetes pod and node-level events. You can configure alerts for pod restarts and crashes, as well as for node conditions such as NotReady, MemoryPressure, and DiskPressure.
Jira Integration – You can now create Jira tickets directly from issues detected by Kerno. This allows you to link production issues with your existing issue tracking system for follow-up and resolution.
Filter and Search – You can now filter and search services and pods by namespace and name. This improves navigation and makes locating specific resources in large clusters easier.
Improvements
External Services on the Service Map – The service map now includes external services that your Kubernetes cluster communicates with, giving you a more complete view of service dependencies.
Pod Health Summary in Service and Node Views – Added pod-level health indicators to both the service and infrastructure views. This gives you a quick visual summary of how many pods are running, pending, failed, or succeeded across your services and nodes, making it easier to detect issues that might not yet impact top-level service metrics.
Pod Restart Events – Pod restarts are now tracked as events in the timeline. You can view individual restart occurrences and see the total restart count per pod.
Encrypted Traffic Detection for Ruby – Updated the agent to support detection of SSL-encrypted traffic in Ruby applications, improving visibility in secure environments.
DaemonSet Toleration Support – During installation, Kerno now detects node-level tolerations and automatically adds them to the DaemonSet configuration. This ensures components like Preon and Vector can run on nodes with custom scheduling constraints.
Preserve Issue Associations for Deleted Components – Kerno now maintains links between issues and services that have been deleted. If a service is removed from the cluster, issues previously associated with it are still displayed and marked as deleted, ensuring historical context is preserved.
Improved Service Navigation – The service preview now links directly to the full service detail view, allowing for a seamless transition when inspecting service-level data.
Toggle Deleted Pods and Nodes – You can now show or hide deleted pods and nodes in the UI. This helps reduce visual noise while preserving the option to inspect recently removed resources.
Faster Event Loading – Optimized event data retrieval to reduce load times in large clusters, improving responsiveness in dashboards and timelines.
Activity Tab UI Improvements – Updated the activity tab to better differentiate between system events and user actions, improving readability and traceability.
Consistent Ordering and Scroll Behavior – Standardized message ordering and scroll position for the chat and activity log views. The latest updates now load predictably at the bottom.
Improved Infrastructure Metrics Display – Refined how CPU and memory usage, requests, and limits are visualized to improve clarity and make resource utilization easier to interpret.
Faster Data Loading in High-Volume Clusters – Optimized front-end queries to improve load performance for clusters with large datasets.
Bugs
Clear Issue List on Cluster Switch – Fixed an issue where switching clusters would leave previously loaded issues visible, potentially causing confusion.
Fix Null in HTTP Traces – Resolved a backend issue where some HTTP trace entries were missing destination identifiers, restoring complete trace information.
Prevent Internal Traffic Duplication in Preon – Fixed a bug in Preon that caused internal service-to-service traffic to be counted multiple times.
Correct Issue Occurrence Counts – Fixed a bug that caused mismatches between reported and actual issue occurrence counts in the UI.
Fix Text Clipping in Overflow Tooltips – Ensured tooltip and popover content is fully visible, even for long strings or overflowing text.
Fix Infinite Scroll in Service Map – Resolved a UI issue that caused the service map to scroll endlessly in certain cases.
Fix Service Map Positioning and Loading – Addressed bugs that caused misalignment or incomplete loading of the service map in some environments.
Fix Paths Table Layout – Adjusted the layout of the paths table to prevent overlapping elements and ensure consistent display.
Display Accurate Load Error Messages – Implemented more specific error messaging when data fails to load, improving debugging and user feedback.
Fix Organization Creation Flow – Resolved an issue that prevented some users from creating new organizations due to missing or restricted permissions.
Fix Slack Issue Links – Updated Slack integration links so they open directly to the correct issue detail view instead of the general cluster overview.
Fix Intermittent Service Map Load Failures – Fixed a bug where the service map would occasionally get stuck during loading.
Last updated