What’s New in Kloudfuse

Improved query performance by automatically using rollup tables for time aggregations when the lookback period is greater than or equal to the rollup interval. This optimization reduces query latency for long-range queries without requiring any configuration changes.

Enabled series limit enforcement by default to protect system stability. Queries without space aggregation are limited to 5,000 series, while queries with aggregation are limited to 20,000 series (prior to aggregation). These limits are configurable via the kfuse chart values.yaml.

Fixed nested resource label caching for namespaces like AWS/ECS. Metrics for ECS services and tasks now correctly inherit and display parent resource labels, improving the accuracy of infrastructure monitoring dashboards.

Added cluster-level metrics scraping for Aurora DB clusters. This enhancement provides visibility into Aurora cluster-wide performance metrics in addition to individual instance metrics.

Added the compare operator for analyzing data across different time periods. Compare current metrics against historical baselines to identify trends, anomalies, and performance changes over time.

Enhanced the transpose operator to support multiple aggregate columns, enabling more complex data transformations in a single query.

The if operator now supports nested function calls, allowing more complex conditional logic in FuseQL queries.

Added support for group by operations on array-type columns, enabling aggregation across multi-valued attributes.

Improved autocomplete suggestions that respect previous filters and query context. The code editor now provides intelligent suggestions based on the current query structure, including function autocomplete for FuseQL and PromQL.

Fixed log table height issues and table cutoff at the bottom of the screen. Improved search bar styling and overall logs screen layout for better usability.

Added support for high cardinality facets in trace analysis with autocomplete, group by, and aggregation capabilities. This enables efficient exploration of trace attributes with many unique values without impacting query performance.

When editing notification templates, you can now load payload data directly from active alerts. This makes it easier to test and preview notification templates with real alert data without manually constructing test payloads.

New dedicated page for viewing active alert notifications. Monitor which notifications are currently being sent and track notification delivery status in real-time.

Added search functionality to the Contact Points page, making it easier to find specific contact points in large configurations.

Added label filter support in the Events panel on the Alert page. Filter alert events by specific label values to focus on relevant alert instances and reduce noise when investigating alert patterns.

This release contains improvements to dashboard performance, rendering, and visualization.

Added Real User Monitoring (RUM) tools to the MCP Server, enabling AI-assisted analysis of frontend performance, user sessions, and browser metrics.

Added APM execution breakdown tools to the MCP Server, providing AI-assisted analysis of service execution time and downstream dependency contributions.

Create Service Level Objectives using custom PromQL queries, not just pre-defined APM metrics. Define custom numerator and denominator queries for precise SLO calculations tailored to your specific requirements.

Promote custom attributes to facets in RUM for enhanced filtering and analytics. Create custom facets from usr and context attributes to organize and analyze your RUM data according to your application’s specific needs.

Provides AI agent access to Kloudfuse observability data through the Model Context Protocol (MCP) server. Enable AI-powered analysis, troubleshooting, and insights across your entire observability stack.

New History tab displays a comprehensive state timeline for each alert with dropdown filters to narrow history by specific label values. Inspect only the events you care about and track alert state changes over time.

Compare two alert revisions side-by-side and restore older versions. Track changes to alert definitions over time and easily revert to previous configurations when needed.

Each alert instance row clearly displays whether it’s silenced. Mute and unmute buttons use Grafana’s matcher logic to precisely target the exact labels you’re viewing, ensuring suppression rules apply only to intended instances.

Creating or editing alerts now prompts for evaluation duration ("for") and execution frequency ("every") with validation for positive time strings. The reducer automatically adjusts based on above/below threshold selection, and deleted annotations no longer reappear on save.

Alert tags in the alert list table now support expand/collapse functionality for better space utilization and readability.

Fixed APM alert editing performance issue that was causing hundreds of queries to be issued. Editing APM alerts is now significantly faster and more efficient.

Contact point creation now opens in a new tab (instead of modal) with a refresh button to update the contact point selector without losing other form inputs.

Enhanced authentication and session management with improvements to oauth2-proxy and user management service. Includes fixes for session handling and rate limiting enhancements.

Kloudfuse now supports ingesting logs from Heroku platform through syslog drains with automatic parsing, authentication, and metadata extraction.

Added support for sort_by_label, sort_by_label_desc, and double_exponential_smoothing functions with Prometheus 3.5.0 engine upgrade.

Added apdex function to supported aggregation functions for SLO calculations.

Migrated all external dependencies from Bitnami and third-party repositories to internal registries for FIPS-140-3 compliance and improved security posture.

global:
  fips:
    enabled: true

yamlCopied!

Kafka has been upgraded to version 4.1 with KRaft mode support, eliminating the dependency on ZooKeeper for metadata management.

Fixed critical issue with log split queries by migrating to KfLogSplit implementation.

Removed maximum columns and rows restrictions for lookup tables, allowing tables of any size.

Fixed CVE vulnerabilities across UI, query-service, and logs-parser components.

In FIPS mode, due to a known limitation in the AWS SDK, decoration for S3 metrics will not work.

Kloudfuse now supports Prometheus native histograms and OTLP exponential histograms. This enables more accurate representation of metric distributions with dynamic bucket boundaries and improved storage efficiency.

You can now import and export multiple alerts in bulk, streamlining migration and backup workflows.

Dashboards can now be imported and exported in bulk using zip files for easier backup and migration.

Alert lists now open a sidebar when viewing details instead of refreshing the entire page, improving navigation and workflow efficiency.

Added template variable support for trace and events analytics dashboards, enabling dynamic filtering and parameterization.

Lookup table creation workflow now supports empty CSV files with headers only and composite primary keys. Update operations properly handle duplicate entries.

Save query option is now available on all subtabs in FuseQL except fingerprints. Export option renamed to "Add to Dashboard" for clarity.

Added reserved field names for filtering on log message content and metadata in FuseQL queries. Users can now filter directly on log line content using kf_msg, log source using kf_source, and log level using __kf_level.

__kf_msg =~ "ERROR.*connection.*timeout" and __kf_level = "ERROR" and __kf_source = "nginx"

sqlCopied!

Added support for AWS CloudFront metrics enrichment and default metric aggregations. CloudFront metrics are now available with proper aggregation types as defined by AWS CloudWatch.

Added support for GCP Stackdriver (Google Cloud Monitoring) metrics ingestion and enrichment. Monitor Google Cloud resources with native metric collection and contextual enrichment.

Data scrubbing now supports Events and RUM data streams. You can preview, filter, and delete Events and RUM data for compliance and data retention requirements.

Fixed issues where stat panels returned "No Data" for valid queries and setting decimal values broke panel rendering.

Fixed inconsistent behavior when toggling between builder and advanced modes. Key exists operator and other filters now translate correctly between modes.

Improved timeseries rendering performance to reduce excessive re-rendering on user interactions.

Fixed scrubber minion stalling issue and improved overall stability.

UI now provides visual confirmation when lookup tables are successfully saved.

You can now import and export multiple alerts in one action, simplifying migration and backup workflows.

Added options to hide/show timeseries on default view.

Fixed UI bugs and made improvements to dashboard import/export workflows.

Fixed a bug where UI showed blank integration pages.

Fixed bugs on infrastructure UI that caused some information cells to be blank.

Fixed a bug on Logs dashboard in Control Plane.

The main navigation has moved from the top bar to a left-hand panel. This update provides more workspace for dashboards and a consistent, modern navigation experience. All modules — such as APM, Metrics, Logs, Events, RUM, Infrastructure, Dashboards, and Alerts — are now accessible from the left sidebar.

You can now import and export multiple dashboards in one action, simplifying migration and backup workflows.

Dashboards now support dragging and dropping row panels for easier layout customization.

Stats panels can now be edited in dashboards, improving flexibility.

Alerts can now be created directly from APM service details panels.

Suppress noisy alerts individually without disabling entire policies.

You can now choose chart styles in dashboards for more tailored visualization.

Create, edit, and delete notification policies directly in the UI.

FuseQL now supports the cat and save operators, enabling more flexible data workflows.

Query lookups are now more powerful with lookupContains.

Dashboards, alerts, and other Kloudfuse objects can now be organized into folders. This improves navigation, access control, and sharing in large environments.

Folders now support hierarchical nesting. You can mirror complex organizational structures and manage resources at multiple levels.

Folder detail pages now show which alerts and dashboards are contained in the folder.

The Hydration UI has been redesigned for clarity and usability. You can now track progress more easily and troubleshoot issues faster when replaying or recovering data.

Audit logs can now be ingested natively into Kloudfuse, simplifying compliance workflows and eliminating the need for external pipelines.

Kloudfuse now supports Kubernetes infrastructure monitoring via the OpenTelemetry (OTel) collector, in addition to Datadog.

We upgraded Grafana to version 12.0, ensuring compatibility and access to the latest visualization features.

Tables in Logs, APM, and Event Analytics now include a download option for exporting data.

We renamed Groups to Teams, removed the concept of Team Ownership (each Team has a default Admin account member), and streamlined role assignment within Teams.

We added service accounts to this release. They allow applications to securely interact with other systems or APIs in the background using bearer tokens. These workflows do not require human involvement, and automate workflows.

This new powerful approach enables users to safely delete data from Metrics, Logs, and APM streams. Create a scrub job by specifying the stream type, one or more filters (label, comparison operator, and value), and the relative time interval. By matching these criteria, Kloudfuse creates a scrub job.

Starting with this release, you can assign RBAC policies directly to both Users and Service Accounts. This is in addition to the existing process of Users inheriting policies through their Group/Team assignment.

On the left sidebar (Filters), we added more search functionality. Each attribute that has a large number of possible values has a search bar.

You can now install Kloudfuse across multiple availability zones (AZs) for high availability (HA). If one AZ fails, Kloudfuse continues operating without disruption. This is critical for large-scale and production-grade deployments.

The Infrastructure view now uses infinite scroll instead of traditional pagination. This improves navigation when working with large datasets.

We are introducing automated Group Memberships and Role Synchronization with OKTA (SAML or oauth 2.0) and Google (only SAML). We do not support Google with oauth 2.0; you must perform these configurations manually.

We added support for parsing arrays inside json log messages. See JSON Operator.

We are now supporting Kubernetes events through Open Telementry (OTel) integration. See OTel for Kubernetes — Events.

We made improvements to the sidebar, enhancing usability, adding new groupings for labels, and calling out the facets. See Events List, and Events Filters.

We improved the performance of charts by streaming the trace data, error report, and error group report. You no longer have to wait for charts that display long intervals. Kloudfuse cancels outstanding queries as you change query parameters.

We added the ability to download Error and Error Group reports.

The if filter operator can now evaluate math expressions. For example, if (C > 1000)`.

The where filter operator can now evaluate math expressions. For example, where (C+5 > 0).

Add single quotes* around the terms filter to process strings that contain symbols and spaces.

In a views table, shifts a column in a table downwards. See fuseql-miscellaneous-operators.adoc#backshift.

Removes duplicate results in a table. See fuseql-miscellaneous-operators.adoc#dedup.

When searching, you can now use the * (wildcard) to find a match within a logline.

We added a new detailed interface for editing and viewing Users, Groups, Folders, and Policies.

We improved the visualization of effective policies.

We added links for navigating between groups and users.

You can now access the Runbooks for Control Panel alerts by clicking the URL on the Alert Details page.

We upgraded the Embedded Grafana to Release 11.6.

We unified the information on consumption in the Consumption Tracking dashboard. You can remove the following dashboards, if present: Consumption Overview, Consumption Overview v2, and Consumption Overview (By Stream, Pie chart).

We refactored this dashboard by adding panels that show agent_type, log lines processed by the ingester for metrics, processed metrics, processed events, and so on.

We created this dashboard so you can see at a glance the information about configured alerts, how many alerts are triggered, how many alerts errored out, how many notifications were sent, and so on.

Expand to see the dashboard.

We changed the tab names for RBAC actions to Users, Folders, Groups, and Policies.

We removed this feature from the Admin section. Instead, we now associate the relevant policy configurations within the Policies tab, after selecting the relevant Groups. In the Policy Config Management section, click Add policy to assign the policy to that group.

Expand to see an example of Policy Config Management.

• Users can belong to multiple Groups.

• Groups can have multiple Policies.

• Policies can have multiple filters (scopes).

Before this release, filters for APM and Metrics streams were combined. You can now configure APM stream filters independently, and apply them correctly to the APM service list and service detail charts.

When combining multiple group policies for a user, Kloudfuse now defaults to a more permissive data access rules, a union (logical OR) of all possible policy filters.

Expand to see an example of combined effective policies.

We combine Filters within a Policy using AND logic. We combine Filters across Policies using OR logic, for each stream.

There are three distinct scenarios for combining policies:

We introduced the matches operator to match strings using the RE2-compliant regex format. See matches for more details.

We introduced the in operator to determine if string or number values are in the search domain. See in for more details.

We now support a single query search across multiple views of the same schema. In effect, we implement a logical OR across two or more scheduled views. See Query multiple views.

We improved loading times for dashboards with streaming queries. The dashboard render on the page much faster; on long term queries, you may notice that Kloudfuse paints the most recent data first, and then proceeds to paint the rest of the time interval in reverse chronological order.

We transformed panel rendering to ensure that all panel types render correctly and fully. We also handle queries that return empty results.

We improved the performance of charts by streaming the data; you no longer have to wait for charts that display long intervals. Kloudfuse cancels outstanding queries as you change query parameters.

We standardized the appearance of the Events list with the APM Traces table. It now has a columnar format, in place of a log-type entry. See Events List.

Starting with this release, we parsed the combined Event Message into the following component columns: Event Id, Aggregation Key, Level (status), Event Type, Title, and Message body. We added the Relative time column, and removed the Tags information.

When you click one of the Events in the list, Kloudfuse displays the event detail definition. It also includes searchable access to relevant facets and several label categories.

You can use the Event detail interface to filter on sources, facets, and labels. See Filter and Search Events.

The Filters navigation lists the Kubernetes Resources: Pods, Clusters, Namespaces, Nodes, Workloads, Network, Storage, and Access Control.

The left navigation also includes a Facet Search, and existing and tracked facets.

The main screen includes a Search and Group by affordances.

The table repords infrastructure information for Pods: their Status, Cluster, Namespace, Age, Readiness (how many of total), Restarts (number), % CPU Usage, and % Memory Usage.

Indicate how many result to show per page (10, 24, 20, or 75), and navigate between pages.

Click on one of the table rows to see the detailed information for that pod: Status, Cluster, Namespace, Node, Deployment, Replica Set, Service, Age, Readiness, Restarts, IP, QoS (quality of Service), Tags, Kubernetes Labels, the YAML definition, Logs, Metrics, and Events.

In the Kubernetes Dashboard section, find the following new entries:

Expand to view the Kubernetes Integration page

In the System Dashboard section, find the following new entries:

Expand to view the System Integration page

We added several bulk actions to the Alert Rules page.

The action formerly known as Mute Alert is now Suppress Alert.

You can create complex, multi-component schedules for suppressing alerts.

This release adds the Evaluate as zero option to account for missing data from alert query conditions.

Implements CRUD operations for applications, by name, type, ID, client token, and so on, through the UI instead of the config file custom-values.yaml.

When you click a panel or chart, it opens a new URL on the right side of the screen.

Overview URL panel

Starting with this release, you can create scheduled views, and query directly from that view.

We added backshift and dedup operators to support querying of Scheduled Views.

Create new dashboards based on results of Advanced Search across logs.

Create new alerts based on results of Advanced Search across logs.

Create and use lookup tables to supplement your data.

This release significantly expands the analytical features for Metrics by adding Top List, Table, and Pie Chart analytical views to Metrics Explorer.

• Top List

• Table

• Pie Chart

This view provides a quick summary of top N metrics that match the filter and time interval criteria. It helps with high-level analysis of metrics in your system.

We added an interpolation function, default_zero, to handle missing data in Metrics Time Series.

We added the existing deployment version markers to the graphical APM Trace List information.

To migrate existing dashboards into the Kloudfuse system, use the consolidated script approach. See Using Kloudfuse Customer Scripts > Dashboards.

To migrate existing alerts into the Kloudfuse system, use the consolidate script approach. See Using Kloudfuse Customer Scripts > Alerts.

We introduced an AZ-service that enables you to switch Kloudfuse to a different availability zone.

For the Service, Traces, Analytics, and Error pages, we restricted the number of displayed results to improve performance. Kloudfuse, by default, sorts labels and attributes in descending order, and displays up to 1,000 items.

Search now leverages regex for the contains operation.

We now support Kubernetes and Host metrics exported by the Otel collector when rendering the APM Infrastructure Dashboard. See System-Level Metrics.

We removed the Show Inactive Services option. Instead, the Services interface has two viewing options: the default Active Services, and All Services.

We added a Tree Visualization to track the values of the overall performance metric. Loading Time is the default metric; you can also choose Largest Contentful Paint, Cumulative Layout Shift, and Interaction to Next Paint. This chart reports the health of the metric across the system, color-coding for Good (green), Needs Improvement (amber), and Poor (red) statistical measurements, based on customer specification.

We added three more visualizations: Loading Time, First Contentful Paint, and User-Centric Page Load Times latency reports. The existing latency reports include Large Contentful Paint (LCP), Cumulative Layout Shift (CLS), and Interaction to Next Paint (INP).

We highlighted the URL addresses of the pages with the poorest performance in terms of Longest INP, Largest CLS, and Slowest LCP.

We restricted the number of displayed results based on Search, to improve performance. Kloudfuse, by default, sorts in descending order, and displays up to 1,000 items on the Log Side Bar Filter. This number is configurable.

We fixed an issue where the overall time range now correctly applies to table data in dashboard panels.

We added support for warning threshold, alert recovery, and warning recovery thresholds to threshold alerts on metrics, logs, and services. You can also link alerts to dashboards and panels, pause alert evaluation, and link back to the Kloudfuse alert page from alert notifications.

Create a Metric Threshold Alert

We added support for linking in log alerts, and display matching log lines in alert notifications.

Permissions now persist for downstream and upstream dependencies that cross cloud, Kubernetes cluster, and namespace boundaries. When you set policy filters using service id labels, Kloudfuse supports cross-boundary permissions, and applies RBAC policy filters to both active and inactive services.

We now support RBAC policies at the level of the data stream. While default RBAC policies apply to all non-RUM streams, you can pin custom RBAC policy filters to a specific stream, such as logs, metrics, or RUM.

Create a Stream-Specific Custom RBAC Policy

You can now create private folders, where only the creator has access to the folder contents.

Create Private Folders

We made changes to warn users with viewer roles about their permission limitations when they attempt to add, edit, and delete dashboards and alerts.

We improved the functionality of Event alerts.

We improved the handling of IPs recorded in RUM sessions.

We added a geographical location extension for RUM, and the ability to handle the user-provided IP:location mapping.

We made improvements to the SDK for Mobile RUM.

We improved the functionality of RUM alerts.

To better assist you in configuring your data streams into the Kloudfuse platform, we added an Integration section to our UI.

The Kloudfuse platform now ingests span and trace data sent by New Relic agents and SDK.

The Kloudfuse platform now ingests ephemeral and batch metrics data sent by Pushgateway.

Operates like a pipeline, progressively narrowing down results to help you find exactly what you need. Each operator, separated by a pipe (|), builds on the results of the previous one. This enables you to filter and focus your search with precision as you move through the pipeline.

FusesQL is part of both the Search Bar and the Query Builder for Logs Analytics: Time Series, Top List, Table, and Pie Chart pages of the Logs interface. Starting with this release, FuseQL is the default query language for logs.

In addition to toInt, we added two more type conversion functions: toLong and toString.

Click the Export icon, and proceed to name the new panel and save it in a new or existing dashboard.

Click the Alert icon, and create alerts based on the current analytical setting. This feature is part of all Analytical interfaces (Time Series, Top List, Table, and so on) for APM, Metrics, Logs, and Events.

Specify source maps to better decode stack traces.

We added UI changes to make Mobile RUM reporting better.

This interactive trace data visualization chart helps you to visually detect deviations and outliers in the latencies reported by APM data.

Our proprietary analysis tool that helps you to narrow down the cause of appliance performance issues; it compares thousands of attributes across span events to their baseline performance, and displays ranked results based on significance of deviation.

For the Trace > Log details interface, we improved the display of the JSON format of the log line.

When examining Trace Latency Breakdown detail, under the Logs option, we added a new filter option. Depending on the nature of your technical stack, you can now select to filter on Kubernetes (pod_name), Docker (container_id), or Host (hostname).

We added the following features to this release:

We added the following features to this realease:

We added the following features to this realease:

We now support for MSK(Kafka), Mongo Atlas dashboards.

We now support for adding custom tags for log groups in AWS and GCP.