AI SRE release notes

The release notes describe recent changes to Harness AI SRE.

March 2026​

New Features and Enhancements​

• ServiceNow Native Actions: Runbooks now include native ServiceNow actions for creating incidents, updating incidents, and adding comments — without requiring custom webhook configuration. Fields are dynamically retrieved from your ServiceNow instance based on your connector. Configure a ServiceNow connector under Project Settings -> Connectors, then use the Create ServiceNow Incident, Update ServiceNow Incident, and Add ServiceNow Comment actions in your runbooks. (IR-1691)

• User-Defined Webhook Templates: You can now save webhook configurations as reusable templates and select them when creating new integrations. Configure a webhook once, click Save as Template, and reuse it across future webhooks. Templates appear in the Select Payload Template dropdown grouped under System Templates and Custom Templates. Templates are organization-scoped and use copy-on-write — changes do not propagate to existing webhooks. See Configure Webhooks for details. (IR-2540, IR-2541)

• Named Alert Rules: Alert rules now support custom display names, making it easier to identify and manage rules across your organization. Set a name when creating or editing an alert rule under Alerts > Alert Rules. (IR-2519)

• Active Pages View: On-call users can now view all currently active pages from the On-Call > Shifts sidebar. The active pages table shows page status, assigned responders, escalation progress, and acknowledgment state, giving on-call managers a single view of all in-progress pages. (IR-2515)

• Documentation Restructured for Admins and Responders: AI SRE documentation has been reorganized into separate paths for administrators and incident responders. Admins get a streamlined path through alert configuration, runbook automation, on-call schedules, and integrations. Responders get a new task-oriented Users section covering incident creation, triage, runbook execution, action items, and the mobile on-call experience.

February 2026​

New Features and Enhancements​

• Service Subscriptions and Status Updates: Users can now subscribe to services in the Service Directory to receive automated status updates during incidents. Navigate to On-Call > Service Directory, select a service, and use the Subscribers tab to add users or teams. Incident commanders can send status updates from the incident detail page with recipients auto-populated from impacted service subscribers. (IR-2291, IR-2095, IR-2096)

• On-Call Notification Rules (SMS, Push, Voice): On-call notification rules now support SMS, push notifications, and voice calls in addition to email and Slack. SMS and voice call support is currently available for US phone numbers, with additional regions enabled within 21 days of request. Push notifications via the mobile apps are available globally. Configure your contact methods under On-Call > Contact Settings. (IR-2513, IR-2411, IR-2410, IR-2520)

• Harness AI SRE Mobile App: Native mobile apps are now available on the App Store and Google Play. Responders can acknowledge, escalate, and resolve incidents and alerts directly from their phone, with critical alert override for Do Not Disturb mode. See the Mobile App Guide for setup instructions.

• Escalation Policy — Individual User Management: Escalation policies now support targeting individual users in addition to groups and schedules. Configure under On-Call > Escalation Policies. (IR-2512, IR-2395)

• Customizable Incident Severity and Alert Priority: Incident severity levels and alert priority mappings are now fully customizable to match your organization's standards. Configure under On-Call > Settings. (IR-2500, IR-2389)

• OpsGenie Migration Tooling: Import schedules, escalation policies, and user configurations directly from OpsGenie. Accessible under On-Call > Settings > Import. (IR-2553, IR-2380, IR-2348)

• PagerDuty Migration Tooling: Import schedules, escalation policies, and user configurations directly from PagerDuty. Accessible under On-Call > Settings > Import. (IR-2455, IR-2370, IR-2361)

• Jira Dynamic Fields Integration: The new Create Jira Ticket V2 and Update Jira Issue V2 runbook actions dynamically retrieve available fields based on the selected Jira project and issue type, replacing the previous static field configuration. Fields render with appropriate input types including multi-select for array fields. A manual key-value mode is available when the issue type is set dynamically using Mustache expressions. (IR-1669, IR-2313, IR-2454, IR-2341)

Fixed Issues​

• Fixed invalid Jira ticket URLs not being rejected, which caused downstream Slack integration failures. (IR-2390)
• Fixed custom field persistence issues after save in Jira runbook actions. (IR-2343)
• Fixed multi-select and labels fields rendering as single-select in Jira runbook actions. Array-type fields (labels, components, multi-select custom fields) now render correctly. (IR-2254, IR-2324)

AI Test Automation Release Notes

The release notes describe recent changes to Harness AI Test Automation.

March 2026​

2026.03.v1​

New Features​

• Task versioning
  Tasks now have full version history aligned with test versioning: each save of an edited task creates a new version instead of overwriting the only copy. Open the Version History tab on the task details page to review past versions, show edits between versions, copy a version to fork a new task, or restore an older version as current. Restoring affects every test that references that task; use copy when you need a separate line of work. Each test version remembers which task version was current when it was saved, so historical runs stay consistent while the latest test can follow the latest task. For details, see Tasks.

  

February 2026​

2026.02.v1​

New Features​

• Jira Integration
  Full Jira integration is now available, including a new Comments & Issues tab in the test run view. You can manage comments, link Jira tickets directly to test runs, and view failure reasons—all within the Harness UI. This streamlines issue tracking and improves collaboration between testing and development teams.

• Nested Tasks Support
  Added support for nested tasks, allowing validation tasks to contain sub-tasks. This enhancement includes schema changes (stepIndexes array in ValidationReport) and IR-level logic for inserting nested steps correctly, enabling more complex and modular test workflows.

• Link to Test Run in Jira Ticket
  Jira tickets created from test reports now include a direct link back to the associated test run. This provides quick navigation and traceability, making it easier to investigate failures and track issues across systems.

• Environment Name Filter in Test Run Page
  You can now filter test runs by environment name on the test run page. This makes it easier to narrow down results in multi-environment setups and quickly find the test runs you're looking for.

• Drag-and-Drop Command for Slider Interactions
  Implemented a drag-and-drop command to support slider interactions, enabling automation for slider UI components. This enhancement improves test coverage for applications with interactive slider elements.

• Parameter Support for Prompt Enhancer
  Added parameter support for the prompt enhancer and replaced the deprecated LLM model method with the default LLMModel. This improves flexibility and future-proofs the integration for enhanced AI-driven test generation.

Enhancements & Bug Fixes​

• Fixed Duplicate Copilot Task Display
  Resolved an issue where copilot tasks were appearing as duplicates in both the Intelligent Analysis (IA) view and the test run page. Users will now see a single, accurate entry per task.

• Renamed "Validate Test" to "Run Test"
  Updated the UI label from "Validate test" to "Run test" to better reflect the action being performed and reduce user confusion.

• Added Busy Tag and Renamed API
  Introduced a "Busy" tag to improve status visibility and renamed the associated API for consistency and clarity across the platform.

January 2026​

2026.01.v1​

New Features​

• AI-Powered Prompt Enhancement
  Get more from your test prompts! The system now automatically evaluates the quality of your natural language prompts for assertions, commands, and tasks. When ambiguity is detected, the AI suggests improved versions that are clearer and more precise, ensuring consistent and repeatable test results. This intelligent prompt optimization reduces variability across test runs, improves reliability, and helps you create more effective tests with less effort.

  Simply enter your prompt—if improvements are recommended, you can accept the suggestion or make further modifications before execution.

  To use this feature, enable enablePromptSuggestion in your settings.

  

• Updated Default LLM Model to GPT-5.2
  The default Large Language Model has been upgraded to GPT-5.2, providing improved AI-driven test generation and smarter element detection for more reliable test automation.

Enhancements & Bug Fixes​

• Optimized Test Suite Parallel Execution
  Test suites now utilize all available parallel workers simultaneously instead of running in batches. This improvement significantly reduces overall test execution time and maximizes resource utilization.

• Enhanced Calendar Date Range Selection
  The calendar agent has been improved to correctly handle date range selections without prematurely closing the date picker modal. This ensures more reliable testing of applications with date range functionality.

• Improved Session Storage Compatibility
  Variables stored in session storage are now JSON parsable, preventing errors when your application attempts to parse session storage values. This enhancement improves compatibility with applications that rely on structured session data.

• Streamlined Slack Notifications
  Slack notifications have been optimized to report only test failures instead of all test suite executions, reducing notification noise and helping teams focus on issues that require attention.

• Fixed Calendar Modal Interaction
  Resolved an issue where the calendar agent would incorrectly click the OK button and close the modal when selecting date ranges, ensuring proper date selection behavior.

December 2025​

2025.12.v1​

New Features​

• SVG Anchor Tag Support
  SVG anchor tags are now supported in test interactive authoring and test run replay. Images with SVG tags are now rendered in the session replayer during authoring and execution, improving visibility of icons and symbol references during test execution.

Enhancements and Bug Fixes​

• Test Suite Parallel Execution Control
  Parallel execution of a test suite on the same environment is now prevented to avoid conflicts. Test suites can still run in parallel across different environments, improving reliability and preventing race conditions.

November 2025​

2025.11.v2​

New Features​

• Task Duplication
  You can now duplicate tasks directly from the UI, making it easier to create similar test scenarios without starting from scratch.

• Configurable Timeout for Copilot Task Execution
  Timeout values for Copilot task execution can now be configured, giving you more control over AI-driven test authoring workflows.

• Grok LLM Support for Smart Selectors
  Grok is now available as an alternative LLM for Smart Selectors, providing more flexibility in AI model selection.

• LLM Model Configuration for Smart Selectors and AI Commands
  You can now configure which LLM model to use for smart selectors and AI commands, allowing you to optimise for performance, cost, or specific use cases.

Enhancements and Bug Fixes​

• Performance & Reliability Enhanced system stability with improved health monitoring and automatic scaling

2025.11.v1​

New Features​

• Configurable Timeout for Test Execution Commands
  Timeout values for test execution commands can now be configured. This allows teams to tune execution behaviour to match their environment and workload.

Enhancements and Bug Fixes​

• Ephemeral Storage Options for Test Executors
  Test executors now support configuration of ephemeral storage, helping improve reliability during storage-intensive test scenarios.

• End Epoch Timestamp on Test Run Completion
  A new timestamp is now captured when a test run finishes. This enables more precise reporting and easier time-based analysis.

• Progress JSON View Improvements
  The Progress JSON interface has been updated to provide a clearer and more consistent visual experience.

October 2025​

2025.10.v2​

New Features​

Intermediate Save in UI Introduced support for intermediate saving in the user interface. Users can now save progress while creating or updating tasks without completing the full workflow. This prevents data loss during long editing sessions and allows task progress to be resumed seamlessly. Navigation prompts have also been added to alert users when leaving unsaved work.

Micro Frontend Automation Support Enhanced the platform’s micro frontend structure to include automation routes. This update ensures automation-related pages load as part of the unified interface, improving modularity and overall responsiveness.

Enhancements and Bug Fixes​

• Faster Test Suite Run Page Loading Optimized backend queries powering the Test Suite Run page, reducing load times from several seconds to nearly instant, even for large test suites. This improvement significantly enhances usability during analysis and debugging.
• Improved Severity Filtering and CSV Reporting Updated report filtering logic to ensure accurate severity prioritization in test run reports. Added a new Test Run ID column to improve traceability and fixed issues with CSV exports for consistent data accuracy.
• Enhanced Test Suite Run Notifications Email and Slack notifications now include a final run status field, allowing users to easily determine whether a suite completed, failed, or was aborted.
• Fixed Step Creation Order in Multi-Step Tasks Resolved an issue where steps generated through AI input appeared in reverse order. Steps are now created in the correct sequence, ensuring consistency during test creation.
• Editable Visual Assert Regions Added the ability to edit assert screenshot regions in both Interactive and Quick Edit modes. Users can now adjust visual assert regions after execution, with clear prompts to prevent editing locked or completed regions.

2025.10.v1​

New Features​

Rerun - Failed Tests Functionality Introduced the ability to rerun failed tests directly from the Test Suite Run History page. This update features a dedicated "Run Failed Tests" dropdown and updated visual indicators consistent with the design system, allowing users to quickly spot and rerun failed executions for faster debugging and recovery.

Added support for rerunning specific tests within any test suite run. Users now have the flexibility to rerun only selected tests or isolate failed tests, making it easier to target issues and streamline the test validation process.

Grok (xAI) Integration Added native support for Grok (xAI) integration within the codebase, extending AI-driven insights .

Download CSV/JSON from Pipelines Added functionality to generate and download test result data in CSV or JSON formats directly from the pipeline.

Run Tasks in Import Test Case Feature The Import Test Case feature now supports executing predefined tasks as part of the import process. Users can include commands in the format Execute task <task_name> to trigger specific automated actions during import. This enhancement enables more dynamic and customizable test case imports.

Enhancements and Bug Fixes​

• Duplicate Test Suite Execution Prevention (AI-1531) An idempotency guard has been added to prevent duplicate test suite executions caused by task redelivery scenarios in Redis-based environments. This fix enhances reliability by ensuring that each suite runs only once per trigger.
• Cross-Origin Iframe Value Verification Fix (AI-1457) Improved the Expected Value Agent to accurately verify values inside cross-origin iframes by incorporating frame path resolution and automatic screenshot capture. This fix resolves validation issues during cross-origin test scenarios.
• Set Custom Headers from Parameters (AI-1530) Enhanced flexibility by allowing users to configure custom headers for test suites through parameter values defined at the app-suite scope. The system now follows a clear precedence chain (parameter → config → default) for header resolution.
• Store Specific Response Paths in API Call Command (AI-1538) Added support for defining a JSONPath within the API call command to capture and store a specific part of the API response. This improvement aligns with the API Interception feature and increases precision in response validation.
• Improved Parameter Field Handling in API Call Command (AI-1546) Enhanced the parameter field in the API call command to support direct selection from existing parameters via paramInputOptions, replacing the old manual naming flow. This update improves the configuration experience and prevents redundant parameter declarations.
• Default Checkboxes in Test Suite Run Page (AI-1545) Added a “Select All” checkbox capability in the Test Suite Run page for improved usability when managing large test suites. This update also replaces the previous custom checkbox logic with the native TableList checkbox implementation, resolving styling and selection behavior issues.
• Auto-Refresh for Custom Headers Parameter (AI-1563) Enhanced header management logic to automatically set or refresh custom HTTP headers whenever the CUSTOM_HTTP_HEADERS parameter is updated. This ensures header configurations remain synchronized without requiring manual refresh or login task completion.
• Added Enum and Intermediate Representation (IR) Support for Prompt Scoring (AI-1529) Introduced support for new Enum and Intermediate Representation (IR) commands to enhance prompt scoring capabilities. This update improves how user prompt understandability scores are handled and serialized across components, ensuring more consistent and testable scoring logic.
• Updated Enums, Mappings, and Helper Functions (AI-1556) Refined the internal command flow by updating Enums and mappings and introducing helper functions for FINALLY_BEGIN and FINALLY_END. These updates improve system stability and ensure better control flow handling during test execution and teardown processes.
• Fix for Prompt Scoring Enum Integration (AI-1574) Resolved a build-time issue where USER_PROMPT_UNDERSTANDABILITY_SCORE was missing from the Command Enum. The fix ensures the new Enum value is recognized correctly during compilation, stabilizing the build process and maintaining prompt scoring functionality.

September 2025​

2025.09.v3​

New Features​

Assert Wait for Question​

This powerful new assertion command lets users pause test execution until a specific user-defined question or condition is true. Built on the wait for text assertion command, it opens up dynamic, context-aware validation capabilities that make your tests smarter and more flexible than ever before.

To know more about Assert Wait for Question, refer to Assert Wait for Question.

Completely Overhauled Test Step Design​

The look and functionality of test steps have been fully transformed to enhance clarity and usability. Highlights include:

• The “Validations” tab renamed to “Step Insights,” with validation details shown directly and auto-validations surfaced much more intuitively.
• New icons communicate step status more effectively, including a retrain badge for steps needing AI retraining.
• Execution time and locator types are now visible for each step, offering deeper insights at a glance.
• The approval workflow is refined for multiple issues, with suggestion actions scoped per issue, making triage simpler and more granular.

Locator Configuration in IA/IE​

Take full control of your element selectors during Interactive Authoring and Interactive Editing! This new feature enables users to configure locators while editing steps, improving precision and reducing test maintenance headaches by ensuring the best possible target identification for elements.

Enhancements and Bug Fixes​

• Max Test Parallelism Error Display: Users now get instant feedback with an error when setting parallelism beyond allowed limits on suite details, enforcing correct configurations upfront.
• Reliable Parallel Task Handling: Errors in AI commands or fast tasks cause immediate task failure rather than just warnings, reducing confusion and improving error tracking.

2025.09.v2​

Enhancements and Bug Fixes​

• CLI Download for Test Results : Quickly download CSV and JSON files from the CLI, to get all the test results in a single file just by clicking the link available after the test it run in the python cli itself.
• Better Gzip Debugging : Troubleshooting compression-related issues is now easier with enhanced debugging support.
• Timezone Accuracy for Indonesia (WIB) : Fixed an issue where some timezone abbreviations were not recognized. Scheduling and reporting now correctly reflect local time in Indonesia, preventing errors.
• Improved Filter Visibility : Active filters now appear as chips, giving you a clear view of the criteria applied when exploring test data.
• Fail Tasks Immediately on AI Command or Fast Task Errors : Tasks now properly fail if AI Commands or Fast Tasks encounter errors. Previously, failures were only flagged as warnings, which could cause confusion.
• Aligned Date Selection : The start and end dates now default correctly and remain consistent in the interface, improving accuracy in reports and dashboards.
• Overseer Task Completion Fix : Overseer now completes tasks reliably, reducing delays caused by screenshot-based prioritization.
• Smarter Element Selection : Relicx-specific ID attributes are now ignored in smartselector, ensuring more reliable element detection and reducing false positives in task execution.

2025.09.v1​

New Features​

• API Response Interception: Added capability to intercept and analyze API responses during test execution for enhanced debugging and validation
• Pagination Enhancement: Added pagination options to display more than 20 items per page across test listings and results
• CSV/JSON Content Generation Control: Introduced configurable settings to control automatic generation of CSV and JSON content during test suite execution
• AI-Powered Parameter Generation: Enabled 'Generate with AI' functionality in parameter creation to support deterministic value generation for dates
• Test Case Import with Assertions: Added support for creating assertions and parameters during the 'Import Test Case' process

Enhancements​

• AI Thoughts Visibility: Enhanced AI transparency by showing AI thoughts during execution of If/elseIf commands and on assertion failures
• Download Directory Navigation: Added support for navigating to DOWNLOAD_DIR for improved file handling workflows
• Copilot Step Interactivity: Made copilot steps clickable during execution in Interactive Authoring mode
• Screenshot Retry Logic: Implemented automatic screenshot retry mechanism when confidence levels fall below retraining threshold

Bug Fixes​

• Parameter Handling: Fixed issues where empty string values were not being properly set in parameters
• Cursor Position: Resolved cursor position reset issue when entering values in input fields
• Live Edit Improvements: Fixed flickering and reload issues in Live Edit functionality
• Test Suite Navigation: Corrected redirection problems when navigating from test details page to view test suite
• User Input Validation: Prevented entry of invalid characters in parameter name input fields
• Login URL Display: Fixed incorrect display of LOGIN_URL and START_URL for navigate commands
• Test Export/Import Error Propagation: Improved error handling and email notifications for test export/import operations
• UI Styling: Resolved double scrollbar visibility issues in Harness integration interface
• Polling Optimization: Decreased polling intervals for All Test Listing Page to improve responsiveness

August 2025​

New Features​

• Binary File Upload: Added support for uploading files during test execution
• Sequential Tasks: Create multi-step workflows that execute in sequence
• LLM Test Naming: Tests now automatically generate descriptive names using AI
• Smart Popup Handling: Improved popup management - blocks unwanted popups while allowing user-triggered ones
• Data-Driven Testing: Upload JSON files via CLI to run tests with multiple data sets

Enhancements​

• AI Models: Added support for Claude 4 and GPT-5 for better test generation
• Test Import: Email notifications now sent when importing test suites
• Certificate Generation: Added UI tool for generating test certificates
• Performance: Improved screencast recording performance

Bug Fixes​

• Fixed test suite run history filters not working properly
• Resolved issues with automatic file upload detection
• Fixed date selection problems in Google Flights testing
• Corrected step ordering issues when importing test cases

July 2025​

New Features​

• Test Suite Export/Import: Export and import complete test suites with email notifications
• Database Testing: Added support for MongoDB and PostgreSQL database queries with secure tunnel connections
• AI Token Usage: View AI token consumption on test run pages with sorting capabilities
• Smart Date Selector: New AI command for intelligent date selection
• Environment Export: Export test environments along with tests

Enhancements​

• Shadow DOM Support: Tests can now interact with elements inside shadow DOM structures
• Parameter Management: Improved parameter creation and editing experience
• UI Updates: Updated branding from "CoPilot" to "Harness AI" throughout the interface
• Error Explanations: Added detailed explanations when AI tasks fail
• Date Handling: Better timezone support when working with dates

Bug Fixes​

• Fixed Slack notification channel selection issues
• Resolved UI styling problems with input fields and buttons
• Fixed replay functionality not showing latest changes
• Corrected AWS Batch test execution reliability issues
• Fixed table alignment and severity text display issues

May 2025​

Relicx Release Summary - Harness Platform Integration

Production Release: May 30, 2025

New Features​

Test Suite Configuration

• Added configurable test suite run timeout settings
• Implemented automatic saving of test schedule configurations when deleted

Visual Testing Enhancements

• Added bounding box highlighting for visual assertions
• Set interactive replay mode to false by default

Enhancements​

Test Execution Reliability

• Improved batch based exception handling during test suite execution
• Enhanced test run process management with proper status updates
• Fixed test suite abort handling for terminal jobs
• Updated admin job timeout status reporting to show "Timed Out" instead of "Executor Failed"

Performance & UI Improvements

• Increased memory and CPU allocation for test session flush operations
• Fixed test name truncation in test run pages
• Increased test runs table header height
• Fixed effective status calculation when test count equals zero

Platform Integration

• Updated configuration for Harness CI/CD release workflow compatibility
• Changed remote entry path for ATA module

Production Release: May 22, 2025

New Features​

Harness Platform Integration​

• AI Test Automation Module: AI Test Automation is now available as a native module within the Harness platform.
• Unified Access: Harness users can now access AI Test Automation directly from within the Harness platform interface.

• Pipeline Studio Integration: AI Test Automation can now be added to Harness CD Pipelines using a graphical no-code interface, enabling test suites to launch as part of Harness CD pipeline workflows.

AWS Batch Test Execution Infrastructure​

• Enhanced Scalability: New test suite execution infrastructure based on AWS Batch for improved performance and scalability
• Better Resource Management: Optimized resource allocation and retry mechanisms for more reliable test execution

Enhancements​

Test Suite Management​

• Complete Run History: Full test suite run history is now available in the test suite details → history tab, providing comprehensive visibility into past executions.

AI & Testing Improvements​

• Google Vertex Claude 3.7: Upgraded AI engine support for better test automation
• Enhanced AI Assertions: Improved AI assertion capabilities with better context handling
• Multi-language Support: Fixed non-English character handling in AI-generated screenshots

Quality Improvements​

• UI/UX Refinements: Enhanced modal displays, loading states, and navigation workflows
• Test Execution Stability: Improved assertion logic and validation handling
• Integration Reliability: Better error handling and recovery for platform connectivity
• Performance Optimization: Asset caching and network capture improvements
• Enhanced Slack Integration: Improved notifications with environment context and better URL formatting

This release marks a significant milestone, bringing AI Test Automation directly into the Harness platform ecosystem while delivering improved scalability and comprehensive test management capabilities.

All Release Notes Modules

Harness Release Notes

Subscribe to all via RSS

Explore recent changes to the Harness Platform, modules, delegates, and the Self-Managed Enterprise Edition.

Harness Platform release notes

Platform release notes

Delegate release notes

Code Repository release notes

Self-Managed Enterprise Edition release notes

Harness Solutions Factory release notes

Module release notes

AI Test Automation release notes

Artifact Registry release notes

Chaos Engineering release notes

Cloud Cost Management release notes

Continuous Delivery and GitOps release notes

Continuous Error Tracking release notes

Continuous Integration release notes

Database DevOps release notes

Feature Flags release notes

Feature Management & Experimentation release notes

Infrastructure as Code Management release notes

Internal Developer Portal release notes

Cloud Development Environments release notes

Security Testing Orchestration release notes

Service Reliability Management release notes

Supply Chain Security release notes

Software Engineering Insights release notes

SAST and SCA release notes

Security advisories

Harness Trust Center - Security advisories

Harness publishes security advisories for every release on the Harness Trust Center.

Artifact Registry Release Notes

The release notes describe recent changes to Harness Artifact Registry.

📌 Release Deployment Status by Cluster​

Progressive deployment: Harness deploys changes to Harness SaaS clusters on a progressive basis. This means that the features described in these release notes may not be immediately available in your cluster. To identify the cluster that hosts your account, go to your Account Overview page in Harness. In the new UI, go to Account Settings, Account Details, General, Account Details, and then Platform Service Versions.

March 2026​

2026.3.v1​

New Features​

Maven plugin for Artifact Registry

The Harness Maven plugin (io.harness.maven:harness-maven-plugin) lets you publish JARs, WARs, POMs, and related artifacts from your Maven build—no one-off scripts or manual uploads. It fits the standard Maven lifecycle, supports parallel-friendly deployments for multi-module projects, and can enforce dependency resolution through Harness upstream proxies so builds pull through the registries you govern.

• Deploy from Maven: Bind the deploy goal and push artifacts to Harness using the same coordinates and repositories your teams already use.
• Credentials via environment variables: Keep tokens out of pom.xml by configuring Harness registry URL and identity token through environment variables in CI and local workflows.

Learn more in the Build plugins overview; open the Maven Plugin tab there for installation and configuration.

Gradle plugin for Artifact Registry

The Harness Gradle plugin (io.harness.gradle) hooks into ./gradlew publish: Harness uploads artifacts to Artifact Registry in parallel for faster multi-module builds and reads registry URL and credentials from environment variables so secrets stay out of Gradle scripts and source control.

• Drop-in Gradle workflow: Apply the plugin in the root or subprojects and keep using your existing publish tasks.
• Built for CI: Matches how Gradle projects already inject registry configuration in pipelines.

Learn more in the Build plugins overview; open the Gradle Plugin tab there for installation and configuration.

February 2026​

2026.2.v1​

New Features​

Dependency Firewall

We're excited to ship Dependency Firewall in Harness Artifact Registry—a major step forward for software supply chain security. Until now, risky or non-compliant packages could flow into your organization through upstream proxies with little gatekeeping at the registry boundary. Dependency Firewall changes that: it evaluates every artifact version pulled from an external source before it is cached in your upstream proxy registry, using the same Policy as Code and OPA-style policies you already trust elsewhere in Harness.

• Policy at the front door: CVSS thresholds, license rules, package age, and custom Rego policies can allow, warn on, or block versions automatically—so violations are caught when dependencies are first fetched, not after they have spread across builds.
• Clear outcomes: Each evaluation is Passed, Warning, or Blocked. In Block mode, non-compliant versions are never cached and cannot be downloaded or used; Warn mode helps you roll out policies safely while you refine rules.
• Built for operators: Enable the firewall on your upstream proxy, attach policy sets, pick Block or Warn, and track everything from the Dependency Firewall tab—no separate toolchain required.

Learn more in the Dependency Firewall overview, enable Dependency Firewall in registry configuration, configure policies and policy sets, and the tutorial Implement Dependency Firewall with OPA policies.

Python registry: Poetry and uv

Harness Python registries now document first-class workflows for Poetry and uv—including publishing, installing, and authenticating with identity tokens—alongside existing pip instructions. Use the same pkg.harness.io endpoints and tokens as for pip; Poetry and uv integrate through explicit sources, pyproject.toml, and lockfiles your teams may already use.

Follow the poetry and uv tabs in the embedded guide on Get started with Artifact Registry (select Python in the format selector) for copy-ready commands.

Enhancements & Fixes​

Harness CLI: Dependency Firewall audit, explain, and npm client configuration

The Harness CLI (hc) streamlines Artifact Registry operations for security and local client setup:

• hc registry fw audit (alias hc registry firewall audit): Parse lock and manifest files and evaluate dependencies in bulk against Dependency Firewall policies. Supported inputs include NPM, Java (Maven and Gradle), and Python files such as requirements.txt, pyproject.toml, Pipfile.lock, and poetry.lock.
• hc registry fw explain: Return firewall scan status (Passed, BLOCKED, or WARN) and details for a specific package version already present in a registry.
• hc registry configure npm: Write Harness registry URLs and authentication into .npmrc for default, scoped, global, or project-level npm configuration.

Learn more in Manage artifacts and registries with the CLI: Audit dependencies from lock files, Get firewall status for an artifact version, and Configure npm client.

January 2026​

2026.1.v1​

New Features​

Artifact Download from UI

Harness Artifact Registry now supports downloading artifacts directly from the UI. You can download all versions of an artifact, specific versions, or individual files. The system prepares your download as a compressed archive and displays a status indicator at the bottom center of the page. Once ready, downloads remain available for 24 hours.

This feature works seamlessly with Docker digests and all supported artifact types, making it easy to retrieve artifacts for offline use, backup, or distribution.

Learn more about downloading artifacts from the UI.

Native CI Integration for Artifact Upload

We've introduced a new native Upload Artifact to Harness Artifact Registry step in Harness CI pipelines, making it easier than ever to publish build artifacts directly to Harness Artifact Registry without custom scripts or third-party plugins.

What's new:

• Built-in CI step: New "Upload Artifacts to Harness Artifact Registry" step available in all CI pipelines
• Multi-format (non-OCI) support: Upload artifacts in formats such as Maven JARs, npm packages, Python wheels, Conda packages, Generic artifacts, and more

This native integration streamlines your CI/CD workflows by eliminating the need for custom scripts and manual authentication setup. Simply add the step to your pipeline, configure your target registry, and let Harness handle the rest.

Learn more about the native CI integration for Artifact Registry.

Enhancements & Fixes​

Enhanced CLI Capabilities for Artifact Registry

The Harness CLI (hc) now includes expanded functionality for managing artifacts and registries:

• Metadata Management: Set, get, and delete custom metadata on registries, packages, and specific versions. Use metadata for tagging environments, tracking ownership, managing approval workflows, and maintaining compliance information.

• Artifact Copy: Copy specific versions of artifacts between registries within your Harness Artifact Registry, with support for artifact type specification (e.g., model, dataset).

• Artifact Version Delete: Delete specific versions of artifacts or all versions of an artifact. This provides granular control over artifact lifecycle management.

• Registry Delete: Remove entire registries from your projects through the CLI.

• Python and NuGet Support: Manage Python (PyPI) and NuGet packages directly from the command line.

These enhancements provide a consistent CLI experience across all supported registry types, making it easier for development teams to integrate Harness Artifact Registry into their existing workflows and automation pipelines.

Learn more about managing artifacts and registries with the CLI.

December 2025​

2025.12.v1​

New Features​

PHP Composer Registry Support

Harness Artifact Registry now supports PHP Composer packages, providing a secure, private registry for your PHP dependencies. You can store, manage, and distribute Composer packages directly within Harness with full compatibility with the Composer package manager.

Key benefits:

• Private package hosting: Host your proprietary PHP libraries and internal packages securely
• Upstream proxy support: Cache packages from Packagist and other public repositories to accelerate builds and reduce external dependencies
• Version management: Full support for semantic versioning and package constraints

To learn more about how to use Harness Artifact Registry with PHP Composer, check out our Composer Registry documentation.

Enhancements & Fixes​

Python PyPI Upstream Proxy Enhancements

Python PyPI upstream proxy configuration now supports specifying a custom registry suffix for non-standard PyPI endpoints. This allows platform teams to integrate private PyPI repositories and enterprise artifact managers such as Artifactory, Nexus, or self-hosted PyPI mirrors that do not expose packages under the default /simple/ path.

You can configure a remote registry URL, optionally define a custom registry suffix, and choose the appropriate authentication method. Harness transparently proxies Python packages from these upstream registries while handling authentication and package resolution.

November 2025​

2025.11.v2​

New Features​

Metadata Support for Artifacts and Registries

Now enhance your artifact management with custom metadata! You can now attach key-value pairs to registries, artifacts, and packages, enabling better organization, searchability, and governance across your artifact ecosystem.

Key capabilities:

• Multi-level metadata: Add metadata at registry, artifact, and package (version) levels
• Flexible filtering: Search and filter artifacts using custom metadata attributes
• Custom attributes: Track ownership, environment tags, build information, security classifications, and more
• Enhanced governance: Maintain audit trails and compliance information with version-specific metadata

To learn more about how to use Harness Artifact Registry to store and manage artifacts with custom metadata, check out our Metadata Support for Artifacts and Registries

Dart Registry Support

Harness Artifact Registry now supports Dart packages with full pub.dev compatibility. You can store, manage, and distribute Dart packages directly within Harness, with complete support for versions, metadata, and immutable release behaviour.

Key benefits:

• Secure, private Dart registry: Provides a dedicated, secure registry for all your teams' Dart packages
• Immutable versions: Ensures immutable package versions matching pub.dev behaviour for safe, predictable builds
• Accelerated CI/CD: Speeds up builds by caching remote dependencies via upstream proxy, reducing external dependencies and improving reliability

To learn more about how to use Harness Artifact Registry to store and manage Dart packages, check out our Dart Registry Quickstart guide.

Enhancements & Fixes​

Download Button for Non-OCI Artifacts

You can now directly download any non-OCI artifact (Maven, npm, PyPI, Generic, Conda, Helm charts, etc.) from the Artifact Registry UI with a single click. Until now, retrieving individual files or packages required using CLI commands or configuring a package manager client. For many workflows: debugging, validation, quick inspections, and offline analysis, users simply want to grab the file instantly.

2025.11.v1​

New Features​

Artifact Registry management via CLI

We're thrilled to introduce comprehensive CLI support for Artifact Registry management through the new Harness CLI v1.0.0 (hc)! This powerful addition brings the full capabilities of Artifact Registry directly to your terminal, enabling seamless automation and developer-friendly workflows.

What's new:

• Registry Management: List, view, and manage your registries with intuitive commands like hc registry list and hc registry get
• Artifact Operations: Push, pull, and list artifacts across all your registries using hc artifact commands
• Developer-Friendly Aliases: Save time with short commands - use hc reg instead of hc registry and hc art instead of hc artifact
• Flexible Output Formats: Get results in JSON, YAML, or table format for easy parsing in scripts and automation pipelines
• Cross-Project Support: Work seamlessly across multiple projects with global flags like --project and --org

Install the new Harness CLI v1.0.0 (hc) and authenticate to your account to start managing your registries and artifacts from the command line. Check out our CLI documentation for detailed examples and best practices.

Conda Registry Support

We have added a new registry type, Conda Registry support, for Python and R package management.

Key capabilities:

• Native Conda client support: Works with conda and mamba out of the box
• Bioconda upstream proxy: Automatically configured to fall back to Bioconda's public repository, giving you access to thousands of packages
• Hybrid package management: Host your private packages while proxying public ones from Bioconda
• Channel organization: Organize packages into channels for better version control and distribution

Configure your Conda client to point to your Harness registry, and you're ready to go - private packages are served directly while public packages are fetched from Bioconda automatically (If some custom source is not configured).

Do refer to Conda Registry Quickstart for more details.

Enhancements & Fixes​

Upstream Proxy to aggregate multiple Artifact Registries

We have enhanced our Artifact Registry experience by allowing it to be configured as an upstream proxy, enabling you to aggregate multiple registries into a single, unified access point. Use any Harness Artifact Registry as an upstream proxy for their respective registry.

When adding an Artifact Registry as an upstream proxy, ensure that registry doesn't have its own upstream proxies configured to avoid circular dependencies.

To know more about Set Proxy for Registry

October 2025​

2025.10.v1​

Enhancements and Fixes​

Public Registry

Users can now define the visibility of an artifact registry as Private or Public. This enhancement allows better control over access to registry contents and image pulls.

By default, all registries are created as Private.

• Public registries make the registry contents and images accessible to all users external to your organization.
• Private registries restrict both visibility and image pulls to authorized users or service accounts with valid permissions or tokens.

Check out our documentation to know more about Creating an Artifact Registry

September 2025​

2025.09.v2​

New Features​

Artifact Quarantine

Protect your software supply chain with Artifact Quarantine! You can now quarantine artifacts to prevent them from being used in pipelines or pulled by users. This powerful security feature works hand-in-hand with built-in container scanning and policy enforcement.

Key Capabilities:

• Manual Quarantine: Quarantine any artifact with a documented reason via the 3-dot menu
• Automated Quarantine: When integrated with Harness Supply Chain Security, artifacts are automatically scanned using AquaTrivy, and Security Tests policy sets can automatically quarantine artifacts based on vulnerability severity
• Easy Management: Remove artifacts from quarantine when they're safe to use again

This feature is available for Docker and Helm registries and provides an essential layer of protection to ensure only secure, compliant artifacts make it into your production environments.

Learn more: Artifact Quarantine

Digest Viewing / Image Referencing

Harness Artifact Registry now provides complete visibility into all your container images with digest-based viewing and flexible tag selection, giving you more control over how you reference and manage images.

Untagged Images Made Visible: Images without tags now appear clearly in the UI with an “N/A” label next to their digest. They remain fully pullable via their digest, so even untagged or cleaned-up images are easy to track and verify. Multi-architecture Docker and OCI images are also grouped neatly by platform, making navigation effortless.

Flexible Tag and Digest Selection: You can now select a tag or version for all artifact types directly from the header selector. For Docker and OCI images, you can also select by digest to reference an immutable version.

• Use Tags to browse familiar labels such as latest or 1.25.2.
• Use Digests to pinpoint a specific, unchanging image for verification or debugging.

Deployment details appear only when a tag is selected.

This enhancement offers a clearer, more dependable way to browse, reference, and inspect your images—whether tagged, untagged, or multi-architecture.

Learn more: Selecting by Tag | Image Referencing

Enhancements and Fixes​

NuGet Visual Studio Integration

We're excited to provide Visual Studio integration for NuGet package management! .NET developers can now configure Harness Artifact Registry as a package source directly within Visual Studio, enabling native IDE integration with secure token-based authentication.

Configure your registry in Visual Studio and start pulling packages from Harness registries with ease.

Learn more: Install and Use NuGet Packages

2025.09.v1​

Enhancements and Fixes​

• Delete Version API: Improved error handling for invalid non-OCI versions. Instead of returning a confusing 500 Internal Server Error, the API now responds with a clear 404 Not Found. This makes debugging easier and ensures a more consistent developer experience. [AH-1302]

• List Versions API: Fixed an issue where requests for unavailable images incorrectly triggered a 500 Internal Server Error. The API now returns a proper 404 Not Found, giving developers accurate feedback and reducing troubleshooting time. [AH-1829]

Chaos Engineering release notes

The release notes describe recent changes to Harness Chaos Engineering.

April 2026​

Version 1.83.4​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.83.0
• harness/chaos-ddcr-faults:1.83.0
• harness/chaos-log-watcher:1.83.0
• harness/service-discovery-collector:0.63.1

New Features and Enhancements​

• Added experiment run report to UI (CHAOS-11384)
• Added support to read from project, org and account scope variables in expression (CHAOS-11379)
• Removed "Show Logs" option for Linux/Windows and K8s v1 infra (CHAOS-11394)
• Added live logging and error handling for load execution in the backend (CHAOS-11322)
• Tabbed structure for probe and action details pages (CHAOS-11170, CHAOS-10611)

Fixed Issues​

• Fixed an issue where the total faults count in the experiment run report was not correct. Users will see the correct fault count now in their experiment run report (CHAOS-11446)
• Fixed log in service discovery (CHAOS-11439)

Version 1.82.1​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.82.0
• harness/chaos-ddcr-faults:1.82.0
• harness/chaos-log-watcher:1.82.0
• harness/service-discovery-collector:0.62.0

New Features and Enhancements​

• Added support for viewing Harness Delegate logs and DDCI logs directly in the logs view, with a new LogsView integrated into the timeline view for improved observability during experiment runs (CHAOS-11344, CHAOS-11343, CHAOS-11342)
• Added support for fault deprecation via Enterprise Hub Sync, allowing deprecated faults to be managed centrally through the enterprise ChaosHub (CHAOS-11324)
• Refactored report generation to a unified data extraction pipeline and added an API to return report data as JSON (CHAOS-11266)
• Added fallback mechanism to list Java PID for running JVM faults, improving reliability when the default discovery method is unavailable (CHAOS-11341)

Fixed Issues​

• Fixed an issue where creating a fault with an org-level fault template as reference was failing when the feature flag was enabled (CHAOS-11358)

Version 1.81.0​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.81.0
• harness/chaos-ddcr-faults:1.81.0
• harness/chaos-log-watcher:1.81.0
• harness/service-discovery-collector:0.61.0

New Features and Enhancements​

• Added enforcement of the load-enabled flag from installation to execution, ensuring load tests run only when explicitly enabled. Execution is now blocked with a clear error if load support is not configured, improving reliability and validation (CHAOS-11321)
• Added variables in the side nav items for experiments similar to pipeline variables (CHAOS-11280)
• Fixed node spanning issues in Chaos Studio Graph (CHAOS-11332)
• Added new infra_id index to chaosExperimentRuns and chaosExperiments to prevent execution context deadline exceeding (CHAOS-11313)
• Migrated execution plane components to Rapidfort (CHAOS-11340)

Fixed Issues​

• Fixed inputs not getting fetched when adding a custom fault under DR stage, where the API was failing (CHAOS-11272)

March 2026​

Version 1.80.3​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.80.0
• harness/chaos-ddcr-faults:1.80.0
• harness/chaos-log-watcher:1.80.0
• harness/service-discovery-collector:0.60.0

New Features and Enhancements​

• Added backend support for expressions in experiment and experiment template (CHAOS-11309)
• Added UI support for Splunk observability probe in chaos (CHAOS-11297)
• Merged CHAOS_LINUX_MIGRATION feature flag with CHAOS_NG_EXPERIENCE (CHAOS-11299, CHAOS-11298)
• Updated linux legacy version to 1.72.2 from 1.72.0 (CHAOS-11287)

Fixed Issues​

• Fixed subscription page not working by updating the authentication token to call the internal API to validate support user (CHAOS-11003)
• Fixed fault template create, update, and fault import as local copy for variable support (CHAOS-11282)

Version 1.79.4​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.79.1
• harness/chaos-ddcr-faults:1.79.0
• harness/chaos-log-watcher:1.79.0
• harness/service-discovery-collector:0.59.0

New Features and Enhancements​

• Introduced user defined variables and utility functions with CRUD support for faults, templates, probes, and actions (CHAOS-11222, CHAOS-11221, CHAOS-11220)
• Added the iterations runProperty support inside the chaos actions (CHAOS-11203, CHAOS-10557)
• Added load test integration with machine infra and machine-chaos IFS integration with load test manager (CHAOS-11154, CHAOS-11152)
• Migrated APIs from mux to gin (CHAOS-11127)
• Added tooltips for APM Probes (APM Type, Connector in Overview, Type, Comparison Criteria, Value in Probe Properties) (CHAOS-11120)
• Discovery Graph Enhancement: highlight connection of selected service (CHAOS-10695)

Fixed Issues​

• Fixed a UI issue where PM times were incorrectly displayed as AM when saving cron schedules (CHAOS-10951)

Version 1.78.1​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.78.0
• harness/chaos-ddcr-faults:1.78.0
• harness/chaos-log-watcher:1.78.0
• harness/service-discovery-collector:0.58.0

New Features and Enhancements​

• Updated chaos module banner to reflect module name change (CHAOS-11181)
• Added support for Secret, Connector, and <+input> field renderers in details for action and probes (CHAOS-11169)
• Added the native windows network chaos faults in machine chaos (CHAOS-11105)
• User can clone a load test using the three dot menu (CHAOS-11134)

Fixed Issues​

• Fixed update for variables in probes/actions/templates not working (CHAOS-11168)
• Fixed UI Infra Creation - Unable to use Environment at Org or Account scope for Chaos Infra (CHAOS-11164)
• Fixed Variables to Input migration issues (CHAOS-11155)

Version 1.77.3​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.77.0
• harness/chaos-ddcr-faults:1.77.0
• harness/chaos-log-watcher:1.77.0
• harness/service-discovery-collector:0.57.0

New Features and Enhancements​

• Migrated Action and Probe Variables to Inputs under Chaos Studio (CHAOS-11103, CHAOS-11102)
• Added egress rules support for ecs network restrict fault for specific rule (CHAOS-11101)
• Added linux chaos faults (network, API, JVM, process, service, DNS, disk fill) in machine chaos and its templates in hub (CHAOS-11081, CHAOS-11079, CHAOS-11078, CHAOS-11038)
• Updated ECS Fargate CPU/Memory Sidecar to Use Multi-Arch DDCR Image (CHAOS-10978)
• Added Disaster Recovery (DR) component support with new entity, APIs (DRTest run, getVariables, DRComponentNodes CRUD), and DDCR execution enhancements (CHAOS-10893, CHAOS-10887, CHAOS-10885, CHAOS-10884)
• Added permission to mitmdump in the install script (CHAOS-11125)

Fixed Issues​

• Fixes the issue with the load task stuck in the Pending state, and the infrastructure not receiving the load task request (CHAOS-11121)

February 2026​

Version 1.76.0​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.76.0
• harness/chaos-ddcr-faults:1.76.0
• harness/chaos-log-watcher:1.76.0
• harness/service-discovery-collector:0.56.0

New Features and Enhancements​

• Added live logging support for linux v2 and windows v2 infrastructure (CHAOS-10975)
• Added UI support for experiment templates for Windows and Linux infrastructure (CHAOS-10920)
• Added Resource Selector for probe, action, faults in chaos module (CHAOS-10900)
• Added new submodule routes behind feature flags in chaos web (CHAOS-11037)
• Added changes in linux infrastructure and infrastructure server to support load tests (CHAOS-11024)
• Updated go-billing package for flex licensing to fix memory leak (CHAOS-11017)

Fixed Issues​

• Fixed HSM secret mechanism in backend for SecretText case for faults - redis/vmware (password) (CHAOS-11026)
• Fixed ACL permission gaps and missing UI error handling across UI/API in Chaos module (CHAOS-11012)
• Implemented start/stop polling control on the onboarding status query. When the user reaches the "Create Application Maps" step, polling is automatically paused so the Network Map table remains stable for interaction (including opening menus and deleting maps). Polling resumes when the user navigates away from the step or advances to the next onboarding phase (CHAOS-9644)

Version 1.75.5​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.75.0
• harness/chaos-ddcr-faults:1.75.0
• harness/chaos-log-watcher:1.75.1
• harness/service-discovery-collector:0.55.0

New Features and Enhancements​

• Upgraded base image for Chaos components to RapidFort (CHAOS-11004)

Fixed Issues​

• Fixes AZ Blackhole target selection to cause chaos on all the derived subnets (CHAOS-10982)
• Fixed Experiment Inputs Not Visible in the Chaos Step (CHAOS-11001)

Version 1.74.1​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.74.1
• harness/chaos-ddcr-faults:1.74.0
• harness/chaos-log-watcher:1.74.0
• harness/service-discovery-collector:0.54.0

New Features and Enhancements​

• Updated Overview Page in chaos to incorporate Resilience Testing and YT videos (CHAOS-10770)
• Added Risk UI present in the project, org and account level scopes (CHAOS-10729)
• Added Resilience Risk backend and DB schema with the new db approach (CHAOS-10870)
• Moved application maps to chaos testing and added banner for simplified nav (CHAOS-10961)
• Added aks-node-down fault (CHAOS-10845)
• Added Experiment Timeline Builder with options menu by hovering over existing node (CHAOS-8897)
• Added initial setup for load test (CHAOS-10915)

January 2026​

Version 1.73.1​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.73.0
• harness/chaos-ddcr-faults:1.73.0
• harness/chaos-log-watcher:1.73.0
• harness/service-discovery-collector:0.53.0

New Features and Enhancements​

• Migrated to new DB interface for fault template and experiment template (CHAOS-10854)
• Added common db impl to remove duplicate code (CHAOS-10851)
• Created hierarchy_lookup table for chaos (CHAOS-10750)
• Updated chaos module license rollover period to 365 days in subscriptions doc page (CHAOS-10732)
• Migrated azure based faults from track1 sdk to track2 sdk version (CHAOS-10726)
• Added support for resource groups for chaos probes and actions in chaos web (CHAOS-10716)
• Added support for resource groups for chaos experiments in chaos web (CHAOS-10715)

Fixed Issues​

• Fixed container probe is erroring out incase of failed condition (CHAOS-10823)
• Fixed apm probe template creation (CHAOS-10819)
• Fixed helper annotation issue for helper daemonset pods (CHAOS-10812)
• Fixed RESILIENCE SCORE and RESILIENCE COVERAGE not updating in application maps with v1beta1 experiments. The fix introduces enhancements to the chaos experiment pipeline by adding logic to update the target network map and target services, thereby improving resiliency coverage for v1beta1 experiments (CHAOS-10844)
• Fixed tune fault functionality broken in beta1 experiments. All fault tunables are working correctly (CHAOS-10809)
• Fixed JVM experiment not respecting JAVA_HOME setting set on the spec. Fixed JVM chaos experiments to support custom JavaHome paths - experiments now succeed when java is not in the system PATH but JavaHome is specified in the experiment configuration (CHAOS-10616)
• Fixed API not sending updated details in response when updating metadata of a fault template (CHAOS-9958)

Version 1.72.7​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.72.0
• harness/chaos-ddcr-faults:1.72.0
• harness/chaos-log-watcher:1.72.0
• harness/service-discovery-collector:0.52.0

New Features and Enhancements​

• Added the windows command probe support to v1beta1 experiment type (CHAOS-10789)
• Added runtime input support for metrics queries in APM probes (CHAOS-10733)
• Added support for chaos template in step template creation (CHAOS-10754, CHAOS-10753)
• Added the harness as event source for the datadog annotation event action (CHAOS-10677)
• Added the HSM secret support to environment variables of Probe and Action (CHAOS-10667)
• Added Whitelist ssh filter for the aws-ec2-network-chaos (CHAOS-10566)

Fixed Issues​

• Fixed Node Network faults not affecting the network connectivity of the nodes, it was only affecting the helper pod where the tc command is executed (CHAOS-10652)
• Fixed issue with Prometheus APM Probe not adding TLS Configuration during experiment execution (CHAOS-10720)
• Fixed Chaos step checks api returning 500 error when pipeline stage has parallel steps or string type res score (CHAOS-10721)
• Fixed New Relic Connector ID derivation while execution (CHAOS-10735)
• Fixed issue with template preview for org and account scoped template in project scope template (CHAOS-10785)
• Fixed the raise condition of datadog metric calculation where Datadog Probe succeeds when it shouldn't (CHAOS-8960)

Version 1.71.3​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.71.0
• harness/chaos-ddcr-faults:1.71.0
• harness/chaos-log-watcher:1.71.0
• harness/service-discovery-collector:0.51.0

New Features and Enhancements​

• Added support for download of experiment manifest via get APIs instead of list APIs (CHAOS-10666)
• Removed usage of token and other keys from localStorage for CHAOS mfe (CHAOS-10540)
• Upgraded crictl in ddcr-faults image to fix CVE-GO-2025-3749 (CHAOS-10510)
• Added support for Container Action & Container Probes allowing users to get complete flexibility to provide all pod specific configurations for the required operations in a cluster (CHAOS-10453, CHAOS-10452, CHAOS-10451)
• Improved handling of pod statuses during container probe execution. The logic now correctly checks for various pod states (e.g., running, succeeded, failed) and provides detailed error messages when failures occur (CHAOS-10676)

Fixed Issues​

• Fixed probe details drawer breaking when probe is having huge details in Resilience Tab under pipelines (CHAOS-10678)

December 2025​

Version 1.70.4​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.70.0
• harness/chaos-ddcr-faults:1.70.1
• harness/chaos-log-watcher:1.70.0
• harness/service-discovery-collector:0.50.0

New Features and Enhancements​

• Added Kafka exception and latency faults in JVM faults (CHAOS-10637)
• Added changes to Chaos Experiment Selection Drawer in Pipeline to support chaos templates (CHAOS-10617)
• Added support for importing experiment as reference (CHAOS-10526, CHAOS-10485, CHAOS-10484)
• Added new statuses for Chaos manager notification triggers (CHAOS-10438)
• Removed the "Enabled" text on the probe resource in Probes page. The probes status on the side of each row has been removed, you can find it beside the name as separate column (CHAOS-9765)

Fixed Issues​

• Fixed issue with the target application validation for application chaos (CHAOS-10609)
• Fixed issue where there's no resilience score, but in the graph it shows 100, and resilience score shows -1% (CHAOS-10598)
• Fixed "Please select chaos infra" warning showing even after we have selected the infra (CHAOS-10590)
• Fixed issue when clicking on any radio button while configuring onboarding screen, it always scrolls screen to top by adding contain: layout to .bp3-form-group (CHAOS-9448)

Version 1.69.6​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.69.0
• harness/chaos-ddcr-faults:1.69.0
• harness/chaos-log-watcher:1.69.0
• harness/service-discovery-collector:0.49.0

New Features and Enhancements​

• Added Datadog annotation action support for custom actions (CHAOS-10563)
• Added bulk ignore support for chaos recommendations (CHAOS-10555)
• Added Grafana Events support in custom actions (CHAOS-10550)
• Implemented bulk operation functionality for chaos recommendations and resource tables (CHAOS-10548, CHAOS-10547)
• Added tooltips to various categories of custom services (CHAOS-10545)
• Addressed vulnerabilities in Linux source-probe (CHAOS-10346)
• Added support for importing faults as reference (CHAOS-10226,CHAOS-9890)

Fixed Issues​

• Fixed Type filter not working on get fault Templates API when trying to import a fault template (CHAOS-10568)
• Fixed experiments getting stuck in Chaos-Manager in 1.68.x release. The change modifies the name parameter from being extracted as a path parameter to a query parameter in the ExperimentExecutionNodeDetails function. This adjustment is intended to fix issues with DDCI calls, ensuring that the parameters are correctly processed in the API handler (CHAOS-10544)
• Fixed actions and probe update when no variables are sent. This fix addresses a bug related to the handling of nil cases in action variables within the UpdateAction function. The previous implementation included a conditional check for nil variables, which has been removed to ensure that the variables are always appended to the updatedAction regardless of their state (CHAOS-10537)
• Fixed DDCI manifest-generation for older k8s version (CHAOS-10572)
• Updated count of chaos recommendations to only reflect pending recommendations (CHAOS-10558)
• Fixed issue where during CD onboarding, Infrastructure View Progress was not opening in new tab (CHAOS-10559)

November 2025​

Version 1.68.5​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.68.1
• harness/chaos-ddcr-faults:1.68.2
• harness/chaos-log-watcher:1.68.0
• harness/service-discovery-collector:0.48.1

New Features and Enhancements​

• Added comprehensive infrastructure management UI with new wizards for creating Kubernetes and Machine chaos infras (Container, Linux, Windows, CloudFoundry), list pages with dedicated tabs for each infra type, and edit/detail views with modern designs (CHAOS-10154, CHAOS-10155, CHAOS-10156, CHAOS-10157, CHAOS-10158, CHAOS-10159)
• Added Probe verification support for Linux and Windows infrastructure (CHAOS-10162, CHAOS-10173, CHAOS-10186)
• Added support for advanced config when onboarding pipeline from CD Resilience Step (CHAOS-10393)
• Added support for k8s based services including node, pod, and workload, along with 'Other' Service option in custom services agent (CHAOS-10309, CHAOS-10380)
• Added filter and sort functionality for both k8s v1 and v2 apis with support for multiple environmentIDs filter (CHAOS-10201, CHAOS-10184)
• Added skeleton loader to timeline view for better loading experience (CHAOS-10231)
• Added ability to use templates from Experiment Builder Page (CHAOS-9970)

Fixed Issues​

• Fixed Chaos Guard not working as expected on Safari browsers where configuring freeze windows and time windows would fail. Time window and freeze window configurations now work correctly on Safari with proper date and time selections saved and displayed across all browsers (CHAOS-10322)
• Fixed ChaosGuard UI issues including tags not being visible on hover (now shown in popover), UI breaking while selecting conditions in rules (improved pagination), and YAML view coming blank before saving (CHAOS-10316, CHAOS-10315, CHAOS-10312, CHAOS-10310)
• Fixed Ask AI not working for questions about windows/linux chaos infrastructures by adding relevant examples in ChaosGuard AI (CHAOS-10314)
• Fixed label issue for Harness infra on Chaos-guard-conditions form page (CHAOS-10503)
• Fixed issue where REST APIs were returning data even when providing -1 as page/limit value by adding validation for page/limit parameters in all REST APIs (CHAOS-10325)
• Fixed issue where deleting a fault from its details page would take users to a blank page (CHAOS-10437)
• Fixed issue where setting only few of tunables under Authentication Tab when creating experiment with a fault would store others as undefined (CHAOS-10196)
• Fixed issue where creating experiment template at org/account level with authentication variables would fail with missing identifier errors (CHAOS-10174)
• Fixed Force parameter being interpreted as runtime argument (CHAOS-10059)
• Fixed import enterprise fault not working properly (CHAOS-9748)

October 2025​

Version 1.67.0​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.67.0
• harness/chaos-ddcr-faults:1.67.1
• harness/chaos-log-watcher:1.67.0
• harness/service-discovery-collector:0.47.0

New Features and Enhancements​

• Introduced file-based secret authentication for GCP VM Service Kill using service account JSON itself (CHAOS-9839)
• Added support for ECS agent stop chaos fault for Kubernetes HD (CHAOS-9885)
• Added support for ECS task stop chaos fault for Kubernetes HD (CHAOS-9883)
• Enhanced chaos-agent with proxy support for installing dependencies, GLIBC_2.32 compatibility for EC2 SSM chaos, and fallback to app.harness.io for binary installation (CHAOS-10007, CHAOS-10006, CHAOS-9972, CHAOS-10020)
• Added support for advance configuration for Chaos onboarding in CD pipeline (CHAOS-10043)
• Added backend validation for the secrets.getValue representation for the APM Probes (CHAOS-9961)
• Added sorting support to listInfrastructure APIs (CHAOS-10093)
• Added backward compatibility support for both v1alpha1 & v1beta1 in windows infra (CHAOS-9674)
• Updated Pipeline pages to refer to Chaos Agent (CHAOS-9980)
• Omitted deleted probes from experiment run details under Resilience Tests Tab (CHAOS-9877)
• Enhanced parsing logic for faultSpec to avoid sending null data (CHAOS-10079)
• Added backend validation to restrict the experiment creation when no fault defined inside the experiment (CHAOS-9962)
• Added support for re-onboarding of chaos experiment in pipeline (CHAOS-9966)
• Disabled GraphQL introspection (CHAOS-10101)
• Added AIEnabled toggle option in advance setting of onboarding (CHAOS-10102)
• Updated command source probe to use the fault-service account (CHAOS-10103)
• Added MTLS support to the API chaos faults (CHAOS-10107)
• Added Pipeline Run support for windows v1beta1 experiments (CHAOS-10200)
• Removed the image, imagePullPolicy and other unrelated fields from windows faults spec (CHAOS-10176)
• Added support for GCP monitoring APM Probe (CHAOS-9955, CHAOS-10090)
• Moved Chaos Guard to Project Settings with better user experience and optimized REST API support (CHAOS-10148, CHAOS-10147, CHAOS-10146, CHAOS-10145, CHAOS-10141, CHAOS-9862, CHAOS-9871, CHAOS-9870)

Fixed Issues​

• Fixed Command Probe issue where marking an environment variable as runtime input would set its runtime variable name as "VALUE" instead of the key/name of the environment variable. This prevented users from setting multiple runtime environment variables inside the chaos studio. The key/name of the environment variable is now correctly used as the name of the runtime input variable (CHAOS-10097)
• Fixed Probe Executions not showing executions for K8s V1 experiments run. The logic now appends recent runs from the v1 experiment list to the existing recent runs, ensuring a comprehensive view of the latest probe activities (CHAOS-9963)
• Fixed Experiment Count In Application Map. The logic for managing the experiment count in application map during deletion has been refined to ensure accurate tracking and decrementing of associated resources (CHAOS-9916)
• Fixed issue where providing volumes/mounts/labels/envs/initContainers from onboarding screen on overview page was not taking effect. Added the new fields in API (CHAOS-9449)
• Fixed Chaos Guard condition not showing probe verification value (CHAOS-10016)
• Fixed internal watcher filter source for the probes inside recommendations page (CHAOS-10024)
• Fixed plain_text/encrypted toggle option to show only for Kubernetes HTTP Probe (CHAOS-10065)
• Fixed issue where entity details would disappear from existing experiment runs after an experiment was updated and certain components were removed. Experiment runs will now continue to show all relevant details even if the experiment configuration changes later (CHAOS-10078)
• Fixed crash in chaos-web when no fault details are available. In cases where a fault is completely templatized or is a custom fault with no tunables, the details tab inside the timeline view would crash. An appropriate error/acknowledged message is now shown (CHAOS-10077)

September 2025​

Version 1.66.8​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.66.1
• harness/chaos-ddcr-faults:1.66.2
• harness/chaos-log-watcher:1.66.0
• harness/service-discovery-collector:0.46.0

New Features and Enhancements​

• Added support for cgroup v2 in AWS ECS container chaos with backward compatibility with cgroup v1 enabling chaos experiments on modern container runtimes that use cgroup v2 while ensuring smooth operation across environments still using cgroup v1 (CHAOS-9924)
• Added VMware Windows disk stress chaos fault in k8s-v2 using DDCR (CHAOS-9166)
• Enhanced error details to be highlighted and shown by default (CHAOS-9401)
• Added hotfix with the ECS Container chaos fixes (CHAOS-9965)
• Updated ChaosResiliencyCard in pipeline to use new resilience score color and logic (CHAOS-9956)
• Migrated cypress pipeline to use k8s connector instead of harness cloud (CHAOS-9950)
• Added variables in probe and action template manifest (CHAOS-9921)
• Added capability to run with readOnlyRootFilesystem: true for chaos-web (CHAOS-9856)
• Added functionality to run vmware stress faults with non-sudo user (CHAOS-9848)
• Updated tooltip for Probe and Fault Auth fields for HSM support (CHAOS-9847)
• Moved Username and Host to the Auth section inside the SSH fault (CHAOS-9844)
• Enhanced error handling in ddcr gracefully when faults execution failed (CHAOS-9834)
• Added authEnabled field inside the faultRef of experiment (CHAOS-9810)
• Added Authentication tab inside the faults based on FaultAuthentication category (CHAOS-9723)
• Added HSM support to the faults (CHAOS-9599)
• Added Select pipeline modal in chaos module with decent filters (CHAOS-9514)
• Added UI changes for NewRelic Probe support (CHAOS-9122)

Fixed Issues​

• Fixed Secret Support/Authentication Tab not coming when adding Fault Templates under Template Studio by calling the submit function directly and adding logic to catch clearing of secret input fields when they remove the value from formik (CHAOS-9941)
• Fixed Redis REDIS_TLS_SECRET not getting saved by calling the submit function directly and adding logic to catch clearing of secret input fields when they remove the value from formik (CHAOS-9940)
• Fixed Node Restart where none of username/secret was getting saved by calling the submit function directly and adding logic to catch clearing of secret input fields when they remove the value from formik (CHAOS-9939)
• Fixed SSH Faults Issues by calling the submit function directly and adding logic to catch clearing of secret input fields when they remove the value from formik (CHAOS-9938)
• Fixed Azure Secret not getting saved by calling the submit function directly and adding logic to catch clearing of secret input fields when they remove the value from formik (CHAOS-9937)
• Fixed VMWare Fault Secret/Authentication Issues by calling the submit function directly and adding logic to catch clearing of secret input fields when they remove the value from formik (CHAOS-9936)
• Fixed AWS Secret as well ARN for AWS both not getting saved as part for Auth Tab by calling the submit function directly and adding logic to catch clearing of secret input fields when they remove the value from formik (CHAOS-9935)
• Fixed Fault Authentication tooltip showing content for tunables by adding tooltip for fault auth (CHAOS-9933)
• Fixed SLO Probe not being removed when NG Exp feature flag is enabled where SLO probe will not be available when CHAOS_NG_EXPERIENCE flag is enabled but editing of pre-created slo probes will be supported (CHAOS-9932)
• Fixed Import Probe as reference for OOTB probe not working by adjusting the logic for determining whether a probe or action is imported, ensuring that the correct references are used throughout the codebase (CHAOS-9901)
• Fixed being able to run experiments that have inputs as runtime when clicking on run button present in 3 dot buttons in exp table page by properly handling runtime variable collection and implementing consistent execution flow across all UI components (CHAOS-9900)
• Fixed not being able to abort experiment when it is in queued state (CHAOS-9899)
• Fixed cleanupPolicy being empty in manifest when launching an experiment from an experiment template by adding logic to add cleanupPolicy while creating experiment from experiment template (CHAOS-9785)
• Fixed none of the filters working on fault template drawer when creating experiment template (CHAOS-9769)
• Fixed being able to save an experiment with non-existing infrastructure ID in windows experiment manifest by adding all infrastructure existence validation from the windows manifest (CHAOS-9543)
• Fixed removing a probe which was added multiple times in a windows experiment removing all instances of probe from fault during editing or cloning where only the intended probe is now deleted (CHAOS-9542)
• Fixed pipeline recommendations for CD services discovered in different project (CHAOS-9893)
• Fixed fault_forge and nsexec links in litmus-go (CHAOS-10001)
• Fixed high vulnerabilities in ddcr-faults docker image (CHAOS-9881)
• Fixed vulnerabilities in ddcr (CHAOS-9880)
• Fixed Experiment status based on the fault, probe and action statuses (CHAOS-9670)
• Fixed experiment resilience score for erroneous experiments (CHAOS-9806)

Version 1.65.12​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.65.0
• harness/chaos-log-watcher:1.65.0
• harness/service-discovery-collector:0.45.0
• harness/chaos-ddcr-faults:1.65.0

New Features and Enhancements​

• Enhanced fault template list page to have correct data in project settings by correcting the endpoint API calls from hubs/faults to /faulttemplates (CHAOS-9872)
• Added field in list fault templates API to return the count of fault templates in all categories (CHAOS-9858)
• Removed "Kubernetes" option from Fault template creation modal as Fault template is only supported for "Kubernetes (Harness Infrastructure)" (CHAOS-9850)
• Enhanced local APIs to use open API client for ng-manager and cg-manager (CHAOS-9842)
• Added permissions and permission links for faults which can be viewed on detail page (CHAOS-9837)
• Reverted changes to rename chaosActionTemplates package name (CHAOS-9752)
• Added Experiment Template frontend with CRUD and launch experiment functionality with intuitive UX (CHAOS-9735)
• Added API to support launch experiment feature (CHAOS-9727)
• Updated action template and probe template APIs (CHAOS-9718)
• Added launch experiment button to create an experiment from a template (CHAOS-9717)
• Added support for Dashboard in unified view for Chaos Module (CHAOS-9713)
• Added statusCheckTimeout support for v1beta1 templates (CHAOS-9705)
• Added logic to remove secret volume mount on GCP vm-stop (by id/label) faults if CLOUD_SECRET_NAME env is absent (CHAOS-9689)
• Added CHAOS_CMD_PROBE_SOURCE_MODE_ENABLED flag in harness-core-ui and hce-saas (CHAOS-9676)
• Replaced the existing resilience tab with the newly implemented resilience tests tab (CHAOS-9662)
• Fixed update probe API call in frontend to support probe and actions templates (CHAOS-9627)
• Added UI changes for delete chaos-hub where users can delete custom chaos hubs with no children (probes, faults, action) (CHAOS-9610)
• Updated infras on updating a chaos fault (CHAOS-9596)
• Added default destination ports in API chaos faults (CHAOS-9585)
• Created fault, probe and action drawers for experiment template (CHAOS-9581)
• Created Experiment template yaml service and schema (CHAOS-9580)
• Added VMWare powerOff by Name Fault (CHAOS-9579)
• Added missing RBAC permissions for node-drain faults (CHAOS-9574)
• Added control for Linux source probe support with feature flag (CHAOS-9573)
• Added pipeline recommendations behind feature flag and handled recommendations on pipeline deletions (CHAOS-9572)
• Standardized empty states for tables, tabular and card lists for chaos (CHAOS-9558)
• Added validation that if target input is available, then at least one of "kind", "namespace", "name", "label" should be provided (CHAOS-9489)
• Added experiment template schema (CHAOS-9435)
• Added support for Merk POV Execution and Enablement (CHAOS-9422)
• Added logwatcher streaming support for daemonset helper (CHAOS-9394)
• Refactored Chaos Studio to centralize all state for template support with major data layer refactoring, greatly reducing bugs/usability issues (CHAOS-9391)
• Added Experiment template list page on chaos hub in all scopes with details drawer (CHAOS-9381)
• Enhanced AZ Blackhole fault to remove ACLs for specific subnet in a Zone (within the VPC) using SUBNET_IDS or SUBNET_TAG as filter (CHAOS-9351)
• Added Pipeline to automate packaging of linux offline installer (CHAOS-9282)
• Added VMWare Power Off Experiment In Chaos V2 with VMWare PowerOff chaos in DDCR (CHAOS-9260)
• Enhanced API chaos faults to print the logs of the mitmproxy (CHAOS-9256)
• Enhanced DDCR to handle nil pointer exceptions (CHAOS-9212)
• Added Probe Certification feature to avoid gamification of Experiment Resilience Score (CHAOS-9097)
• Added auto-population of resources in ChaosGuard Conditions Wizard (CHAOS-9096)
• Added delete experiment template functionality with delete button (CHAOS-8847)
• Added delete experiment template API (CHAOS-8846)
• Added update experiment template UI with edit flow (CHAOS-8845)
• Added List and get Experiment template and revision UI (CHAOS-8841)
• Added Create Experiment template UI (CHAOS-8839)
• Added Create Experiment template API (CHAOS-8838)
• Added DB schema and index for experiment template (CHAOS-8675)
• Removed Image registry feature flag from Chaos-manager (CHAOS-8600)
• Added UI Tasks for HSM support for Chaos (CHAOS-8476)

Fixed Issues​

• Fixed incorrect experiment runs count and faultTemplate count (CHAOS-9571)
• Fixed GetExperiment returning null due to missing manifest apiVersion check (CHAOS-9695)
• Fixed CVE-2025-22868 security vulnerability in chaos-manager (CHAOS-9820)
• Fixed probe environment variables name and value not being properly displayed under probe details section in UI after import (CHAOS-9761)
• Fixed scoping issue for launch experiment API (CHAOS-9742)
• Fixed API call for ApplicationDetails for Target App Tab not being called for template studio (CHAOS-9845)
• Fixed get and list API for chaos experiment where infra id is empty (CHAOS-9721)
• Fixed AI recommendation banner having padding issue when Experiment_list feature flag is enabled where left-right padding will be present for recommendation banner on new list page (CHAOS-9807)
• Fixed authentication validation error when trying to create http probe template without auth where authType cannot be empty (CHAOS-9805)
• Fixed non-cron option not being selected by default in schedule tab when editing an existing non-cron v1beta1 or v1alpha1 experiment (CHAOS-9798)
• Fixed descriptions for fault templates not coming as formatted where bullet points are now properly respected (CHAOS-9797)
• Fixed authentication not being a required field while configuring http probe template (CHAOS-9791)
• Fixed Target App details Tab not coming up when adding a fault template while configuring an experiment template (CHAOS-9790)
• Fixed project dropdown showing projects from other orgs when launching an experiment from org level chaoshub (CHAOS-9789)
• Fixed chaos infrastructure selection input being enabled before project is selected (CHAOS-9788)
• Fixed "Launch Experiment" label to "Create Experiment" for better clarity (CHAOS-9787)
• Fixed Advance Options not being available for k8sv2 when editing an existing experiment (CHAOS-9786)
• Fixed infrastructure required field error even though infrastructure is already selected when launching experiment (CHAOS-9784)
• Fixed unknown action templates being listed even though there are no action templates in selected chaoshub (CHAOS-9783)
• Fixed unknown probe templates being listed even though there are no probe templates in selected chaoshub (CHAOS-9782)
• Fixed incorrect count of fault/experiment templates showing by default as 3 & 2 even though it's an empty chaoshub (CHAOS-9781)
• Fixed Infrastructure Type missing on experiment template (CHAOS-9778)
• Fixed error for child entities when trying to update an existing experiment template (CHAOS-9777)
• Fixed option to add experiment template to Enterprise-Chaoshub being available inappropriately (CHAOS-9775)
• Fixed enterprise probes showing custom probes instead of OOTB probes when trying to select probe templates (CHAOS-9774)
• Fixed Empty Experiment Templates Tab showing text wrt Fault Templates (CHAOS-9773)
• Fixed Probe Template Type filter not working under probe template selection drawer (CHAOS-9771)
• Fixed Action Template Type filter not working in Action Template selection drawer (CHAOS-9770)
• Fixed options to add probes/faults/actions showing texts wrt resource and not templates (CHAOS-9768)
• Fixed Template Builder showing options wrt "add a probe/action/fault" & not "add a probe/action/fault template" (CHAOS-9767)
• Fixed Overview Tab under Experiment Template construction page showing "Version" input inappropriately (CHAOS-9766)
• Fixed GetProbesInExperiment API giving 500 error when experiment manifest is not found (CHAOS-9684)
• Fixed "Open template in New tab" button of referenced probe redirecting to probe template name instead of identity (CHAOS-9641)
• Fixed Linux IFC restarting in Prod0 under edge case where task definition would delete while event orchestration is still underway (CHAOS-9638)
• Fixed experiment description box having misaligned text upon hover (CHAOS-9631)
• Fixed experiment showing Queued status in execution view though history indicates stopped OR timed-out (CHAOS-9630)
• Fixed unable to configure properties as part of edit operation on actions (CHAOS-9629)
• Fixed wrong import icon used on Import Button in Probes & Action (CHAOS-9568)
• Fixed error and blank screen when trying to edit an experiment after deleting chaos infrastructure (CHAOS-9548)
• Fixed Fault Type filter not clearing on press of remove button on ImportFaultDrawer (CHAOS-9547)
• Fixed experiment manifest of windows experiment containing unused labels (CHAOS-9545)
• Fixed experiment manifest for linux containing unused label "type: standalone_workflow" (CHAOS-9544)
• Fixed probes status/data not coming correctly under execution view for windows experiments (CHAOS-9541)
• Fixed custom script action giving error "delay action inputs is nil" when running (CHAOS-9505)
• Fixed input validation on duration input for probes not working/available (CHAOS-9494)
• Fixed Target Label or Target Name Runtime Input not updating to Fixed Value with empty string (CHAOS-9484)
• Fixed descriptions for faults not coming as formatted where bullet points are now respected (CHAOS-9481)
• Fixed Infrastructure Type for default probe created via onboarding showing "k8sV2" & execution history not coming (CHAOS-9459)
• Fixed "Chaos Infrastructure disabled successfully" message when removing/deleting an application map (CHAOS-9458)
• Fixed configuration labels referring to all as "chaos infrastructure" instead of "discovery agent" for discovery agent (CHAOS-9450)
• Fixed Infrastructures screen showing "ACCESS_TYPE" keys as empty values for Chaos V2 (CHAOS-9445)
• Fixed headers not being visible on probes details screen for http probes (CHAOS-9442)
• Fixed custom action having envs showing visual representation as "[object, object]" (CHAOS-9439)
• Fixed type dropdown not showing APM probe type when importing probes (CHAOS-9438)
• Fixed placeholders for metricsSelector/EntitiySelector under APM probe created with Dynatrace connector showing message wrt prometheus (CHAOS-9436)
• Fixed http-probe template (k8s) headers not being visible in UI mode but available in YAML (CHAOS-9432)
• Fixed Edit template button getting mis-aligned when action name is too long (CHAOS-9431)
• Fixed Env for custom script action not being visible in UI mode but available in YAML (CHAOS-9430)
• Fixed Cross button getting mis-aligned when there is an error in form while creating action template (CHAOS-9429)
• Fixed Import/Export Chaoshub being disabled on ChaosHub (CHAOS-9428)
• Fixed custom fault not showing run count and execution history even after running experiment (CHAOS-9418)
• Fixed missing icons in kubernetes hd card which was misleading (CHAOS-9400)
• Fixed updating a non-existent action not resulting in an error (CHAOS-9341)
• Fixed deleting a non-existing action not resulting in an error (CHAOS-9340)

August 2025​

Version 1.64.14​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.64.1
• harness/chaos-log-watcher:1.64.1
• harness/service-discovery-collector:0.44.0
• harness/chaos-ddcr-faults:1.64.3

New Features and Enhancements​

• Added in-house logic for windows memory stress and removed Testlimit dependency for better performance and reliability (CHAOS-9597)
• Added resource access restrict support to 1.64 with cherry-pick changes (CHAOS-9577)
• Added UI changes to migrate recommendation into separate tab for better organization (CHAOS-9567)
• Added support for outbound traffic in resource access restrict functionality (CHAOS-9566)
• Enhanced probes to execute in parallel mode even if duration is provided (CHAOS-9424)
• Added logic to derive autopilot enabled or disabled from helper pod ownerRef (CHAOS-9395)
• Added packaging of all dependencies and binaries into a single installable archive for online and offline installation (CHAOS-9365)
• Added auto pilot support for chaos infra and onboarding from frontend (CHAOS-9310)
• Added ability to clear inputs for a fault once provided in Chaos Studio (CHAOS-9003)
• Added changes for ddcr-faults-dual-binary support (CHAOS-9389)
• Updated frontend nginx configuration (CHAOS-9360)
• Upgraded AWS V2 SDK to support the V2 AWS faults with IRSA (CHAOS-9348)
• Added Resilience tab fixes for v1beta1 experiments (CHAOS-9333)
• Added hce-sdk changes to add environment variables in custom action scripts (CHAOS-9330)
• Added dual (pod and daemonset) mode support in ddcr-faults (CHAOS-9315)
• Enhanced mongo queries and added indexes for chaosFaults collections for better performance (CHAOS-9313)
• Added permissions and links in chaosFaults and chaosFaultTemplates collections (CHAOS-9309)
• Added support for generation of AI recommendations for the v1beta1 schema (CHAOS-9308)
• Changed icon of AI related items from co-pilot to standard AI icon (CHAOS-9265)
• Made the duration input for probes as FIXED VALUE until runtime support is implemented (CHAOS-9232)
• Updated CHAOS_ECS_ENABLED flag to the new name CHAOS_CONTAINER_ENABLED in the CHAOS_MANAGER (CHAOS-9197)
• Added support to run experiment with fault resource including v1alpha1 (CHAOS-9193)
• Added infraType to path params for getMachineInfraDetails API and added infra name to the deleteMachineInfra API response (CHAOS-9187)
• Added support for providing environment variables in custom script action in UI (CHAOS-9153)
• Added support for generation of v1beta1 experiment manifests via Onboarding to support timeline view in execution (CHAOS-9098)
• Enhanced chaos experiment DB schema to include reference of used faults (CHAOS-8905)

Fixed Issues​

• Fixed AI recommendation banner having padding issue when Experiment_list feature flag is enabled where left-right padding will be present for recommendation banner on new list page (CHAOS-9591)
• Fixed source filter not working in recommendations API (CHAOS-9587)
• Fixed execution details being blank on blocked chaos experiment (CHAOS-9370)
• Fixed audit event for Run recommendations API (CHAOS-9491)
• Fixed status under executionData of V1Beta1 experiments remaining always Queued by enhancing the execution data management for version v1beta1 and updating the status within the execution data structure (CHAOS-9529)
• Fixed openSSL dependency for wkhtmltopdf binary in chaos-manager (CHAOS-9327)
• Fixed chaos v2 experiment configured as cron returning wrong workflowType as "experiment_v2" instead of cron by correcting the handling of cron scheduling types during experiment updates (CHAOS-9528)
• Fixed probe search not listing probes imported as reference by fixing the probe name filtering logic to ensure it correctly matches the intended probe names (CHAOS-9520)
• Fixed update conflicts when adding multiple entities (fault/probes/actions) in parallel in an experiment by implementing a new RetryTransaction function to encapsulate MongoDB transactions with automatic retries for write conflicts (CHAOS-9507)
• Fixed step duration showing end time even though the step is still in progress where the shown value remained constant (same as start time) while running (CHAOS-9506)
• Fixed not being able to scroll YAML editor of Spec under Timeline View (CHAOS-9504)
• Fixed Target Application Details being visible for faults which don't have target details in Timeline View (CHAOS-9503)
• Fixed activity usage not getting updated for v1beta1 related services/experiments by adding logic to capture target services for beta1 experiments (CHAOS-9460)
• Fixed cmd probes not importing environment variables for linux and windows type of http probe (CHAOS-9443)
• Fixed probe or action imported as reference returning empty manifest API by adding manifest details while fetching data from template reference (CHAOS-9441)
• Fixed probe/action import not getting description from the referred template by adding description from template during import if not provided by user (CHAOS-9440)
• Fixed apiTokenSecretName coming as empty string for linux dynatrace probe under YAML (CHAOS-9437)
• Fixed not being able to scroll to env values while configuring k8s cmd probe template when source is enabled (CHAOS-9433)
• Fixed error when creating a probe template under a chaoshub having same identity as other probe in different chaoshub by updating the create operation to consider hub reference (CHAOS-9426)
• Fixed error when creating an action having same identity as any other action but in a different chaoshub by updating the create operation to consider hub reference (CHAOS-9425)
• Fixed fault selection drawer and modal for custom fault creation using "agent" terminology by adding agent type name in the fault Selection subtitle for clarity (CHAOS-9417)
• Fixed input validations not working on fault template creation modal where empty array/array elements will not be allowed from the UI (CHAOS-9397)
• Fixed fault template under chaoshub keeping loading and giving "internal server error" by defaulting api revision to v1 (CHAOS-9396)
• Fixed ProbeTemplates template creation succeeding with non-existent hub ref by adding validateActionTemplateCreation and validateProbeTemplateCreation methods to ensure that the specified hub exists and that the template identity is unique (CHAOS-9386)
• Fixed cron scheduling for experiments not working where multiple experiments targeting same network map was causing write conflict errors by adding a retry mechanism (CHAOS-9296)
• Fixed empty args being allowed during action creation (CHAOS-9185)
• Fixed not being able to run VMware experiments by passing vCenter credentials as secret (CHAOS-9174)
• Fixed "Completed with probe Error" status not coming when a probe failed due to genuine reason by updating status logic for execution data (CHAOS-9173)
• Fixed HTTP Probe summary text indicating success but actual probe execution reflecting error (connection refused) by adding http "connection refused" as a failure instead of an error for probes (CHAOS-9169)
• Fixed creating a probe or action with a duplicate name resulting in "Internal Server Error" by updating error messages to provide clearer feedback instead of generic internal server error messages (CHAOS-9060)
• Fixed creating an action with an existing id or name failing with "Internal Server Error" by updating error messages to provide clearer feedback instead of generic internal server error messages (CHAOS-9059)
• Fixed not being able to delete tag filters from Probe list page filters where probe filter tags can now be deleted (CHAOS-9056)
• Fixed SLO probe evaluation start and end time not getting saved by fixing the issue with evaluation window not showing up in SLO Probe (CHAOS-9055)

July 2025​

Version 1.63.7​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.63.3
• harness/chaos-log-watcher:1.63.0
• harness/service-discovery-collector:0.43.1
• harness/chaos-ddcr-faults:1.63.3

New Features and Enhancements​

• Enhanced and simplified image registry support for experiment v1 and v1alpha1 images (CHAOS-9052)
• Added support for "Complete with probe error" status on timeline view screen (CHAOS-9207)
• Added recent execution details for both probes and actions in the database schema, allowing for better tracking and management of execution history (CHAOS-8953)
• Added UI/UX support for windows command probe using windows infrastructure (CHAOS-9255)
• Added support for "INTERRUPTED" status for probes in probes screen as well as timeline view screen (CHAOS-9208)
• Enhanced experiment result display to show "CompletedWithProbeFailure" when probe failed (CHAOS-9205)
• Added support for import probe as a reference in UI (CHAOS-9114)
• Added targets support to the timeline view (CHAOS-9048)
• Added support for running windows process kill fault on windows infrastructure using SYSTEM user (CHAOS-9013)
• Added support for node faults on GKE Autopilot (CHAOS-8981)
• Enhanced healthcheck support to pod-io-stress fault (CHAOS-8896)
• Added name sorting support for Experiment list (CHAOS-8875)
• Added ENV support to the inline command probe for both backend and UI (CHAOS-8850, CHAOS-8816, CHAOS-8814)
• Added OOTB probes as inline command probes for kubernetes infrastructure (CHAOS-8815)
• Enhanced UI to surface Common Kubernetes Errors (ImagePullBackOff, Evicted, OutOfCpu) for Kubernetes Infrastructure (CHAOS-8709)

Fixed Issues​

• Fixed gameday execution error when trying to run a gameday with same ID as another gameday in different project due to missing identifiers in query filters (CHAOS-9247)
• Fixed chaos license page not being accessible in production environments where subscriptions page was crashing when refreshed (CHAOS-9215)
• Fixed list probes API not working for older probes/experiments and panicking due to user details not being available by adding nil check for user details in recent probe execution details (CHAOS-9206)
• Fixed multiple probes getting selected upon using import probe in Local Hub (CHAOS-9202)
• Fixed error in CMD probe when Float data type is provided and value is a runtime input where selecting Runtime Input for Comparator Value was throwing a type validation error (CHAOS-9171)
• Fixed inputs being duplicated for a given probe pulled at various points in an experiment by fixing the support for multiple probes/actions inside one experiment (CHAOS-9167)
• Fixed not being able to abort experiments in production environments (CHAOS-9144)
• Fixed action and probe search filtering not working in Chaos Studio by updating parameter names for clarity where actionType has been changed to entityType and infrastructureType has been simplified to infraType (CHAOS-8994)
• Fixed Local Hub showing 0 faults and 0 actions although fault and action templates exist within the hub by updating the mongo query to fetch the count of chaos hub resources (CHAOS-8962)
• Fixed Timeline View showing action name instead of the type (Delay/Custom Script) by adding a new field Type to the ActionData struct within the execution data (CHAOS-8951)
• Fixed input validation issues in DataDog/Dynatrace probes in both linux/k8s when providing runtime inputs where support for runtime inputs in time frame and test type field was broken (CHAOS-8943)
• Fixed V2 experiments showing status and type as empty strings and showing error in YAML UI by updating the hce-sdk to allow for optional fields in the experiment manifest (CHAOS-8931)
• Fixed Resilience Score coming as 0 even though all probes passed by correcting the calculation of the resiliency score to ensure accurate scoring based on the status of chaos execution nodes (CHAOS-8930)
• Fixed Custom Script Action argument being a mandatory field when it should be optional as commands may not have arguments necessarily (CHAOS-8874)
• Fixed probe status being reported as failed on abort when it should be N/A or Pending by updating probe status to be Stopped or Skipped on experiment abort (CHAOS-8804)
• Fixed Timeline View experiments keeping running indefinitely by ignoring the DDCR pod from the target pod selection (CHAOS-8801)
• Fixed pod-delete fault being randomly reported as Error even though it succeeded by resolving the issue where the fault status was incorrectly reported (CHAOS-8800)
• Fixed target not showing in the timeline view representation of the fault by adding target details to the fault data structure (CHAOS-8796)
• Fixed Actions "Stop on Failure" radio button not being applicable (CHAOS-9025)
• Fixed empty cards for fixed inputs in the experiment inputsets page in Chaos Studio (CHAOS-9094)

June 2025​

Version 1.62.11​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.62.4
• harness/chaos-log-watcher:1.62.1
• harness/service-discovery-collector:0.42.1
• harness/chaos-ddcr-faults:1.62.2

What's New​

New Chaos Experiments​

• VPC Route Misconfiguration for AWS - Simulate network connectivity issues by disrupting connections to Transit Gateway, NAT Gateway, or VPC Peering connections. (CHAOS-8181)
• Windows Memory Stress Experiments - Run memory stress tests on Windows systems without requiring administrator privileges. (CHAOS-8468, CHAOS-8456)
• AWS Windows Chaos Experiments - Execute chaos experiments on Windows EC2 instances using AWS Systems Manager. (CHAOS-8139, CHAOS-8141)
• Linux Experiments with Non-Root Users - Install and run chaos experiments on Linux systems without root access (limited fault selection available). (CHAOS-8121)

Enhanced User Experience​

• Redesigned Experiments Page - Cleaner, more intuitive interface for managing your chaos experiments. (CHAOS-8185)
• Improved Timeline View - Better visualization of experiment execution with support for new probes, actions, and faults. (CHAOS-8558)
• Enhanced Chaos Studio - New properties and variables tabs for easier probe and action configuration. (CHAOS-8745)
• Action Management - Dedicated details screen for managing standalone actions. (CHAOS-8178)
• Permission Indicators - Visual indicators on fault cards showing basic or advanced permission requirements with helpful tooltips. (CHAOS-8397)

Platform Improvements​

• APM Probe Integration - Monitor application performance during chaos experiments with Application Performance Monitoring probes. (CHAOS-8518)
• Parallel Linux Experiments - Run multiple chaos experiments simultaneously on Linux infrastructure in SaaS environments. (CHAOS-8748)
• Vanity URLs - Support for vanity URLs across chaos management services. (CHAOS-8528)
• Enhanced Experiment Construction - Streamlined process for creating experiments with experiment-level probes, actions, and templates. (CHAOS-8062)
• Unified Validation - Consistent schema validation across all infrastructure types supporting experiment-level probes. (CHAOS-8608)

Improvements and Bug Fixes​

Timeline and Execution​

• Fixed experiment status not updating correctly when experiments fail
• Improved timeline view to properly display skipped and error nodes with accurate duration
• Resolved missing data issues in timeline execution details
• Fixed experiment execution getting stuck in queued state

User Interface​

• Fixed dropdown pagination controls not working properly
• Improved error messaging to show clearer descriptions instead of technical error codes
• Resolved display issues with Windows infrastructure troubleshooting commands
• Fixed various UI components for better user experience

API and Data Management​

• Improved API responses for experiment lists, faults, probes, and actions
• Fixed data counting issues in Local Hub and Chaos Hub
• Enhanced authentication handling to prevent unnecessary impersonation data
• Improved schema handling for better compatibility
• Added restriction in backend to prevent the deletion of default kubernetes system probe via API.

Infrastructure and Performance​

• Fixed vulnerabilities by upgrading jwt and ff-go-sdk go packages in all chaos components.

May 2025​

Version 1.61.11​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.61.1
• harness/chaos-log-watcher:1.61.1
• harness/service-discovery-collector:0.41.0
• harness/chaos-ddcr-faults:1.61.2

What's New​

Enhanced Platform Support​

• Istio and Virtual Services Support - All SMP services now fully support Istio service mesh and virtual services for better network management and traffic routing. (PL-59078)

APM Monitoring Improvements​

• Splunk Observability Integration - Added Splunk Observability support for APM probes, expanding monitoring capabilities during chaos experiments. (CHAOS-8463)

• AppDynamics Comparator Support - Enhanced APM probes with comparator functionality for AppDynamics connector, enabling better threshold-based monitoring. (CHAOS-8513)

• APM Probe Details View - Improved experiment execution page to display APM probe properties and details for better visibility. (CHAOS-8163)

Experiment Management​

• BYOC (Bring Your Own Container) Enhancements - Added runtime input support and improved user experience for custom container-based chaos experiments. (CHAOS-8162, CHAOS-8128, CHAOS-8153)

• Kubernetes V2 Beta1 Support - Added versioning support for Experiment Factory with new experiment schema for Kubernetes v2 beta1 manifests. (CHAOS-8137)

• Custom URLs Support - Enhanced chaos management services with vanity URL support for better branding and accessibility. (CHAOS-7920)

Backend and Infrastructure​

• Improved ChaosGuard - Fixed YAML mode issues for ChaosGuard conditions in Linux and Windows infrastructures. (CHAOS-8146)

Improvements and Bug Fixes​

Experiment Creation and Validation​

• Fixed YAML Validation Issues - Resolved problems where unrelated errors would block experiment creation and saving when using blank canvas or YAML upload methods. (CHAOS-8585)

• Enhanced Linux Experiment Execution - Fixed execution issues in Linux environments by updating TaskDefinition CRD charts with required response body fields. (CHAOS-8573)

Probe Management and Validation​

• Command Probe Fixes - Resolved multiple issues with Kubernetes CMD probes including source attribute handling and editing functionality. (CHAOS-8555, CHAOS-8550, CHAOS-8581)

• Dynatrace Probe Improvements - Fixed probe creation issues and added proper validation for numerical values in Float type comparisons. (CHAOS-8554, CHAOS-8547, CHAOS-8546)

• Prometheus Probe Validation - Added mandatory validation for query and query path fields to ensure proper probe configuration. (CHAOS-8544)

• APM Probe Display - Fixed issues where APM probe properties weren't showing correctly on experiment execution pages. (CHAOS-8565)

• Bulk Probe Operations - Improved bulk disable functionality for probes with better error handling and logging. (CHAOS-8552)

Platform Stability​

• Windows Build Improvements - Resolved Windows-specific build issues for better cross-platform compatibility. (CHAOS-8535)

• Helm Chart Updates - Fixed missing Kubernetes event watcher image references in Helm charts. (CHAOS-8453)

• Pod Status Monitoring - Enhanced pod-delete fault to properly check status for all affected pods. (CHAOS-8539)

• TLS Configuration - Improved handling of empty TLS configuration objects for APM probes. (CHAOS-8158)

April 2025​

Version 1.59.0​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.59.0
• harness/chaos-log-watcher:1.59.0
• harness/service-discovery-collector:0.39.0
• harness/chaos-ddcr-faults:1.59.0

New features and enhancements​

• Added support for Windows CPU stress on Windows dual socket servers. (CHAOS-7902)
• Adds support for dynamic updates in Active Discovery and the Application Map after each discovery run—new services are added, removed services are marked accordingly, and resources from newly excluded namespaces are hidden, with their services shown as removed if referenced in the Application Map. (CHAOS-7745)
• Added Windows global blackhole experiment to cause global blackhole chaos on a Windows VM. (CHAOS-7091)

Fixed issues​

• Fixed an issue that was causing runtime input-based fault additions to fail in Kubernetes (Harness Infrastructure) experiments when using a private image registry with a secret. The system now correctly handles secrets during fault creation in these scenarios. (CHAOS-8006)

• Added an index on the environment_id field to improve query performance and reduce latency for operations involving environment-specific data. (CHAOS-7726)

• Updated the title header on the Getting Started page from a question format ("Create Chaos Experiments on your ______?") to a clear statement: "Create Chaos Experiments on your Infrastructure" for better readability and user experience. (CHAOS-7802)

• Resolved a goroutine leak in chaos-manager by properly closing response bodies, preventing memory spikes and unbounded memory usage. (CHAOS-7972)

• Corrected the experiment count display in Bulk Run on HCE-SaaS UI by replacing the hardcoded value with a dynamic variable to reflect the actual number of selected experiments. (CHAOS-7818)

March 2025​

Version 1.57.2​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.57.0
• harness/chaos-log-watcher:1.57.0
• harness/service-discovery-collector:0.37.1
• docker.io/harness/chaos-ddcr-faults:1.57.0

New features and enhancements​

• Replaces the large text blocks in the pod details and probes tabs with concise tooltips and "Learn More" links, directing users to relevant content in the Harness Developer Hub for improved readability and navigation. (CHAOS-7621, CHAOS-7619)

• Adds support for targeting multiple keys in the Linux Redis Cache Expire and Kubernetes Redis Cache Expire faults.(CHAOS-7283)

Fixed issues​

• Fixed an issue where services were not appearing on the application map in the Chaos module when the discovery agent was created through the Discovery page. (CHAOS-7743)

• Fixed scrolling issue with the ChaosHubs side navigation bar. (CHAOS-7728)

• Resolved an issue with the timeline view for an experiment with multiple probes, which showed incorrect probe information when user clicked any probe. (CHAOS-7718)

• Resolved an issue in the experiment timeline view where all timestamps were unnecessarily converted to Unix milliseconds, even when already in Unix millisecond format. (CHAOS-7719)

February 2025​

Version 1.56.3​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.56.0
• harness/chaos-log-watcher:1.56.0
• harness/service-discovery-collector:0.36.0
• docker.io/harness/chaos-ddcr-faults:1.56.0

New features and enhancements​

• Running timeline view in the Execution View: You can track the real-time execution of chaos experiments for improved visibility. (CHAOS-6672)

• Improved chaos infrastructure search navigation: Searching for chaos infrastructure from the Chaos Experiment page now includes a search bar in the list view, making navigation easier when dealing with multiple entities. (CHAOS-7556)

Fixed issues​

• Fixed issue that prevented auto-creation of experiments with unsupported kinds. (CHAOS-7600)

• Linux Dynatrace probe details did not appear correctly during creation. This issue has been fixed. (CHAOS-7596)

• Fixed caching issue when selecting chaos faults in Chaos Studio. (CHAOS-7595)

• Resolved ChaosGuard failure for non-Kubernetes experiments. (CHAOS-7636)

• Fix typo in category for template validation. (CHAOS-7692)

Version 1.55.1​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.55.0
• harness/chaos-log-watcher:1.55.0
• harness/service-discovery-collector:0.35.0
• docker.io/harness/chaos-ddcr-faults:1.55.0

New features and enhancements​

• Modifies the Chaos Select Pipeline component to accept additional resources as properties and support allowed infrastructure types. (CHAOS-7035)

Fixed issues​

• Fixed an issue where the Kubernetes chaos infrastructure was being auto-selected even when other infrastructure types were enabled. (CHAOS-7549)
• Updated the useGetChaosExperimentStats API to support project and account-scoped data retrieval. (CHAOS-7530)

Version 1.54.0​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.54.0
• harness/chaos-log-watcher:1.54.0
• harness/service-discovery-collector:0.34.0
• docker.io/harness/chaos-ddcr-faults:1.54.0

New features and enhancements​

• Upgrades the base image for all chaos services from UBI-8 to UBI-9 to resolve vulnerabilities. (CHAOS-7504)

• Supports using a chaos fault template with boolean values. (CHAOS-7514)

• Supports including a templated fault for an experiment that is part of an application map. (CHAOS-7516)

Fixed issues​

• Fixed the issue where clicking a Kubernetes Harness Infrastructure from the Experiments table threw a 400 (Bad request) error. (CHAOS-7512)

• Fixed an issue where Kubernetes API, HTTP, and network faults failed due to a malfunctioning container ID derivation command in the customer's Docker environment. This has been resolved by allowing user-defined input for the command pattern. (CHAOS-7255)

January 2025​

Version 1.53.1​

Images required​

Listed below are the images to download to use image registry with Harness Delegate.

• harness/chaos-ddcr:1.53.0
• harness/chaos-log-watcher:1.53.0
• harness/service-discovery-collector:0.33.0
• docker.io/harness/chaos-ddcr-faults:1.53.0

New features and enhancements​

• Upgrades all Go Services to use Go 1.23 to address the vulnerabilities.(CHAOS-7324)

• Upgrades all the third-party binaries like kubectl, crictl and gcloud to the latest versions to resolve the vulnerabilities. (CHAOS-7322)

• Introduces support for specifying Transaction Percentage in DNS chaos faults. (CHAOS-7244)

Version 1.52.3​

• New videos: Pod Network Rate Limit, Pod API Modify Header, Pod IO Attribute Override, Pod API Block, Pod API Modify Body. Adds a playlist with Kubernetes Pod Chaos experiments.

Fixed issues​

• Fixed the issue where fault templates were not working without enabling the feature flag associated with it. (CHAOS-7210)

• Fixed the issue where automatic experiment creation was unable to create the required number of experiments in Advanced/Maximum mode. (CHAOS-7214)

Previous releases​

Cloud Cost Management Release Notes

Review the notes below for details about recent changes to Harness Cloud Cost Management. For release notes for Harness Self-Managed Enterprise Edition, go to Self-Managed Enterprise Edition release notes. Additionally, Harness publishes security advisories for every release. Go to the Harness Trust Center to request access to the security advisories.

Important Notice:​

We've migrated to LabelsV2, which preserves your original label keys while dramatically improving perspective load times—from 1 minute down to under 2 seconds. Action required: Please update your automated scripts to ensure compatibility with the new system.

Instructions to Update

April 2026 - Version 1.86.15​

Deployment Date: April 17, 2026 (Prod-1)​

New Features and Enhancements​

• Cost Settings now respects RBAC permissions: Users with CCM Viewer (view-only) access will no longer see an enabled Save Preferences button in the Cost Settings tab. The Cost Settings preferences section in both the Recommendations and Anomalies pages now correctly reflects manage permissions — consistent with how all other preference sections behave. [CCM-31477]

• Rejected and Ignored lists moved to the main Recommendations view: These lists have been relocated from the drawer into the Recommendations list itself, enabling users to export rejected and ignored recommendations directly without any extra steps. [CCM-30485]

• GCP invoice month filter in Perspectives shows correct months on month-end dates: The invoice month filter in Perspective filters now accurately reflects the current month when viewed on the last day of the month. [CCM-31433]

• Chart tooltips and x-axis in Overview page now display time in UTC: The overview page chart now consistently shows UTC time in both the tooltip and x-axis labels, removing ambiguity across time zones.

• Budget alerts section shows the correct label per budget type: The alerts section on the budget details page now displays the appropriate label based on the budget type configured. [CCM-31190]

March 2026 - Version 1.84.4​

Deployment Date: March 26, 2026 (Prod-1)​

New Features and Enhancements​

• Cloud Billing Tags for Node Pool Recommendations: Kubernetes node pool recommendations now support cloud billing tags (such as allocation and environment tags from AWS cost data) in addition to Kubernetes node labels. (CCM-31030)

• Tags for Node Pool and Workload Recommendations: Node pool and workload recommendations now display tags, consistent with AWS EC2 recommendations. (CCM-30822)

Fixed Issues​

• Node Pool Preset UI Breaking on AWS Region Addition: Fixed an issue where editing a node pool preset and adding an AWS region caused the UI to break upon saving, preventing navigation back to the Presets page. (CCM-31303)

• JIRA Reporter Field Not Passing Correct Value: Fixed an issue in the JIRA creation flow where selecting the reporter field did not pass the correct value in the payload. (CCM-31249)

• RDS Rules Showing Incorrect Traffic-Based Rule Type: Fixed an issue where RDS rules displayed an incorrect traffic-based rule type selection. (CCM-31115)

• AutoStopping Rule Edit Button Disabled for Deleted Connectors: The AutoStopping rule edit button is now disabled when the associated connector is deleted. (CCM-31107)

• Workload Anomalies Showing "undefined" for Service Label: Fixed an issue where workload anomalies displayed "undefined" for the service field label. (CCM-30994)

• Recommendation Presets Screen Broken Due to EC2 Presets: Fixed an issue where the recommendation presets screen was broken due to new EC2 presets. AWS EC2 passthrough support is hidden until fully implemented. (CCM-30983)

March 2026 - Version 1.83.1​

Deployment Date: March 18, 2026 (Prod-1)​

Fixed Issues​

• Anomalies Drilldown List Not Rendering Multiple Resources: Fixed an issue where the anomalies drilldown list failed to render details correctly when multiple resources were present. The list now properly displays resource names (Cost Category/Cost Buckets) instead of showing empty results. [CCM-30735]

• Asset Governance Rules Not Generating Recommendations: Resolved an issue where custom rules in Asset Governance stopped generating new recommendations, even though they appeared under "Rules Generating Recommendations." We have added a retry mechanism that automatically restarts BQ load jobs if they remain in progress for more than one day. [CCM-30639]

• Kubernetes Cluster Anomalies Reporting Incorrect Costs: Fixed an issue where service-level anomalies within a Kubernetes cluster displayed the total cluster cost instead of the cost for the specific service, resulting in inflated anomaly amounts. Additionally, resolved a related issue where the anomaly drilldown view for cluster resources failed to load. [CCM-30591]

• Jira Ticket Comments Not Saving for CCM Recommendations: Resolved an issue where adding a comment while creating a Jira ticket for a CCM recommendation was not working. [CCM-30525]

• Budget Amount Changing on Scroll: Fixed unexpected behavior in the budget UI where scrolling while the cursor was in a budget textbox caused the amount to change, including going negative. Scroll events are now disabled for numeric input fields. [CCM-30434]

• AWS Passthrough Recommendations Not Showing Overridden Savings: Fixed an issue where overridden savings for AWS passthrough recommendations were not displayed on the recommendation details page. Overridden savings now display correctly, consistent with other recommendation types. [CCM-30432]

March 2026 - Version 1.82.5​

Fixed Issues​

• AWS Perspectives Anomaly Data Returning Empty Results: Anomaly data for AWS perspectives was returning empty results due to a field ID case-sensitivity mismatch. This has been resolved. [CCM-30345].
• Cost Category Rule Validation Error on Save: When creating a cost category with a cost bucket rule referencing another Cost Category as "not null", saving the configuration would produce an error on the Cost Category field within the Cost Bucket rule. This has been fixed.[CCM-30294].
• Memory Cluster Capacity Limits Missing Default Unit: When setting Memory Cluster Capacity Limits, a unit was required despite the UI implying GiB as the default. The system will now automatically append "Gi" to Cluster Capacity Limits Memory values when a numeric-only value is entered.[CCM-30217]
• Perspective and Budget API Fixes: The API now correctly handles perspective and budget operations that previously failed or went untracked. Specifically: folder validation has been tightened to reject empty inputs on update and move; duplicate perspective names are now permitted as designed; and perspective moves are fully logged in the audit trail. Budget management reflects the same improvements.[CCM-29269]

March 2026 - Version 1.81.2​

Deployment Date: March 03, 2026 (Prod-1)​

Feature Improvements​

• Disable Atomization in Commitment Orchestrator: We've added a new Disable Atomization option in Commitment Orchestration setup, giving you greater control over your commitment purchase strategy. This is useful for:
  • Extremely stable, predictable workloads with no expected changes for 3+ years
  • Organizations prioritizing absolute lowest per-hour costs over flexibility
  • Environments with regulatory requirements for fixed, long-term capacity commitments
  • Teams preferring simplified "set and forget" commitment management

Turning off Atomization will revert RI expirations to an annual schedule instead of monthly. This will affect your RI optimization strategy due to:

• Reduced flexibility
• One renewal opportunity per year or 3 years instead one per month
• Higher lock-in risk
• Full-year commitments with fewer exit opportunities
• Lower potential savings
• Limited ability to optimize based on usage changes

[CCM-29445]

• Folder Name Validation: Leading and trailing spaces are no longer allowed in folder names. [CCM-29248]

• Cost Category–JIRA Mapping Cleanup: Empty rows will no longer be added when users attempt to create multiple mappings without saving them. [CCM-29161]

Fixed Issues​

• Perspective Rule Reset Issue: Fixed an issue on the Perspective page where editing rules in the rule builder and then changing the name or folder caused the rules to reset to the originally saved configuration. [CCM-29971]
• Applied Recommendations Chart Aggregation: Improved the Recommendations vs. Savings chart to prevent clutter when large time ranges are selected. Aggregation now switches to monthly when the range exceeds three months and to yearly when it exceeds one year. [CCM-25390]
• Recommendation createdAt Field Population: Fixed an issue where the createdAt field was not populated when generating recommendations. The field is now correctly populated upon recommendation generation. [CCM-29613]

February 2026 - Important Updates to Dashboard​

We are officially deprecating reporting_timeframe field and replacing it with a much more intuitive and powerful workflow.

To ensure a smooth transition with minimal disruption to your daily operations, we are rolling out this change in two careful phases: an automated migration and the official retirement of the field

Please note that this entire deprecation process will be fully completed by March 1, 2026.

Phase 1: Automated Dashboard Migration​

We are running an automated migration script across all sub-folders within your account's dashboard folder. The script will first update your user attributes to ensure your data access remains perfectly aligned. From there, it will iterate through your dashboards and seamlessly update your filters.

Because we know you rely on historical comparisons, the script uses specific logic to update both your dashboard-level and element-level filters without breaking your current views.

Dashboard-Level Changes: The script evaluates the elements on your dashboard and updates the old reporting_timeframe filter based on the following:

Element-Level Changes: For individual tiles and elements, the script will swap out the old reporting_timeframe filter for a standard start_date filter.

The technical detail: If an element uses legacy cost trend reporting, the script will automatically set the new start_date to 2x the original timeframe (or a default of 60 days). We do this intentionally to ensure Looker pulls enough historical data to calculate your period-over-period trends accurately.

Below is a high-level diagram detailing what the script will do:

Phase 2: Retiring the Field​

Once the migration is complete on 03/01/26, the reporting_timeframe field will no longer be a viable filter for standard time filtering.

However, we are not deleting it entirely. It will remain as a hidden field. This ensures that any highly specific legacy period-over-period reporting that relied on it will not suddenly break.

Cluster Orchestrator Release - February 23, 2026​

Version 0.8.1 (Operator) & 0.4.1 (Helm Chart)​

What's New

• In-Place Pod Resizing with VPA 1.5.1: The highlight of this release is the upgrade to VPA 1.5.1, which introduces in-place pod resizing. This feature allows the Vertical Pod Autoscaler to adjust CPU and memory resource requests for running pods without requiring pod restarts, significantly reducing disruption to your workloads.

• Enhanced Security:

  • AWS IMDSv2 Support: Full compatibility with AWS Instance Metadata Service Version 2, providing improved security for EC2 instance metadata access
  • Rapidfort Hardened Container Images: All container images are now hardened using Rapidfort's security scanning and optimization, reducing attack surface and improving security posture

If you're using VPA features, ensure that the feature flag is enabled for your account. Contact support@harness.io if you need assistance enabling VPA capabilities.

February 2026 - Version 1.80.3​

Deployment Date: February 19, 2026 (Prod-1)​

Feature Improvements​

• AutoStopping Logs Improvements: AutoStopping logs now include pagination and 30-day retention. You can download logs and select your preferred timezone when exporting. [CCM-25281]

• Anomalies Widget Enhancement: The Anomalies legend in the Cloud and Cluster Spend widget is now automatically disabled when both Cluster Spend and Cloud Spend are deselected, ensuring anomalies are only displayed alongside relevant cost data. [CCM-26683]

• Cluster Orchestrator Refresh Button: Added a refresh button to Cluster Orchestrator tables, allowing you to quickly check for updates to nodes, workloads, schedules, and logs. [CCM-27972]

• Total Anomalies Widget: A new Total Anomalies widget is now available on the Perspective Details screen. This widget displays data when supported filters are applied. When unsupported filters are selected, a helpful message appears indicating which filters support anomaly detection. [CCM-28637]

• Commitment Setup Display: The commitment setup review section now displays account names instead of account IDs for improved readability. [CCM-28983]

Fixed Issues​

• Cluster Orchestrator Pagination: Fixed two pagination-related issues:
  • Page numbers now reset correctly when filters are changed in Cluster Orchestrator Logs.
  • Switching between Cluster Orchestrator tabs no longer persists page numbers in the URL, preventing unintended pagination carryover. [CCM-29746]

February 2026 - Version 1.79.21​

Deployment Date: February 16, 2026 (Prod-1)​

Feature Improvements​

• Anomaly Alert Recipients Display: The Anomalies Overview screen now displays the email and Slack recipients to whom alerts were sent for each anomaly, providing better visibility into alert distribution. [CCM-28638]

• Anomaly Filtering by Cost Buckets: Added a quick filter to view anomalies by Resource or Cost Buckets. In the Cost Buckets view, you can see which cost bucket each anomaly belongs to, drill down for details, and redirect to a Perspective with the cost bucket filter applied. You can also filter by cost categories within the Cost Buckets view. This feature is behind a feature flag. Contact Harness Support to enable it. [CCM-27460]

• Enhanced Anomaly-to-Perspective Mapping: The anomaly fetch API now supports enhanced perspective query filters. Anomalies can now be mapped to perspectives using AWS Account Name/ID, AWS Service, and AWS Usage Type filters, providing more granular anomaly analysis. [CCM-29488]

• Improved Budget Alert Email Accuracy: Daily budget alert emails now display the date when the cost was actually incurred, rather than when the alert was generated. This improves accuracy when alerts are processed with a delay due to cloud provider cost data latency. [CCM-29490]

• Kubernetes AutoStopping Rules V2: Kubernetes AutoStopping rules have been upgraded to V2. The new template is now available on the K8s rule creation step:

apiVersion: ccm.harness.io/v1
kind: AutoStoppingRule
metadata:
  name: <rule_name>
  namespace: <namespace_of_workload>
  annotations:
    harness.io/cloud-connector-id: autostopping_ccm_play
spec:
  idleTimeMins: <idle_time_in_minutes>
  hideProgressPage: <decides_whether_to_hide_progress_page_defaults_to_false>
  workloads:
    <unique_id_of_workload>:
      namespace: <namespace_of_deployment>
      name: <name_of_deployment>
      id: <unique_id_of_workload>
      desired: <replica_count>
      type: <type_of_workload_Deployment/Statefulset>
      warmup_delay_sec: <duration_in_sec_before_depender_is_started_on_warmup>
      traffic_routes: 
        - ingress_name: <name_of_ingress>
          ingress_type: <type_of_ingress.current_support_nginx>
          service_name: <name_of_k8s_service_for_the_workload>
          service_host: <host_name_of_the_workload>
          service_port: <port_specified_on_service>
          service_path: <service_path_specified_on_ingress_default_/>
          path_type: <k8s_ingress_path_type>
          target_path: <target_path_if_any_override>
  relations:
    <unique_id_of_dependent_workload>:
      - <unique_id_of_dependee_workload>
  dependencies: [<unique_id_of_workload>]

[CCM-26821]

Fixed Issues​

• We identified and resolved an issue affecting Cost Category rules that use the NOT_IN operator. When a rule was configured to exclude specific values, the system was unintentionally also excluding line items where the field value was empty or undefined.

After the fix, only the values explicitly listed in the NOT_IN condition are excluded. Cost Categories now accurately classify costs according to your defined rules [CCM-29692]

February 2026 - Version 1.78.21​

Deployment Date: February 7, 2026 (Prod-1)​

Feature Improvements​

• Enhanced Anomaly Detection Logic: Improved the machine learning models to handle limited data scenarios more effectively. The system now provides more accurate anomaly detection when working with sparse cost data points by better aligning the Prophet and Statistical models. [CCM-29409]

• Improved Budget Reporting: Enhanced the display of budget variance percentages by rounding to 2 decimal places for yearly budgets, providing more precise financial reporting. [CCM-29332]

• Redesigned Cost Category Creation: Streamlined the user experience by moving the Cost Category creation from a drawer to a dedicated page, matching the familiar workflow of Perspective creation. [CCM-28931]

• Standardized Operator Naming: Reverted the ! IN and ! NULL operators back to the more widely recognized NOT IN and NOT NULL syntax in the Perspective and Cost Category builder for better clarity and consistency. [CCM-25286]

• Unallocated Cost Label: Earlier, when the Hidden strategy was selected for Unallocated Cost in cost categories, we stamped a default label (“Cost categories default”) in the data used by the dashboard. Now, we stamp the Unallocated Cost label irrespective of what strategy is selected. This ensures that the dashboard data remains consistent even if the strategy is changed later. [CCM-29236]

Fixed Issues​

• Governance Recommendations: Resolved an issue where multiple "remove from ignore list" options were appearing for ignored Governance recommendations. The interface now correctly shows a single option to remove from the appropriate ignore level. [CCM-28071]

January 2026 - Hotfix: Changes to Anomaly Detection​

Deployment Date: January 30, 2026 (Prod-1)​

This release focuses on improving the clarity, accuracy, and usability of Anomalies across the product. We’ve enhanced the UI, corrected spend calculations and date logic, expanded API filtering capabilities, and improved alert accuracy so teams can detect, understand, and act on cost anomalies with greater confidence.

What’s new​

• Fixed an issue where the resource name and tags overlapped in the Anomalies resources table.
• Replaced the Cost Impact column with One-Day Change in the Anomalies drill-down resource table for clearer day-over-day insights.
• Improved UX and consistency across the Anomalies home and drill-down pages. Spend metrics were renamed for clarity, Expected Spend and % Change were added, layouts were simplified to reduce visual noise, and date comparisons and spend calculations were corrected. We also fixed several long-standing UI issues (multi-cloud sorting, table labels/layout, missing tooltips, and comment ordering) to improve reliability and usability.
• Enhanced the Anomalies fetch API to support Perspective-based filters. Anomalies can now be mapped using AWS Account Name/ID, AWS Service, and AWS Usage Type.
• Daily budget alert emails now display the date the cost was incurred instead of the alert generation time, ensuring accurate context when cloud cost data arrives with a delay.

January 2026 - Version 1.75.1​

Deployment Date: January 24, 2026 (Prod-1)​

Feature Improvements​

• Multi-Container Workload Downloads: Since multi-container workload recommendations are split into individual line items during download, the total count increases even though the actual recommendation count remains the same. We’ve added a message in the product during download to surface this behavior. [CCM-25080]

• Default Folder Visibility: Perspectives moved to Default folder after a folder deletion will now show up during the creation of a new folder [CCM-26370]

• Recommendations Breakdown: We’ve added realized savings from Recommendations to the Recommendations Breakdown widget on the CCM Overview page. [CCM-26915]

• Cost Tooltips for Cluster Orchestrator: We’ve added tooltips across Cluster Orchestrator to clarify that in Cluster Schedules savings, on-demand instances use netamortizedcost and spot instances use unblendedcost. In Cluster Savings, on-demand instances use publicondemandcost and spot instances use unblendedcost. [CCM-27616]

• Cluster Orchestrator Pagination: We’ve added support for search and pagination on the Cluster Orchestrator list page. [CCM-27805]

Fixed Issues​

• Governance Evaluation CSV Export: We fixed an issue where exporting a Governance evaluation with Keep Filters Applied selected, but no filters actually applied, resulted in an empty resource file being downloaded. [CCM-25498]

• Bug Fix for Cost Category mapped Jira Ticket: We fixed an issue where, when the Jira project came from a Cost Category mapping, the issue type wasn’t being fetched automatically on the initial load of the Create Jira Ticket form. [CCM-29166]

January 2026 - Version 1.75.1​

Deployment Date: January 16, 2026 (Prod-1)​

Feature Improvements​

• “Others” cost preference now supports negative values: Previously, when Show Others (not in top 12) was enabled, negative costs in the “Others” category were shown as zero. We’ve fixed this—negative values will now display correctly in Perspectives. [CCM-28678]

• Perspective chart x-axis now shows only valid dates: Fixed an issue where extra dates appeared at the beginning and end of the chart timeline. The x-axis now reflects only the dates present in the dataset. [CCM-28640]

• Security Groups dropdown now loads correctly for Azure AutoStopping Proxy: Resolved an issue where the Security Groups list wasn’t populating during proxy creation. [CCM-28559]

• Improved navigation from Anomaly Details when no contributing resources exist: If an anomaly has no contributing resources, we now display the root resource details and automatically redirect you to the relevant Perspective view. [CCM-28219]

• New SP roll-down policy option in Commitment Orchestrator renewals You can now configure a roll-down percentage to control how much of an expiring Savings Plan commitment is renewed during setup. [CCM-27577]

January 2026 - Hotfix: Changes to AutoStopping Rules​

Deployment Date: January 12, 2026​

We have rolled out a hotfix that includes the following major changes:

• We've refined our savings calculation timeframe. Savings will now be computed with data up to two days prior to the current date (T-2), rather than the previous approach of using data up to the previous day (T-1) so that complete data is received.

• AutoStopping rules cannot be created in a disabled state. This option was not available in the UI but could be done through the API to create disabled rules. Moving forward, any attempt to create or edit a disabled rule will result in a validation error. AutoStopping rules can now only be disabled after creation.

• Savings will be calculated for rules that are created and resources are being utilized. Rules created with valid resources will display savings numbers as long as the resource is valid and cost data is available in the Cloud Usage Report (CUR). Resources that are onboarded in "Down" state and not used after rule creation will not show any savings, as the details of such resources will be completely missing in the CUR.

January 2026 - Version 1.75.1​

Deployment Date: January 8, 2026 (Prod-1)​

[New Feature] AutoStopping Alerts​

We have introduced alerts for AutoStopping rules. Configure notifications for critical events related to your AutoStopping rules. You can set alerts for Rule create, Rule update, Rule delete, Rule warm up failures and Rule cool down failures and receive real-time notifications directly in Slack or e-mail.

December 2025 - Version 1.73.5​

Deployment Date: December 12, 2025 (Prod-1)​

Feature Improvements​

• Enhanced Rule Management: Improved rule update logic to prevent unnecessary exceptions when updating existing rules with the same name. The system now verifies if the rule being updated is the same as the existing one before throwing any exceptions. [CCM-27875]

• CSV Export for Anomalies: We’ve added support for CSV export in Anomalies, similar to Recommendations. Users can now export detailed anomaly data including Anomaly Link, Start Date, End Date, Duration (Days), Criticality, Resource Name, Resource Info, Cloud Provider, Total Spend, Expected Amount, Anomalous Spend, Anomalous Spend Percentage, Status, Status Updated At, Status Updated By, and Ongoing Anomaly. [CCM-27889]

• Accurate Budget Reporting: Refined budget spend calculations to show true costs-to-date by excluding future pre-allocated costs. This provides more accurate and actionable "Spend till date" values for better financial planning. [CCM-27253]

• Enhanced Governance Exports: Extended CSV export functionality for Governance recommendations to include Action Type and Evaluated At columns, providing more comprehensive compliance reporting. [CCM-26912]

• Perspective Cloning: Implemented perspective cloning capability, allowing users to create copies of existing perspectives with custom names and save them to their preferred destination folders. [CCM-25667]

December 2025 - Cluster Orchestrator Release - 0.7.0​

Deployment Date: December 12, 2025​

Feature Improvements​

• Karpenter 1.7.3 Support: Cluster Orchestrator now supports Karpenter 1.7.3 features. Users must re-run the enablement script or Terraform template and perform a Helm upgrade to add new permissions for Cluster Orchestrator related to Karpenter 1.7.3.

• Expanded Kubernetes Support: Now supports Kubernetes (EKS) version 1.33; previously it supported only 1.32.

• Enhanced Karpenter Features: New major Karpenter features (from 1.2.4 to 1.7.3) for non-Karpenter customers include:

  • Support for ODCRs (on-demand capacity reservations)
  • SSM Parameter support for AMI selection
  • Improved IAM instance profile management)
  • Bug fixes and performance improvements

• Auto-Discovery Support: We have enhanced the Cluster Orchestrator with automatic discovery capabilities for AWS configuration parameters, eliminating the need to manually specify certain environment variables in the deployment. Environment Variables with Auto-Discovery Support:

  • AWS_DEFAULT_INSTANCE_PROFILE
  • AWS_NODE_ROLE_ARN

  Previously, these environment variables were required to be explicitly defined in the deployment configuration. Now, the Cluster Orchestrator automatically discovers these values on startup.

  While auto-discovery is enabled by default, users can still manually configure these environment variables if needed. When set explicitly, the manually configured values take precedence over auto-discovery.

December 2025 - Version 1.71.3​

Deployment Date: December 3, 2025 (Prod-1)​

Feature Improvements​

• Enhanced Recommendation Management: We’ve added support for temporarily ignoring recommendations. Users can now ignore a recommendation for 7, 15, or 30 days, or ignore it permanently. Once the selected duration expires, the recommendation will automatically move back to the Open tab on the Recommendations page. [CCM-27645]

• Improved UI Navigation: Enhanced the user experience on Perspectives and Budgets pages with collapsible folder navigation, providing a cleaner interface and more screen real estate for data visualization and analysis. [CCM-26679]

December 2025 - BI Dashboards Release​

AWS Billing and Inventory Dashboard Updates:​

• Optimized Tag Filtering: We've introduced an improved approach for filtering data using resource tag values, simplifying queries and reducing load times.
• Optimized Filter Fields: Users can filter AWS billing data by one or more key/value pairs using the "Resource Tags: All Values" fields, which have been optimized for better performance
• Efficient Cost Grouping: For use cases requiring cost grouping by tags, queries can now group by one or more key/value combinations efficiently
  • Note: It is recommended to limit dashboards to a maximum of 10 tiles to ensure optimal rendering and usability

When you need to filter AWS billing data by one or multiple key/value pairs without grouping, use the new filter fields under Resource Tags: All Values.

AWS Explore Updates:​

• Key Filters Migration: Key filters have been relocated from labels to labelsv2 for improved organization and performance. This migration includes naming convention changes for the keys - refer to the example image below for details. For more details regarding Label V2, see Label V2.
• Dimension Changes: Keys/values have been moved from dimensions to filter-only fields
  • Note: As a result of this change, grouping by keys/values in the results will no longer be possible

AWS RDS, AWS EBS, and AWS EC2 Inventory Explore Updates:​

• Dimension Changes: Keys/values have been moved from dimensions to filter-only fields
  • Note: As with AWS Explore, grouping by keys/values in the results will no longer be possible

November 2025 - Version 1.71.3​

Deployment Date: November 25, 2025 (Prod-1)​

Feature Improvements​

• Realized Savings from Governance Enforcements: We’ve added support for showing realized savings for enforcements. On both the enforcement list page and the enforcement details page, monthly savings will now be displayed for non–dry run enforcements. [CCM-27472]

Fixed Issues​

• Anomaly Criticality Classification: We've enhanced the anomaly detection system to correctly flag anomalies based on their actual cost impact. This update ensures that high-impact anomalies are properly classified as critical, providing more accurate severity indicators for better cost management decisions. Note that this fix applies to new anomalies only; previously detected anomalies will maintain their original classification. [CCM-26720]

November 2025 - Version 1.70.0​

Deployment Date: November 18, 2025 (Prod-1)​

Feature Improvements​

• Cluster Orchestrator Deletion: We have added a deletion flow for Cluster Orchestrator, enabling users to fully remove all orchestrator resources and revert clusters via API, script, or Helm uninstall. This operation is only available for disabled clusters. [CCM-26694]

• Recommendations Cost Settings: We have added support for configuring AWS cost type settings (Net Amortized or Unblended) for EC2 recommendations. These settings inherit from your Account Settings defaults but can be overridden specifically for this feature. This feature is currently behind a Feature Flag. Please contact Harness support to enable it for your account. [CCM-27040]

Fixed Issues​

• Shared Cost Buckets in Budgets: Added support for shared cost buckets in budget calculations, enabling more comprehensive and accurate budget management. [CCM-24034]

November 2025 - Version 1.69.2​

Deployment Date: November 10, 2025 (Prod-1)​

Feature Improvements​

• Enhanced GCP Discount Preferences: We've expanded the single GCP discount preference into five distinct options for more precise cost management
  • Subscription Credits: Corresponds to SUBSCRIPTION_BENEFIT which are credits earned by purchasing long-term subscriptions to services in exchange for discounts.
  • Sustained Use Discounts: Corresponds to SUSTAINED_USAGE_DISCOUNT.The sustained use discounts credit type is an automatic discount that you receive for running eligible Compute Engine resources for a significant portion of the billing month, with no commitment required.
  • Resource-based CUD Credits: Corresponds to COMMITTED_USAGE_DISCOUNT. Resource-based committed use contracts purchased for Compute Engine in return for deeply discounted prices for VM usage.
  • Legacy Spend-based CUD Credits: Corresponds to COMMITTED_USAGE_DISCOUNT_DOLLAR_BASE. For legacy spend-based committed use discounts (CUDs) that aren't part of the new pricing model, this is the credit earned in exchange for your commitment to spend a minimum amount for a service in a particular region.
  • Other Discounts: Corresponds toDISCOUNT. The discount credit type is used for credits earned after a contractual spending threshold is reached. Note that in the Cloud Billing reports available in the Google Cloud console, the discount credit type is listed as Spending-based discounts. For existing perspectives with the previous Include Discounts preference enabled, all five new preferences will be automatically enabled. You can adjust these individually according to your reporting needs.

Complete GCP Preferences Categories:

• Savings Programs

  • Legacy spend-based CUD credits
  • Resource-based CUD credits

• Other Savings

  • Promotional credits
  • Sustained use discounts (SUDs)
  • Spending-based discounts
  • Subscription credits
  • Negotiated savings

• Invoice Level Charges

  • Tax

  [CCM-24740]

• Improved Anomaly Alert Email Validation: Added UI validation for email format when creating anomaly alerts, preventing invalid email submissions. [CCM-25070]

• Improved Permissions for Governance: Previously, users with granular permissions on Governance Alerts couldn’t open the alerts drawer because it checked for global view permissions. This issue has now been resolved. If a user doesn’t have access to any alerts, the drawer will now display an appropriate error state indicating that. [CCM-26473]

• **Persistent behaviour across Perspective Rules **: Previously, when adding new rules, the Perspective’s default group-by and time range values were being reset to their defaults. This has now been fixed — they will persist across rule changes. [CCM-25493]

• Folder ID Now Mandatory for Budget Creation via API: To ensure better organization and visibility of budgets in the UI, the folderId field is now mandatory when creating budgets through the API. Previously, budgets created without a folder ID did not appear in the UI, leading to inconsistencies in reporting. With this update, any API request that omits the folderId will return an error. This change ensures that all budgets are created under the correct folders, improving overall organization, visibility, and reporting accuracy. Users integrating with the API should update their automation or scripts to include the folderId parameter to avoid errors during budget creation. [CCM-27057]

October 2025 - Version 1.68.9​

Deployment Date: October 21, 2025 (Prod-1)​

• Enhanced Perspective Reports Delivery Options: [CCM-26936]

  • Download as CSV: Instantly export reports in CSV format for quick, ad-hoc analysis. Supports all time range options.
  • Create Scheduled Report: Automate recurring deliveries to up to 50 recipients with selected time ranges. Learn more

  Note that, Download as CSV supports all time range options but scheduling is limited to some time range.

• Improved Anomaly Detection in Perspectives: Removed the 'Apply Filter' functionality from anomalies tooltip. Anomalies now correctly honor the perspective data aggregation value. [CCM-26647]

• New Filters in Cloud Integration: We have added new filters on Account Settings > Integration for Cloud Cost.

  • For Kubernetes Cluster: Connectivity Status, Features Enabled
  • For Cloud Accounts: Connectivity Status, Features Enabled, Cloud Provider, AWS Account ID(s), Azure Tenant ID, Azure Subscription ID, GCP Project ID [CCM-17003]

October 2025 - Version 1.67.2​

Deployment Date: October 22, 2025 (Prod-1)​

Feature Improvements​

• Streamlined Autostopping Rule Creation: CCM has introduced a use case-based setup with intuitive Traffic-based and Schedule-based mode selection. The interface dynamically adjusts to display only relevant configuration fields, simplifying the setup process and improving user experience. [CCM-24600]

• AWS Service Detail (Beta): CCM has added a new beta field in AWS Explore that provides granular insight into mixed services within the CUR. This feature delivers standardized service categorization by mapping service codes, product families, and usage types. It properly identifies mixed services (such as EBS, VPC, and EMR under EC2) and includes fallback logic to ensure comprehensive coverage across all AWS usage patterns. [CCM-25768]

Fixed Issues​

• Anomalies Table Sorting: The default view of the anomalies table now sorts by "Cost Impact" in descending order. [CCM-26665]
• Perspective Page Improvements: CCM has removed the 'Apply Filter' functionality from the anomalies tooltip on the Perspective page. Anomalies now correctly honor the perspective data aggregation value. [CCM-26647]

October 2025 - Version 1.66.0​

Deployment Date: October 15, 2025 (Prod-1)​

[New Feature] Bulk Management and Filtering in Budgets​

[CCM-24072] | Docs

This release introduces two powerful features to streamline budget management:

Bulk Management enables efficient budget administration by allowing you to:

• Select and modify multiple budgets simultaneously
• Adjust budget amounts by percentage or fixed value
• Manage alerts across multiple budgets (add/remove recipients, delete alerts)
• Clone budgets with customizable options for thresholds and recipients

Saved Filters improves budget discoverability with filters for:

• Created By
• Last Modified
• Max/Min Budget Amount
• Period
• Perspective

All filter dropdowns are now context-aware, showing only options relevant to your budgets.

[New Feature] Ticketing Tool Mapping​

[CCM-25483] | Docs

We've introduced Jira integration features to streamline your recommendation workflow:

1. Status Mapping with Resolution Codes: Earlier, we introduced the ability to automatically update recommendation statuses based on Jira ticket status changes. Now, you can also choose Resolution Codes that will automatically update recommendations to Applied status.

2. Default Project Mapping: Route recommendations to the right teams by mapping cost categories to specific Jira projects. When creating tickets, projects are auto-selected based on the resource's cost category, ensuring recommendations reach the appropriate stakeholders without manual routing.

Feature Improvements​

• Disable Functionality: Added support for disabling Cluster Orchestrator on enabled clusters. Disabling a Cluster Orchestrator stops all its components, removes finalizers and configurations, disables Harness-managed nodepools, and prepares the cluster for Karpenter (or any autoscaler) fallback. [CCM-25739]

• Enhanced Recommendation Tags: Added a "tags" field to each data item in the API response for the Recommendations overview list endpoint (/ccm/api/recommendation/overview/list) which contains all of the cloud tags associated with each recommendation. If no cloud tags exist for a Recommendation, "null" is given instead. [CCM-25737]

• Enhanced Ticket Management for ServiceNow: Added support for "Service Request" ticket type for our ServiceNow integration for recommendations. Using Service Request tickets, customers can manage recommendations more easily with automated workflows and auto-discovery. Ticket lifecycle can be managed using proper fulfillment and approvals.

• Vanity URL Support: Added support for vanity URL redirection in budget email alerts. Earlier, alerts for customers using vanity URLs redirected to app.harness.io, which was invalid. They now correctly redirect to the customer’s vanity URL. [CCM-25920]

Fixed Issues​

• Bar Chart Fix: The Harness CCM Overview page bar chart duplicated Azure costs due to missing filters for actual vs amortized values. This issue was resolved by fetching azureCostType from the Active Spend API and applying it as a condition. [CCM-23539]

[New Feature] Budget Folders​

October 2025 - Version 1.65.9​

Deployment Date: October 8, 2025 (Prod-1)​

[New Feature] Budget Folders​

[CCM-24072] | Docs

We are introducing Budget Folders to improve usability and streamline budget management. Budgets are now automatically associated with the Perspective folder under which they were created. This provides a consistent folder structure across both Perspectives and Budgets.

The Budgets tab displays all the budgets in a new consolidated view. Users can drill down into a folder or access the complete budget list for broader visibility. Users can also select which columns they want to see in the Budgets list. This gives them the flexibility to refine their view with the following attributes: Time Period, Budget Amount, Current Spend, Forecast, Forecast vs. Budget, Current Spend vs. Budget, Thresholds, Created By, and Last Modified.

These changes improve the overall usability and discoverability of budgets, allowing users to identify and organize budgets more efficiently.

Feature Improvements​

• Resource ID Support in Perspectives: Added "Resource ID" as a group by, filter, and operand across all cloud service providers, enabling more granular cost analysis and reporting. Resource ID - AWS Global Resource Name - GCP

Added for the following scenarios: Perspective Operand definition Group by Support by Resource ID Filter support by Resource ID

These two new fields excluded from cost categories. [CCM-25078]

• Azure Resource Group Filtering: Added support for filtering recommendations by Azure Resource Group, aligning with Perspectives functionality and enabling more targeted cost optimization strategies. [CCM-25175]
• AutoStopping Enhancements: Simplified health check configuration by removing status enforcement requirements, allowing single status inputs. We have also stopped using the instance groups to populate the port configuration in the rule creation flow. Now user can fill the ports based on their usecase. [CCM-24597]

October 2025 - Hotfix: Added new columns for Recommendation CSV Export​

Deployment Date: October 3, 2025​

What Was Changed

CCM has added new columns in CSV of recommendations when you export CSV, these new columns are:

• Recommendation Link: Direct URL to view the recommendation details in the CCM portal
• Recommendation Date: Date when the recommendation was generated
• Account ID: Unique identifier for the cloud account
• Account Name: Display name of the cloud account
• Resource ID: Unique identifier for the resource
• Resource Name: Display name of the resource
• Cloud Provider: Cloud service provider
• Region: Geographic location where the resource is deployed
• Resource Type: Category of cloud resource
• Cost Category: Classification of cost for the resource
• Tags: Resource labels used for organization and filtering
• Recommendation State: Current status of the recommendation
• Resource Count: Number of resources affected by the recommendation
• Custom Recommendation: Only for Governance Recommendations if they are not OOTB
• Applied At: Date and time when the recommendation was applied (only for applied recommendations)
• Applied By: User who applied the recommendation (only for applied recommendations)
• Overridden Savings: Savings amount that has been manually adjusted for applied recommendations

September 2025 - Version 1.64.7​

Deployment Date: September 23, 2025 (Prod-1)​

Feature Improvements​

• Bin Packing Notification Enhancement: If Bin Packing is disabled for a cluster within the Cluster Orchestrator, a notification banner and confirmation dialogue will appear upon saving, indicating that “Bin Packing must be enabled for cluster scale-down and optimal node utilisation.” [CCM-25324]

• Mandatory Fields for Bin Packing: With the enabling of Bin Packing, the fields for Node Deletion Criteria and Delay have now been made mandatory. [CCM-25312]

• JIRA Integration for Recommendations: We have addressed an issue regarding removing JIRA tickets from the shared recommendations. An API is implemented to unlink recommendations from the associated JIRA tickets. [CCM-25455]

• AutoStopping Schedule Ordering Fix: We fixed an issue where when AutoStopping schedules from the Backend were unordered (e.g., 4, 1, 2, 3), it caused the first schedule on UI edit to overwrite the wrong one. We now sort the Backend response before iterating or updating items to prevent mismatches. [CCM-25489]

• AutoStopping Timezone Display Correction: We have resolved an issue where the savings timestamp in AutoStopping Rules was coming from the BE in UTC while the UI was plotting them against the user's local timezone, causing mismatches. The chart now shows the savings in the user's local timezone.[CCM-25450]

• Cluster Orchestrator Setup Flow Improvement: We have made a minor improvement to the setup flow of the Cluster Orchestrator. After selecting Cluster Preferences, the workflow will now proceed to Spot Preferences, followed by Replacement Schedules, and finally to the Finish stage. This adjustment is intended to prevent any oversight in configuring essential settings during the setup process. [CCM-25325]

• Governance Rule Permissions Fix: Previously, the Save button in the Governance Rule Details page did not check for granular edit permissions regarding the rule. Consequently, the button was disabled if the user lacked global permissions. This issue has now been addressed. [CCM-25304]

• Event Logs Performance Enhancement: We have implemented pagination and sorting capabilities in the Event Logs in Cluster Orchestrator to enhance performance when managing large lists. [CCM-24277]

September 2025 - Version 1.63.3​

Deployment Date: September 12, 2025 (Prod-1)​

Feature Improvements​

• Enhanced Cluster Configuration UI — Added informative tooltips for disabled fields in the Cluster Orchestrator Enablement flow. These tooltips clearly explain prerequisites and requirements needed to enable each configuration option, improving user experience and reducing confusion during setup. [CCM-25141]

• Consolidated CSV Export — Consolidated CSV export functionality on the Perspective Details page. Users now have access to a single, comprehensive CSV format that includes all relevant data in a well-structured format. If you wish to use the older CSV format, please do so using Perspective API. [CCM-24775]

Bug Fixes​

• Fixed LIKE Operator — With the recent revamp of the Perspective Rule Builder UX, the LIKE operator was not always working as expected. This issue has now been fixed, and the Perspective Preview will correctly display data when available. [CCM-25271]

September 2025 - Hotfix: Dynamic Cost Categories Toggle in Perspectives​

Deployment Date: September 10, 2025​

[New Feature] Dynamic Cost Categories Toggle in Perspectives​

[CCM-24072] | Docs

We’ve added a new Dynamic toggle on the Perspective page that gives you control over how cost category rules are applied:

• Dynamic ON → The system evaluates cost category rules dynamically at query time. This approach is best suited for testing scenarios where you want to understand how changes in rule definitions impact cost attribution. Please note that performance may vary based on the volume of cost data and the selected time range; in some cases, evaluations may be slower.
• Dynamic OFF → Perspectives use pre-computed cost category rules for cost attribution. Since CCM ingests cost data once per day, cost category rules are evaluated during ingestion and then persisted in the database. This approach significantly improves load times compared to the Dynamic ON option. The system also ensures that the data displayed is consistent and accurate across both modes. When data ingestion is in progress and the Dynamic OFF option is selected, queries will temporarily fall back to the dynamic mode to maintain continuity.

By default, Perspectives will have Dynamic Toggle as OFF.

Note: This feature is behind a Feature Flag. If you wish to enable this feature for your account, please contact Harness Support.

August 2025 - Version 1.62.3​

Deployment Date: August 29, 2025 (Prod-1)​

[New Feature] Jira Status Mapping in Recommendations​

[CCM-23844] | Docs

We’ve added support for Jira Status Mapping in Recommendation Settings. Users can now map Jira statuses so that recommendations are automatically moved to either the Applied section or the Ignore List, based on their selection.

Noted that you need to have a Jira connector configured successfully for this feature to show up in UI and it takes about an hour for the changes to reflect.

[New Feature] Cluster Capacity Limits​

[CCM-24203] | Docs

We have added Cluster Capacity Limits in Cluster Orchestrator’s Cluster Configuration. For Karpenter Nodepools, users can now set maximum CPU (cores) and memory (bytes) limits as guardrails to prevent uncontrolled cluster scaling.

Feature Improvements​

• Draft Indicator for Perspective Creation: All Perspectives created from now will have a "Draft" prefix to their name in the Perspective Builder to indicate that the Perspective creation is still in process.

• Enhanced Policy Evaluation Transparency We have added an Evaluated Rule tab in the Evaluation Output to display the policy YAML used for evaluation. An option "Include custodian-metadata.json" has been included in the bulk export functionality to export the metadata of the policies evaluated.

Bug Fixes​

• Chart Visualization Improvement: We recently switched the chart's Y-axis from logarithmic to linear based on feedback. However, this made it difficult for users to view both cluster and cloud costs together, as one would get overshadowed unless the other was deselected. We've reverted the change for now while we work on a better solution to visualize both.

• Azure VM Recommendations Field Mapping: In the Azure VM recommendations, the namespace and clusterName fields were mistakenly mapped in reverse. This mismatch caused the unique combination to fail to align with the ignore list, resulting in items remaining unexcluded.

• GCP Connector Command Fixes: Previously, for Cloud Costs GCP connector, the command to create a custom role for AutoStopping did not have backslashes for new lines, which caused the command to fail. We have also added compute.snapshots.list and compute.disks.list permissions by default as they are used for visibility.

August 2025 - Version 1.61.1​

Deployment Date: August 22, 2025 (Prod-1)​

[New Feature] Dynamic Perspective Reports​

[CCM-23836] | Docs

CCM is introducing a new capability Dynamic Perspective Reports, a powerful way to generate, schedule, and manage cost reports directly from your Perspectives. Create reports from your perspectives to bookmark specific filter and grouping configurations. No need to rebuild the same view repeatedly — just save it once and access it anytime.

Key Highlights

• Create Reports from Perspectives with custom grouping, filters, time periods, and columns.
• Flexible Delivery Options: download instantly or schedule recurring deliveries to up to 50 recipients.
• Centralized Management: access, edit, or delete all saved reports under Cloud Costs > Perspectives > Saved Reports.
• Email Subscriptions: subscribe/unsubscribe from scheduled report deliveries anytime.

This feature is rolled out behind a Feature Flag. If the flag is enabled for your account, report creation via the Perspective creation flow will be disabled.

August 2025 - Version 1.60.1​

Deployment Date: August 18, 2025 (Prod-1)​

Feature Improvements​

• Revamp Perspective Rule Builder UX: We’ve revamped the UX for the Perspective Rule builder. The time range selection in Perspective creation is now on the right side near the chart, and the Rule Name and Folder fields are moved into a separate modal. This updated rule builder is also used in Cost Categories. [CCM-24935]

Bug Fixes​

• Permission Handling Improvement: We have fixed an issue with folder permission enforcement. Previously, users with read-only access could sometimes edit and move perspectives from one folder to other provided the destination folder allows create/edit permissions. This update ensures proper permission validation across all folders. [CCM-24670]

• Container Cost Display Enhancement: Previously, when a container's lastDayCost was unavailable, recommendations displayed 'NaN' (Not a Number) values in the UI. We've implemented fallback logic that automatically retrieves the lastDayCost, ensuring consistent cost visibility across all recommendations. [CCM-23235]

August 2025 - Hotfix: Dashboard Cost Totals Issue​

Deployment Date: August 17, 2025 (Prod-1)​

What Was Fixed

• We addressed an issue where cost totals in Dashboard reports did not always align with the detailed breakdowns. When viewing cluster costs by specific categories (such as namespace ), the “Total” row could sometimes differ from the sum of the detailed rows. This occurred because filters were applied differently to totals versus detailed costs.

What has changed now?

• The “Total” row now always matches the sum of the detailed rows.
• Reports remain reliable across all views whether filtered by namespace, labels, or other categories.

Note: This was due to a display issue. The actual cloud costs and allocations did not get affected.

August 2025 - Version 1.59.1​

Deployment Date: August 11, 2025 (Prod-1)​

⭐ [New Feature] Rules Generating Recommendations​

[CCM-24188] | Docs

• Added support for Rule and Target Account Exclusions in Governance Recommendations, enabling you to define custom default rules that apply globally or to specific account subsets. This enhancement gives you precise control over which governance policies apply to specific accounts and which rules generate recommendations, streamlining compliance management across your organization.

Feature Improvements​

• Governance Rule Operators: Updated op operator type from less-than to greater-than for improved rule evaluation accuracy. [CCM-24446]

• Budget Forecasting Enhancement: Added support for overriding forecast values in yearly budgets with monthly breakdown, providing more granular budget management capabilities. Also, improved the graph to show the forecasted values if override is enabled. [CCM-24440]

• New Feature Flag: Introduced a new feature flag which, when enabled for specific accounts, will exclusively bypass any cost calculations using public pricing sources. To enable the feature flag, please contact Harness Support. [CCM-22370]

• Enhanced Node Pool Recommendations Display: We've improved the Node Pool Recommendations listing page to show cloud account identifiers, providing better context for multi-account environments. For AWS, you'll see the Account ID; for GCP, the Project ID; and for Azure, the Subscription ID. This enhancement helps you quickly identify which cloud account each recommendation belongs to. [CCM-24291]

Bug Fixes​

• Anomaly Email Links: Fixed an issue where the "Investigate" button in anomaly alert emails was resulting in a 404 error. The problem occurred with anomaly alerts based on all data that had no associated perspective, causing malformed redirection links. [CCM-24542]

• Anomaly Access Control: Resolved an issue with anomaly visibility filtering where users with no access to specific anomalies could see all anomalies in the system ensuring proper access control enforcement. [CCM-24353]

• Connector Health Check: Improved error messaging for storage bucket connectivity issues. When a bucket doesn't exist, users now receive a specific error message instead of the generic "Unknown error occurred" message. [CCM-20815]

• Perspective Bookmarks: When a user accessed the perspective using a bookmarked URL containing filters, CCM wasn't prioritising the URL filters over those stored in local storage. We have fixed this issue. [CCM-24726]

July 2025 - Version 1.58.3​

Deployment Date: July 28, 2025 (Prod-1)​

[New Feature] Bulk Evaluation Export​

[CCM-23906]

CCM has introduced a new "Export" option in Cloud Asset Governance that lets you easily download results and logs from multiple evaluations run across different accounts and regions directly from the Test Terminal (for multi-target evaluations) or the Evaluations page.

You can now export a ZIP archive containing:

• metadata.json
• resources.json
• custodian-run.log
• actioned-resources.json

Key Highlights:

• Export results and logs from multiple evaluations at once
• Downloads as a single ZIP file
• Available when all evaluations are in terminal states (Succeeded or Failed)
• Supports up to 100 evaluations per export

Feature Improvements​

• Monthly alerts for Yearly Budgets: Define a spend threshold for each month within an yearly budget and receive alerts as soon as that month’s spend approaches the limit. [CCM-24187]

Available when:

• Budget Period: Yearly
• Budget Breakdown: Monthly

• GCP subscription-benefit credits now included : We have added support for GCP subscription-benefit credits in Perspectives. This ensures that any credits applied through GCP subscription-based benefits are now accurately identified and included. [CCM-24197]

• Perspective Preferences Persistence: We've introduced automatic saving of Perspective preferences, eliminating the need to reconfigure each time. This includes (but not limited to):

  • Group By selection
  • Filters applied
  • Cost preference
  • Cost granularity When preferences are selected in a Perspective, those settings are saved automatically. Upon returning to the same Perspective, the previously selected preferences are reapplied. Once the user logs out, all view preferences stored in the cache will be cleared. [CCM-24210]

Bug Fixes​

• Perspective search with parentheses: Fixed an issue where Perspective searches containing parentheses would fail. [CCM-24027]
• Perspective exports match the grid: In the last release, we added support for aggregation in Perspective exports. However, this led to issues where the Perspective grid export displayed incorrect data. This has now been resolved to ensure the exported data matches the grid view accurately. [CCM-24328]
• Budget search works on every page: In case Budget Cascades is not enabled, we use a paginated API. Previously, if the user navigated to a later page and then searched for a budget, the page number (offset) was not being reset, which caused the search to fail. This issue has now been fixed and the search will work correctly regardless of which page the user initiates it from.. [CCM-24083]
• Accurate budget alerts: Fixed early triggers;alerts now fire only when spend truly exceeds your set threshold. [CCM-24053]

July 2025 - Version 1.57.1​

Deployment Date: July 18, 2025 (Prod-1)

Feature Improvements​

Perspectives​

• Line chart view: Perspectives now displays a Line chart instead of an Area chart, eliminating redundancy with the stacked bar view. [CCM-23889]

• CSV aggregation fix: CSV exports from Perspectives now respect the selected aggregation setting, matching the on-screen view. [CCM-20404]

• GCP credits regex support: Perspective queries now support regex-based matching for GCP discount types aligned with dashboard behavior. [CCM-24110]

• Tooltip enhancements: The Group By Chart tooltip now shows the total daily cost plus individual data-point costs, with the selected point highlighted for easier analysis. [CCM-23888]

Cost Categories​

• Cluster Cost Category (CC) Stamping Support: We now support Cost Category for Cluster data. Set Billing Source to CLUSTER to use Cost Category as a filter or dimension in Dashboards. This capability applies only to data generated after 10 July 2025 (submit a support ticket to backfill older data). You can also create Cost Buckets based on cluster-level rules. View cluster Cost Categories in Unified Explore rather than Cluster Explore. [CCM-22980]

AutoStopping​

• Connector permission checks: The AutoStopping rule flow now validates that the selected connector already has the granular permissions required to create the target gateway resource (ALB, Azure Application Gateway, or Harness AutoStopping Proxy). For detailed permission requirements, see the granular-permissions sections for each cloud provider: AWS, Azure, and GCP. [CCM-21579]

Recommendations​

• Readable ServiceNow timestamps: Applied recommendations linked to ServiceNow tickets now show human-readable date-time values, making audit trails easier to follow. [CCM-24082]

• Tag filter search: Recommendation filters now list all available cloud tags and provide search to quickly find specific tags. [CCM-23792]

Asset Governance​

• Enforcement evaluation cap increased: The previous limit of 30 Rules or Rule Sets per enforcement has been removed. Each enforcement can now run up to 10,000 evaluations (Rules × Accounts × Regions). [CCM-23953, CCM-21995 ]

Connectors​

• AWS connector role retention: When updating an AWS connector from the default authentication type to OIDC, the Role field is now preserved instead of being set to undefined. [CCM-23765]

• Single CCM K8s connector per CD connector: Each CD Kubernetes connector can now be linked to only one CCM Kubernetes connector; the UI enforces this limit during setup. [CCM-18900]

July 2025 - Version 1.56.3​

Deployment Date: July 7, 2025 (Prod-1)

Feature Improvements​

• Jira Integration Optimization: We have improved Jira ticket creation performance for recommendations by optimizing API call timing. Jira projects now load immediately and API call frequency is reduced, improving overall performance and backend load. [CCM-23837]

• Anomaly Detection Filter Support: We have added support for filter inheritance when navigating from Perspectives to Anomaly Detection (V2) screens. Anomaly results now respect all filters applied on the source Perspective page.[CCM-23544]

• Increased Folder Limit: We have increased the maximum number of folders that can be created from the previous limit of 500 to 2,000. [CCM-23784]

• New GCP “Include Promotions” preference:

  • Located in Perspective Preferences, this toggle includes promotional credits (for example, Free Trial or marketing credits) in cost calculations when enabled.
  • Promotions are no longer bundled with the regular Discount option; enable this preference separately if you want promotional amounts reflected. [CCM-20075]

Bug Fixes​

• Governance Recommendation Unignore Options: Previously, the Recommendations list menu showed multiple options for unignoring a Governance recommendation. We’ve now fixed this by showing a single option to unignore it at the same level it was ignored, and we’ve also corrected the payload that’s sent when unignoring. [CCM-23591]

• Budget Growth Rate Display Logic: We have fixed budget creation workflow to properly handle growth rate visibility. When creating a monthly budget, users will not see the growth rate option. [CCM-22148]

• Cost Category Filter Logic: We have fixed an issue with compound filter conditions where cost category filters were not properly handled when no matching recommendations existed. The fix ensures that when no recommendations match the cost category criteria, the entire compound condition properly evaluates to false, maintaining logical consistency in filter operations. [CCM-23645]

June 2025 - Version 1.55.3​

[New Feature] OIDC Authentication Support​

[CCM-23638] | AWS OIDC Documentation | GCP OIDC Documentation

Harness CCM now supports OpenID Connect (OIDC) authentication for enhanced security and streamlined connector setup. This authentication method is available for the following cloud providers and features:

AWS Support:

• Asset Governance
• Commitment Orchestration
• AutoStopping

GCP Support:

• Asset Governance
• AutoStopping

[New Feature] Event Driven Anomaly Detection​

[CCM-22730] | Docs

CCM now provides event-driven anomaly detection that triggers automatically when cost data is ingested, complementing the existing scheduled anomaly detection jobs.

This feature provides immediate detection by running anomaly analysis instantly when cost data becomes available, enabling faster identification of cost anomalies. The feature is available for AWS, GCP, and Azure cloud accounts.

Feature Improvements​

• Cost Category Management Improvements: Cost category name updates now propagate automatically across all usage locations, with changes reflected in CCM Perspectives and nested cost category rules. [CCM-23698]

• Commitment Setup UI Refinements: We have removed the Potential Savings widget and the Current Coverage section from the Commitment Setup Review screen. These changes eliminate generic data that didn't reflect user-specific selections from previous configuration steps, providing a more streamlined and relevant user experience. [CCM-23604]

June 2025 - Version 1.54.5​

⭐ [New Feature] Granular AutoStopping Permissions Support for Azure and GCP​

[CCM-21574, CCM-21575] | Docs for Azure | Docs for GCP

We have introduced Granular permissions support for Azure and GCP Autostopping. With this update, users can now select the specific resource types they want to enable for Autostopping such as virtual machines and instance groups. Based on the selected resource types, only the minimal required set of permissions will be requested. This feature simplifies onboarding, and aligns with security best practices.

Granular AutoStopping Permissions Support for Azure:​

Granular AutoStopping Permissions Support for GCP:​

Feature Improvements​

• RBAC Support for Nested Cost Categories: We've extended Role-Based Access Control (RBAC) to support nested cost buckets within CCM data scope, providing more granular access control for nested Cost Categories. Note that RBAC support is not available for the Unallocated Cost bucket. [CCM-23579]

Bug Fixes​

• Fixed Duplicate Discounts with Net-Amortised Cost: Resolved an issue where duplicate discounts were applied when net-amortised cost was selected with discounts enabled. Improvements include:
  • When creating or updating a perspective with net-amortised cost, discounts are automatically set to false during save/update
  • The discount field is now greyed out when net-amortised cost is selected in perspective preferences as discounts are already included in the net-amortised cost.
  • Backend validation prevents duplicate discounts when using the API, ensuring accurate cost calculations in all scenarios [CCM-22152]

May 2025 - Version 1.53.3​

Bug Fixes​

• Missing Name in Enforcement Update Toast: There was an issue where the name didn’t appear in the toast message after updating an enforcement rule. This has now been fixed, and the name will display correctly. [CCM-23256]

• Error Toast on Navigating Away from Perspective Details: Navigating away from the Perspective Details page before it fully loaded triggered an error toast due to an aborted API call. This behavior has been fixed, and users will no longer see this error toast when navigating quickly.[CCM-23244]

May 2025 - Version 1.52.1​

Feature Improvements​

• New creation flow for AutoStopping Load Balancers and Proxies: We’ve revamped the creation flow for Load Balancers and AutoStopping Proxies by moving it to a side panel for better input organization. The Custom URL is no longer a required step for Load Balancer creation—legacy LBs will still show it as a static field, and edits must be done via Route53. We’ve also introduced a new Internal/External toggle for ALBs to define access type. Additionally, there’s now a toggle for Encrypted EBS volumes. [CCM-21226]

• Sorting for Clusters List: The Clusters List in Cluster Orchestrator Overview is now sorted with enabled connectors first (alphabetically), followed by AWS connectors that are not enabled (alphabetically), and then the rest. [CCM-23021]

• Improvements in Budgets and Perspectives home page: We have now added support for showing "Created By" and "Modified By" in Perspectives and in Budgets. [CCM-22932]

• Standardized Region Format for Recommendations: We are now normalizing region formats across all recommendations. Previously, EC2 recommendations used formats like US_EAST_1, while others used us-east-1, leading to inconsistencies. Going forward, only the standardized us-east-1 format will be used, ensuring filters work consistently across all recommendation types. [CCM-22998]

• Cluster Orchestrator UI changes: We’ve now moved the Base On-demand setting below the Spot/On-demand split. Additionally, if the On-demand percentage is set to 100%, the fields below will be disabled. [CCM-22759]

Bug Fixes​

• Improved Cloud Asset Governance Rule Saving Experience: We've resolved an issue where saving a new Cloud Asset Governance rule incorrectly triggered a confirmation dialog to stay or leave the page. This dialog will now only appear when there are unsaved changes. [CCM-22995]

May 2025 - Version 1.51.4​

⭐ [New Feature] Replacement Schedules​

[CCM-21859]

We have added support for replacement schedules with options: Always, Never, or Custom. Users can now define specific windows (e.g., Tuesdays and Fridays, 9 PM–11:59 PM IST) for bin packing operations. Currently this is only available for "Harness Pod Eviction". During these scheduled periods, node replacements and system updates may occur, temporarily affecting workload availability. So it is advisable to choose time windows that minimize impact on critical operations.

Feature Improvements​

• Added validation for Cost Category rules during save and update operations: The system now constructs and dry-runs queries based on the defined cost bucket rules to identify any invalid rules before saving. As a result of this additional validation step, you may experience a slight delay during Cost Category save or update actions. [CCM-21901]

• Added support for quarterly reports in Perspectives: Previously, only monthly and annual reporting options were available. With the new quarterly option, users can now receive more balanced and actionable insights per quarter.

• Nodepool Recommendations UI changes: "Largest Resource Requirements" has been renamed to "Minimum Resource Requirements".The positions of Preferred and Minimum resource requirements have been swapped in the UI for better clarity and consistency. Preferred resource requirements will always be greater than Minimum resource requirements. We have also added tooltips on the UI to make the UI more user-friendly. [CCM-22826]

• Traffic analysis chart on resource level for AutoStopping Rules: We’ve introduced a new interactive chart to help you better understand resource activity and idleness over the past 30 days. It is available on every AutoStopping rule page to make it easy for users to analyze trends at a glance.

• More Options for Date Ranges: We’ve added support for additional shorter time ranges on the Governance Evaluations page, including:
  • Last 1 Minute
  • Last 5 Minutes
  • Last 10 Minutes
  • Last 15 Minutes
  • Last 30 Minutes
  • Last 45 Minutes
  • Last 1 Hour
  • Last 3 Hours
  • Last 6 Hours
  • Last 12 Hours
  • Last 1 Day
  • Last 2 Days
  • Last 7 Days
  • Last 30 Days.

• Editing Instance Families: We’ve added support for editing Instance Families for regions that have already been added in the create/update Nodepool preset flow.

• Validation Enhancements for Cost Categories: To prevent naming conflicts we’ve added new validations following which Cost Buckets and Shared Costs cannot have the same name within the same Cost Category. A validation check has been added to ensure Cost Bucket names are unique within a Cost Category, regardless of case sensitivity. Similarly, Shared Cost names must also be unique within the same Cost Category (case-insensitive check). [CCM-21900]

Bug Fixes​

• The Harness API for Nodepool Recommendations previously did not return the Account ID in the response. This issue has now been fixed, and the API will include the Account ID as part of the response payload. [CCM-21082]

• Improved Accuracy in Cost Category Anomaly Detection: We’ve improved how cost category rules are applied to anomaly filters within Perspectives. Specifically, OR filters were incorrectly being converted to AND filters, which resulted in the expected anomalies not appearing in the Perspective view. We have fixed this and this update ensures that anomalies tied to cost category rules ( combinations of AND/OR conditions) are shown accurately. [CCM-21712]

• UI enhancement for Governance Recommendations: We’ve resolved an issue in the Governance Recommendations’ Resource Breakdown tab where Azure Resource IDs were overflowing into adjacent cells. The text now wraps onto the next line to maintain proper layout. [CCM-22850]

May 2025 - Version 1.50.2​

⭐ [New Feature] Label V2​

[CCM-22075] | Docs

We're rolling out Label V2, a major enhancement to how labels (tags) are handled and displayed across the platform, delivering better visibility, performance, and alignment with cloud-native formats.

Key Benefits​

• Improved Query Performance
• Accurate and original label visibility

⚠️ Action Required

Users will need to manually update their Perspectives, Cost Categories, and Dashboards to use the new LabelsV2 format.

Feature Improvements​

• Preferred Instance Families: We’ve added support for saving Preferred Instance Families in Nodepool presets, along with Buffer and Minimum Node Count. Users can now select specific instance families per cloud provider and region, and these preferences will be considered when generating the next set of Nodepool recommendations. [CCM-21987]

• Granular RBAC for Governance Alerts: You can now assign granular permissions for Governance Alerts to specific resource groups and roles, enabling more precise access control. [CCM-21921]

  For Resource Groups:

  1. Navigate to Account Settings > Access Control > Resource Groups
  2. Select an existing Resource Group or create a new one
  3. Enable the Cloud Asset Governance Alerts permission
  4. Choose between All alerts or Specified alerts for more granular control
  

  For Roles:

  1. Navigate to Account Settings > Access Control > Roles
  2. Select an existing Role or create a new one
  3. Enable the Cloud Asset Governance Alerts permission
  4. Assign specific permissions such as View or Edit/Delete

Bug Fixes​

• We have resolved an issue where, in the Setup HTTP/HTTPS step while creating an AS Rule, clicking the Proxy option first and then closing it would cause the LB option to incorrectly open the Proxy dialog. It will now always open the correct dialog. [CCM-22677]

April 2025 - Version 1.49.1​

Feature Improvements​

• Asset Governance Alert Fix: We've improved the alert update experience to ensure that only relevant resources and cloud accounts are displayed based on the selected cloud provider. This enhancement ensures a more accurate and streamlined configuration flow for Governance alerts. [CCM-22618]

April 2025 - Version 1.48.1​

⭐ [New Feature] Alerts for Governance Rule Evaluations​

[CCM-21921] | Docs

We’ve added Alerts in Cloud Assets Governance, allowing users to configure alerts based on Cloud Provider (AWS, GCP, Azure), Resource Type (Cloud Custodian-defined), Cloud Accounts, Minimum Number of Resources Found, Minimum Cost Impact, Email IDs, and the option to attach evaluation output as a .json file.

Feature Improvements​

• Preferred Instance Families in Nodepool presets: We’ve added support for saving Preferred Instance Families in Nodepool presets, along with Buffer and Minimum Node Count. Users can now select specific instance families per cloud provider and region, and these preferences will be considered when generating the next set of Nodepool recommendations.

• Extend Warmup Period: We've improved the AutoStopping experience by introducing the Extend Warmup Period button. Previously, once a resource was manually started or warmed up via the AutoStopping rule page, users couldn’t extend its uptime during the active warm-up window. With this update, you can now extend the warm-up duration while the resource is still running, eliminating interruptions and giving you more control.

April 2025 - Version 1.47.2​

⭐[New Feature] Granular permissions support for AWS Autostopping​

[CCM-21572]

We have introduced Granular permissions support for AWS Autostopping. With this update, users can now select the specific AWS resource types they want to enable for Autostopping such as EC2, ASG, or RDS. Based on the selected resource types, only the minimal required set of permissions will be requested. This feature simplifies onboarding, and aligns with security best practices.

Feature Improvements​

• Improved Recommendation Tracking: Users can now specify estimated savings when marking a recommendation as applied. Upon marking a recommendation as applied, users can now confirm whether the estimated savings matched the actual savings or enter the actual amount saved if different from the estimate. Additionally, after a recommendation has been applied, this savings data from the Applied Recommendations section can be edited. [CCM-21629]

• Cost Categories Integration for Recommendations: The Filter panel in the Recommendations view now includes the option to filter by Cost Categories. This update is especially valuable for large-scale organizations that manage thousands of recommendations and require structured views to take meaningful action. This improvement allows for efficient sorting and quick isolation of recommendations based on relevant cost buckets and labels. [CCM-21439]

• Recommendations Filter Revamp: The Filter panel in the Recommendations view has been updated to provide a more streamlined experience. As shown in the image below, the new design improves navigation and makes it easier to apply multiple filters. [CCM-21438]

• With the release of granular Autostopping permissions for the AWS connector, only those AWS resources for which the required granular permissions are present in the AWS connector will be available for selection. [CCM-21577]

• Improved Budget Start Date Validation: To ensure accurate budget tracking, we've added validation to prevent the creation of new budgets with a start date later than the 28th of any month. [CCM-22132]

Bug Fixes​

• For Budgets on Perspectives using Cost Categories with shared cost buckets, we’ve fixed an edge case where budget history data was not displaying in the UI. Budget computations are now accurately reflected to ensure complete visibility into historical usage and trends [CCM-22006]

• We’ve enhanced the logic for converting Cloud Asset Governance Recommendation output from JSON to CSV by better handling of edge cases that previously caused issues in the CSV view. [CCM-21834]

Before

After

April 2025 - Version 1.46.2​

⭐ [New Feature] Recommendation Preferences​

[CCM-20954] | Docs

We have introduced a new feature to enhance the personalization of recommendations: Recommendation Preferences. This allows users to create, apply, and save custom tuning preferences to better suit their usage. Users can now also select a default preset preference. This ensures that the chosen tuning preferences are consistently applied across all future recommendations removing the need for repeated manual adjustments.

Feature Improvements​

• New Filters for Autostopping: We have added two new filters in the Autostopping page filter panel:
  • Minimum Savings: This allows users to filter rules based on minimum savings threshold.
  • Hide Disabled Rules: This filter is designed to help users manage and view only the active rules in their system by hiding disabled rules. [CCM-21788, CCM-21766]

Fixed Issues​

• Anomaly Detection Timezone Fix: We have resolved an issue where anomalies appeared with a one-day delay due to a timezone mismatch. The chart implementation has been updated to reflect accurate timestamps. [CCM-21942]
• Cloud Asset Governance Timezone Label: We have fixed a display issue where the "UTC" timezone label was incorrectly shown as "CUT" in enforcement logs. [CCM-21934]

March 2025 - Version 1.45.7​

⭐ [New Feature] (Beta) External Cost Data Ingestion​

[CCM-20954] | Docs Harness Cloud Cost Management now supports External Cost Data Ingestion, enabling you to bring in cost data from third-party vendors using a standardized CSV format. Key Capabilities:

• Ingest and analyze external costs using Perspectives, Budgets, Dashboards and Cost Categories
• Support for FOCUS-formatted CSV files (20MB max per file)

This feature is currently in Beta and behind a feature flag CCM_EXTERNAL_DATA_INGESTION . Please contact our support team for enablement. [CCM-21297]

Feature Improvements​

• Anomaly Comments Support: You can now add and view comments directly on the Anomalies Details page, making collaboration and investigation easier. [CCM-21384]

• Date Range in UTC Format: Date ranges are now processed in UTC for anomaly drilldown chart, ensuring accurate time representation across time zones like IST and PST. This also ensures drilldown chart data is fetched correctly based on the selected time range. [CCM-21630]

Fixed Issues​

• Fixed Incorrect Timestamps in Budget Monthly Breakdown: Previously, the List Budgets API returned monthly breakdowns with Unix timestamps for the year 2024 instead of the configured 2025 values. This issue has now been resolved, and the API returns accurate timestamps consistent with the UI configuration. [CCM-21657]

March 2025 - Version 1.43.5​

Feature Improvements​

• Improved Accessibility for Rule Set Creation: Previously, if a user added multiple Rules or Rule Sets while creating Rule Sets or Enforcements, the button to proceed or save would be pushed off-screen, preventing the action. This issue is now resolved. [CCM-21512]
• Improved Error Handling for AutoStopping Pages: Previously, navigating away from a loading AutoStopping page caused aborted API calls to trigger error toasts. Now, toasts will only appear for valid errors.[CCM-21199]
• Improved UI Behavior for User Dropdown in AutoStopping : Previously, users with access only to AutoStopping encountered an error in the UI due to missing permissions for listing other users. To enhance the experience, we have modified the implementation to hide the user dropdown in filters if the user does not have access to list other users. [CCM-19404]

February 2025​

Version 1.42.4​

New features and enhancements​

• Reverse Fallback Retry in Cluster Orchestrator: We have introduced support for Reverse Fallback Retry in the Cluster Orchestrator configuration. This ensures improved reliability by allowing the orchestrator to retry failed tasks using a reverse fallback mechanism. [CCM-17784]
• Fix for Anomalies Alert Scope Visibility: Some anomalies alerts created from the existing flow were incorrectly assigned an alert scope. Due to this, the alerts were not appearing in the anomalies list. This issue has now been fixed by setting the correct scope, ensuring proper visibility of anomalies alerts. [CCM-21278]
• Confirmation Dialog for Unsaved Cloud Asset Governance Rules: To prevent accidental loss of unsaved changes, a confirmation dialog will now appear when navigating away from an unsaved Cloud Asset Governance rule. [CCM-20972]
• Time-To-Live (TTL) Support for Karpenter Nodes in Cluster Orchestrator: We have introduced Time-To-Live (TTL) support for Karpenter nodes in the Cluster Orchestrator configuration. This feature allows users to automatically terminate idle or underutilized Karpenter nodes after a specified period, optimizing resource usage and reducing costs. [CCM-20051]
• Role-Based Access Control (RBAC) for CCM Side Navigation: Now, the side navigation dynamically adjusts based on a user's permissions. Features such as Overview, AutoStopping, and Governance will only be visible if the user has the necessary RBAC permissions. This ensures a cleaner interface and prevents access to unauthorized sections. [CCM-20789]

January 2025​

Version 1.38.3​

New features and enhancements​

• Granular Governance Recommendations: We have introduced granular governance recommendations, providing greater detail and actionable insights at the individual resource level. This enhancement improves operational scalability by offering more targeted recommendations for both custom and out-of-the-box (OOTB) rules. Users can now specify the scope when adding recommendations to the Ignore List, with options for Rule-level, Subscription-level, or Resource-level exclusions. [CCM-20482]

• New Table View in Output Terminal of Evaluations: The output terminal for evaluations in Asset Governance now includes a new table view, with an option to flatten the table for easier analysis. This update enhances the visibility of governance evaluations and allows users to better manage large sets of data. [CCM-20448]

• Recommendations Insights Tab: Harness CCM now offers a new Recommendations Insights tab in the Asset Governance module, providing users with clear visibility into the evaluation status of each rule. This feature includes detailed insights into the cloud account and region associated with the recommendations. The tab also supports troubleshooting by displaying information on why recommendations may fail, be ignored, or succeed, enabling users to take immediate corrective actions when necessary.

• Extended Idle Time for AutoStopping Rules: The AutoStopping idle time limit in the UI has been extended to allow users to set up to 24 hours (1440 minutes). [CCM-20484]

• Overlapping Schedules for AutoStopping Rules: Users can now define multiple fixed schedules within an AutoStopping rule, with support for overlapping schedules. To determine the final resulting schedule at any given time, users can set the priority order of the schedules using drag-and-drop functionality. [CCM-19196]

December 2024​

Version 1.36.1​

New features and enhancements​

• "Hibernate" Option in AutoStopping Flow: We have introduced the "Hibernate" option for Azure VMs and EC2 instances in the AutoStopping flow, enabling enhanced control over resource state management. [CCM-19907]
• Azure Inventory Metric Data Update: Inventory metric data is now fetched for all Azure VMs that are not in a deleted state, extending beyond just running VMs. [CCM-19960]
• Budget Amount History in Drilldown Chart: The budget drilldown chart now accurately preserves the history of budget amount changes, improving visibility into budget adjustments. [CCM-19494]

Fixed Issues​

• Evaluation List Resource Count Checkbox Persistence: The Resource Count checkbox on the Evaluation List page now retains its previous value correctly after a page refresh in Asset Governance. [CCM-20264]
• GCP Missing Current Daily Costs: We have handled the new invoice.publisher_type column in the GCP Billing export to ensure costs now appear correctly in Perspectives. [CCM-20214]

Version 1.35.6​

New features and enhancements​

1. UI for Governance Custom Savings Percentage: We have added an option to configure and display custom savings percentages in Governance.[CCM-20193]

2. Error Dialog for Governance Rule Validation Errors: The error dialog used for AI-generated policies during rule validations will now also appear for standard rule validation errors, ensuring consistency and clarity. [CCM-20170]

3. UI for Actioned Resources and Savings: Asset Governance now displays the count of actioned resources along with the associated savings. Users can view the "Actioned Resources" tab in the test terminal output for better visibility. [CCM-20108]

4. Account Names in Governance Recommendations Dashboards: we have added support for displaying account names in the namespace field for governance recommendations. Also we extract and expose the account ID in the AWS Account ID field, similar to EC2 or ECS recommendations. We have applied similar functionality for Azure Governance Recommendations. [CCM-19921]

5. Ignore All Recommendations feature: Users can now choose to ignore all recommendations associated with a specific rule, simplifying management of rules. [CCM-19917]

8. Bulk Processing of AutoStopping Rules: We have added support to select multiple AutoStopping rules to enable, disable, or dry-run them simultaneously, streamlining operations. [CCM-19786]

November 2024​

Version 1.34.3​

New features and enhancements​

• UI for Hibernate Support for EC2 and Azure VM: We have introduced the "Hibernate" option in the AutoStopping flow for Azure VM and EC2. This allows resources to enter a hibernation state until a request is received. [CCM-19907]

  

• UX Enhancements for Rule Sets and Enforcements in Asset Governance:
  [CCM-19861] and [CCM-19860] We’ve improved the creation and update flow for the following:

  • Enforcements

• Rule Sets

• Support for Pasting Comma-Separated Values: Users can now paste comma-separated values in account, region, project, and subscription input fields across Azure, GCP, and AWS. [CCM-19837]

• Enhanced Governance Recommendations Details: Asset Governance dashboards now display both Account Name in the Namespace field and AWS Account ID in a dedicated field. [CCM-19921]

Fixed Issues​

• Improved Schedule Creation in AutoStopping UI:
  Users can now directly input numbers when creating schedules for AutoStopping in the UI, eliminating the need to rely solely on arrow keys. [CCM-19425]

October 2024​

Version 1.32.1​

New features and enhancements​

• Ad-Hoc Enforcement Runs: Users can now dry run or execute enforcements directly from the Enforcements list or details page in Asset Governance, improving control over ad-hoc enforcement. [CCM-19438]

• New Resource Name Column in Recommendations View: A new column, Resource Name (No drilldown) has been added to the Governance recommendations view. This allows users to group Governance Recommendations by a specific rule in Dashboards. [CCM-19922]

• AWS Payer Account ID Field Added: We’ve introduced the AWS Payer Account ID in the AWS fields dropdown for Perspectives and Cost Categories Rules. Only account IDs will be shown, excluding account names. [CCM-19843]

• Increased Connector Limit for GCP: We raised the default limit to 6,000 for GCP cross combinations (target account * target region) to prevent unannounced system loads, aligning with the 200 account and 30 region limits in Asset Governance. [CCM-19838]

• ‘Select All’ Option for Filtered Search: We have added a 'Select All' option to the filtered search results in Enforcement creation flow and Testing Terminal in Asset Governance. This option will select the entire filtered list. If the filtered list changes, the 'Select All' option will be deselected. If all filtered options are already selected, the 'Select All' option will be automatically selected. [CCM-19794]

• Date Filter Sync Issue on Evaluation Details Page: Fixed a date filter sync issue displaying "undefined - undefined" on load. Users previously had to manually select a date range, which is now resolved. [CCM-19785]

• Fixed ECS Metrics Chart Issue: CPU recommendations in ECS metrics charts previously displayed incorrect graphs, and CPU usage showed as 0. This has now been corrected. [CCM-19715]

• AutoStopping Controller Update Notification: Outdated AutoStopping Controller versions will now trigger a UI notification, guiding users through the update steps. [CCM-19356]

Version 1.32.1​

New features and enhancements​

• Account Names Now Displayed in Governance Overview:** We now display Account Names in the Recommendations widget and Overview list, improving clarity.

• Anomaly Alert Upsert Functionality: When adding a new anomaly alert, the system was updating the existing alert instead of creating a new one when an email address was added. We have now implemented an upsert function for the channel list to ensure a new alert is added correctly, enhancing alert management.

Fixed Issues​

• Date Filter Sync Issue on Evaluation Details Page: There was a mismatch between the date filter and URL-passed dates, causing an "undefined - undefined" display. Users had to manually select a date range. This is now fixed for a smoother experience.

Version 1.31.4​

New features and enhancements​

• Updated Cloud Custodian Versions for Cloud Asset Governance: The cloud-custodian versions have been updated to the following:

  • c7n==0.9.41
  • c7n_azure==0.7.40
  • c7n_gcp==0.4.40 [CCM-19608]

• Support for AWS GovCloud in Asset Governance: Asset governance now extends support to AWS GovCloud, enabling governance features. [CCM-18909]

• UI Improvement for AutoStopping Windows: We have added enhanced CSS to provide adequate space for selection of source IP-based custom exclusions in the exclusion list. [CCM-14294]

September 2024​

Version 1.30.3​

New features and enhancements​

• Prerequisites for Cluster Orchestrator: We are now listing all prerequisites required for Cluster Orchestrator, providing clearer guidelines to ensure smooth setup. [CCM-18892]
• Export Perspective Chart as CSV: We now support exporting the Perspective chart as a CSV, allowing users to easily extract and analyze data from the chart. [CCM-18812]

Fixed Issues​

• Toggle Fields in Recommendations Filters: The Toggle fields in the Recommendations screen Filters were not being saved on Create or Update. This has been fixed. [CCM-19267]

• Saved Filter Option Not Visible on Anomalies Page Load: When first loading the anomalies page, the option to "Select a saved filter" was not visible until a filter was applied. This has been corrected so the saved filter option appears immediately. [CCM-19345]

• Overflowing Text in 'Resume Setup' Button: The button text for "Resume Setup" in the orchestrator setup screen was overflowing. This has been fixed. [CCM-19212]

• Cloud Connectors Redirect Fix: Clicking on cloud connectors in the integration did not lead to the actual connector configuration. Now, it redirects to the connector details page, improving troubleshooting and configuration access. [CCM-17586]

• Cost Category Bucket Overflow Handling: Display issues with overflowing content in cost category bucket conditions when multiple choices were selected have been fixed. [CCM-14719]

• Search Bar Missing for Label Value Selection in Cost Categories: The search bar was not appearing when selecting label values while creating cost categories. This has been resolved. [CCM-15437]

• Asset Governance Savings Displayed as $0 on Overview Page: Asset Governance Savings were showing as $0 on the CCM Overview page. This issue has now been resolved, and the correct savings are displayed. [CCM-19450]

Version 1.29.1​

Fixed issues​

• Budget Amount Reset: Earlier, navigating between the budget creation steps caused the budget amount settings to reset to defaults. The budget form fields will now retain their values when switching between steps. [CCM-10953]
• Cost Category Name Updates: We have addressed an issue where cost category name updates were not properly reflected in perspective rules, ensuring the correct category names are used in all relevant areas. [CCM-18977]
• Recommendations Slider Action: Previously, using the buffer slider in the CCM "recommendations" view and hitting the back button moved the slider instead of returning to the main recommendations page. Now, the back button no longer includes slider actions. [CCM-11303]

August 2024​

Version 1.28.7​

New features and enhancements​

• Custom Recommendations in Cloud Asset Governance: We have introduced custom recommendations in Cloud Asset Governance, allowing users to create, manage, and operationalize their own policies. A new custom recommendations page has been added where users can view, manage, and drill down into their custom recommendations based on various filters. [CCM-18955]

• Support for Multi-Status in Governance Evaluation: We've added support for multiple statuses within a single Governance Evaluation, enabling more detailed policy evaluations. [CCM-18942]

• GCP Data Sync Migration to Gen2: We have successfully migrated our GCP data sync Cloud Function from Gen1 to Gen2, significantly increasing the timeout from 9 minutes to 60 minutes, enhancing the reliability and performance of GCP data sync. [CCM-13672]

Fixed issues​

• Budget Percentage Display Issue: Previously, when a budget amount was set to 0, the percentage spend comparison displayed as "Infinity." This has been fixed. [CCM-]

• Cluster Data Query Optimisation: We've started using Mongo hint to optimize the cluster billing calculation query, ensuring much faster calculation. [CCM-19140]

• Cost Category Name Updates: There was an issue with cost category name updates that affected the rules in perspectives where the cost category was referenced. This issue has now been fixed. [CCM-18977]

Version 1.27.2​

Fixed issues​

• Filtered Output Reset: Previously, when a user re-ran a rule with a filtered output, the output was not reset. This issue has now been fixed, ensuring accurate output after each rule execution. [CCM-18859]
• Spot and On-Demand Split Validation: Previously, no error was shown when the split ratio between Spot and On-Demand did not equal 100%. Now, if a user attempts to complete enablement in Commitment Orchestrator with an incorrect split ratio, an error message will be displayed, ensuring proper configuration. [CCM-18857]
• Commitment Orchestrator Tooltip Accuracy: Enhanced the chart re-rendering after applying filters, hence ensuring accurate tooltips. [CCM-18739]

July 2024​

Version 1.25.0​

New features and enhancements​

• Dry Run and Run Once Buttons: Resolved an issue where the "Dry Run" and "Run Once" buttons were not enabled if there were issues with policy YAML validation. [CCM-18754]
• Regions Filter: Added a new Regions filter in the Recommendations Filter Panel to allow more granular filtering. [CCM-18591]
• Autostopping UI Changes: Improved the UI for Autostopping, including updates to the page index on the URL to match the page in the list, a wider first column based on available space, and the "Last Activity" column now reads the updated_at value directly instead of calling a separate API. [CCM-18585]

Fixed issues​

• Governance Evaluations List: Resolved an issue where the page number was not resetting when changing the time filter on the Governance Evaluations List page, which resulted in retrieving the incorrect page with no results. [CCM-18571]

Version 1.23.0​

New features and enhancements​

• Policy YAML Validation Enhancement in Asset Governance: Previously, we were not enabling the Dry Run and Run Once buttons if there were issues with policy YAML validation. This should now be resolved, allowing users to proceed with these actions even if there are validation issues. [CCM-18754]

• Regions Filter for Recommendations Page: We have added a Regions filter in the Recommendations Filter Panel on the Recommendations Page, enhancing the ability to filter recommendations by specific regions. [CCM-18591]

• Load Balancer Pagination Fixes: Some improvements have been made to the load balancer pagination:

  • The page index on the URL now matches the page in the list.
  • The first column is now wider based on the available space.
  • The last activity column now reads the updated_at value instead of calling a separate API. [CCM-18585]

Fixed issues​

• Governance Evaluations List Page Pagination: Previously, on the Governance Evaluations List page, the page number was not resetting when changing the time filter, causing the current page to match the filters incorrectly and resulting in no results. This issue has now been resolved. [CCM-18571]

June 2024​

Version 1.23.0​

New features and enhancements​

• Asset Governance Recommendations page enhancements: Previously, Governance Recommendations were grouped by Account and Region for AWS, Subscription and Region for Azure and Project for GCP. Now, each Account-Region pair, Subscription-Region pair and Project will have individual recommendations listed in the Recommendations Overview page. This would allow users to create individual Jira/ServiceNow tickets per pair and "Recommended Action" now shows the rule name that is recommended.

• Memory Metrics Tooltip for EC2 Recommendations: Added a tooltip to EC2 recommendations to give more information on memory metrics. This tooltip provides documentation on what users need to do to enable memory metrics, ensuring they understand how to gather necessary data for accurate recommendations, especially for memory-optimized instance families.

• Disabled Enforcements for Accounts without a Valid CCM License: We have disable Enforcements related to accounts that do not have a valid CCM license.

Early access features​

• Early access to Commitment Orchestrator for RDS: We have introduced Commitment Orchestrator for RDS with support for Standard RI. Being in the very early stages, it is only available on request. If you would like to test it out, kindly reach out to Harness support

Version 1.22.0​

New features and enhancements​

• Azure Data Sync Frequency Update: The Azure data sync frequency has been changed from 1 day to 1 hour. This enhancement will allow data to flow faster for Azure customers, reducing wait times and improving data availability [CCM-18014]

Version 1.21.0​

New features and enhancements​

• Savings by Harness Integration Enhancement: Last release we added "Savings by Harness" integration on Commitment Orchestrator. This release, we have added a minor enhancement with a loading state for savings and utilization widgets. [CCM-18212]

• Azure VM Inventory Duplicate Entries Enhancement: Duplicate entries within AzureVMInventory will now be handled better after this fix, ensuring appropriate data is displayed on the dashboards. [CCM-17313]

• Autostopping Rule Search by VM ID: Previously, users could not search their rule by VM ID. Now, users will be able to search their rule in the rule list using the ID of any VM managed by the rule. [CCM-16437]

Version 1.20.3​

New features and enhancements​

• "Savings by Harness" Integration on Commitment Orchestrator : Added a new feature to display savings by Harness on the Commitment Orchestrator Overview page, providing users with detailed insights into savings achieved by using Harness Commitment Orchestrator. [CCM-17919]

May 2024​

Version 1.19.1​

New features and enhancements​

• Asset Governance for AWS : Added new governance recommendations and cost computation support for AWS resources including cache-cluster, S3,Redshift, and Redshift-snapshot. [CCM-17852]
• Asset Governance for GCP : Added new governance recommendations and cost computation support for GCP resources including redis, gke-cluster, bq-dataset, function, bucket, dataflow-job, loadbalancer-address, cloud-run-service and also, added cost computation for gcp.loadbalancer-forwarding-rule. [CCM-17852]
• Editing Enforcements: Improved the enforcement editing process. Now, when viewing and editing an enforcement, it will no longer get enabled automatically if it was previously disabled. This ensures that the enforcement status remains consistent unless intentionally changed by the user. [CCM-18050]
• Email Validation Enhancement: We've introduced better validation for email addresses in the recipients list for perspectives report. This enhancement ensures that only correctly formatted email addresses are accepted in the recipients list, enhancing data integrity and security. [CCM-17850, ZD-63324]
• Azure Governance Subscription selection: In the Azure Rule window's Subscription drop-down menu, previously, only the Subscription ID was displayed. Now, both the Subscription Name and ID are shown for better clarity and ease of selection. [CCM-17650]
• Node New K8s Labelling: We have added support for the new K8s labels starting from K8s v1.17 for Instance Type, Operating System, Region, and Zone respectively. We use these labels to get the public pricing data for a given cloud provider. [CCM-17979]

Fixed issues​

• Azure perspectives Previously, only fields in "group by" with available data were displayed in Azure perspectives, leading to incomplete views and unnecessary errors. With this update, all relevant fields, including those without data, will now be visible. This eliminates any unnecessary errors arising from missing data at the source level. [CCM-17573, ZD-62691]

Version 1.18.1​

New features and enhancements​

• Governance JSON Viewer UI Improvement: We have moved the Copy button in the Governance's JSON viewer beside the Filter button for better accessibility and user experience. [CCM-17651]
• Cost Correlation Refresh Button: Expanded the functionality of the refresh button to include multi-policy evaluations, allowing users to update and refresh data related to savings for multi-policy evaluations. [CCM-17648]
• Time Filter Options for Recommendations: We have introduced UI changes to support time filter options for the last 60 days and 90 days in recommendations. This enhancement provides users with greater flexibility in analyzing recommendations over specific time periods. [CCM-17725]

Fixed issues​

• Edit Flow for TCP-Based Autostopping Rule: Fixed the issue where users were unable to edit the auto-stopping proxy from the dropdown and proceed with the flow. Users can now see the preselected value of the proxy and complete the process as expected.
• Azure Data Fix: Resolved an issue where adding a perspective rule with Azure subscription ID set to null resulted in no data being displayed. [CCM-17414]
• Auto Stopping Page Filter: Added a fix for a minor issue where using filters on the auto-stopping page resulted in an unknown error. [CCM-16195]

Version 1.17.0​

New features and enhancements​

• Commitment Active Actions Details Enhancements:

  • Added formatting to the expiry date for improved readability.
  • Aligned savings summary chart colors with graph colors for improved visual coherence.
  • Added option to expand details about a recommendation by clicking on it. This enhancement allows users to gain deeper insights into their recommendations. After clicking on it, users can view both the potential savings after applying the recommendation and the potential spend without applying the recommendation. Additionally, users can access details about the source Reserved Instance (RI).[CCM-17844, CCM-17474]

• Cloud Perspective Recommendation Display: Recommendations alongwith their total potential savings is now be displayed on the cloud perspective, providing users with actionable insights directly within their Perspectives overview page. [CCM-17639]

• Enhanced JSON Viewer: The same JSON Viewer used in the Evaluation details page has been added to the Governance Recommendation details page, providing users with consistent and user-friendly access to the output data with all the available filters. [CCM-17436]

Fixed issues​

• Updated Messaging in Commitment Orchestrator: In Commitment Orchestrator, we have improved the UI by adding appropriate messaging about recent actions once they are completed, and for actions that do not require user intervention, ensuring clarity and accuracy in the displayed information. [CCM-17655]

Version 1.16.0​

New features and enhancements​

• Expanded Azure Region Options: We have introduced a new allregions option in Azure regions, allowing users to execute any Rule or establish an Enforcement across all regions with just one Evaluation for a Subscription. This boosts efficiency and helps prevent exceeding Azure API limits when executing a Rule for a Subscription across multiple regions. Instead of running on multiple regions it will now run only on one single region i.e allregions. [CCM-17185, CCM-16771]

• Anomaly Drilldown Support: We have introduced support for Anomaly drilldown, allowing users to precisely view anomalies on the Anomaly List Page that were visible on the Perspective Details Page. [CCM-17137]

• Perspective List Page Enhancement: We have added a minor change on the Perspective List Page of removing the total cost and changing the default view to list format. [CCM-17380]

Fixed issues​

• Label Adjustment for Clarity: We have added a minor change to update label from "Active Rules" to "Total Rules" for enhanced clarity and accuracy in Autostopping UI. [CCM-16450]

• SMP Overview Screen Total Cost: We have updated the query used to retrieve active spend on the overview page for SMP to ensure accurate data representation. This fix ensures alignment between the total cost displayed on the SMP Overview Screen and the data presented in perspectives. [CCM-17380]

• Perspective Dashboard Optimization: As part of this fix, we have disabled redundant parentheses from BigQuery SQL query to optimize the perspective and cost category queries, enhancing query efficiency and performance.

Version 1.14.3​

New features and enhancements​

• RBAC Support for Active Recommendations: To enhance security and control over active recommendations, we have implemented Role-Based Access Control (RBAC) support. With this update, only admin users with edit access or commitments can approve or reject recommendations. This feature ensures that sensitive actions related to active recommendations are limited to authorized users to improve overall security. [CCM-17337]

• Introduction of Anomaly Workflows: We have added Anomaly Workflows to enhance anomaly management within the CCM platform. With this feature, when a user marks an anomaly as a False Anomaly, it will be automatically moved to the Ignore list within the Anomaly Settings drawer. Furthermore, users now have the option to undo this action directly from the Anomaly Settings drawer, providing greater flexibility and control over anomaly management processes. [CCM-17311]

• Multi-Select Cloud Providers Filter: We have enhanced the Recommendations and Governance Evaluations filter panel by adding a multi-select Cloud Providers filter. This addition allows users to select multiple cloud providers simultaneously, providing more flexibility with governance evaluations. [CCM-17150]

• Child Account Exclusion in Commitment Orchestrator: With this functionality, users can selectively exclude child accounts from commitment utilization calculations, allowing for more precise management of Reserved Instances (RIs) and Savings Plans (SPs) across their organisation. [CCM-17184]

• Audit Trail Integration for changes in Commitments: Commitment orchestration flow changes like change of coverage or split between SP & RI and commitment actions approval or rejection is now captured in audit trails. [CCM-17267]

Fixed issues​

• Governance Rules List Pagination: We have added a minor fix on the Governance Rules List page where the page number now resets properly when applying either the cloud provider or rule type filter. [CCM-17279]

• Granularity Adjustment for Cluster Data Retrieval: With this fix, we have ensured that data retrieval aligns with the specified granularity settings: fetching from the daily data table for DAILY granularity and from the hourly table for HOURLY granularity. [CCM-16061]

April 2024​

Version 1.13.0​

Fixed issues​

• Improved Governance Enforcement: We have added a minor fix for governance enforcement when recommendations contain multiple accounts or subscriptions so that the enforcement is created efficiently. [CCM-17192]

• Improved Precision in Perspective Reports: Previously, when choosing the current month during perspective setup, the report would present data from the past 7 days. We have promptly added a fix so that it accurately portrays data for the entire current month. [CCM-17100]

Version 1.12.0​

New features and enhancements​

• Native User Approval in Commitment Orchestrator: We have implemented native user approval in Commitment Orchestrator. With this new feature, users can approve the recommendations generated by the orchestrator. This feature enhances the functionality of the orchestrator by enabling user actions such as approval or rejection of recommendations generated by the engine. Read more about it here. [CCM-16648]

• Pagination for Budgets: We have added pagination for our Budgets page. This allows users to navigate through multiple pages of budget entries, improving the overall user experience by making it easier to manage and access information efficiently. [CCM-16978]

• Asset Governance Overview Page Revamp: We have changed our Overview screen for Asset governance and introduced new UI enhancements such as:

  • Widget for All-time Evaluations
  • Widget for Savings in Timeframe
  • Widget for Evaluations in Timeframe
  • Widget for Evaluations trend (per day)
  • Resource count in recommendations
  • Cloud filters

Fixed issues​

• Commitment Orchestrator UI Column Overlap: We have added a minor fix to address the issue of overlapping columns in Commitment Orchestrator overview page. [CCM-16971]

• Resolved Cluster Data Visibility Issue: We resolved a problem regarding inaccurate cost tracking within the specified date range due to a recent change in the primary node's TimescaleDB timeout configuration, which led to failures in the utilization data query. By rerouting read queries to a secondary node, we fixed the query performance issue. [CCM-17048]

Version 1.11.3​

New features and enhancements​

• Updates in Output Terminal in Asset Governance:

  • Filter: We've added a new feature for filtering the JSON in Evaluations. Users can filter based on the keys of the output and use comparison operators like ==, <, >, etc. for comparison against numeric fields and use the LIKE operator for comparison against alphabetic fields.
  • Sort: Users can also sort the output based on a specific key in ascending or descending order. For showing outputs with only one particular key, they can use the pick option.
  • Search: Searching in the output terminal in Asset Governance is now available. Users can search for any piece of text in their output. [CCM-16309]

• JSON and CSV Zip Download for Multi Policy Evaluations: For multi-policy evaluations, users can now download them in zip files. Each policy would have a sub-directory inside the zipped directory. Also, we have added two new options to include log files and filters applied (to the original JSON) in the downloaded folder. [CCM-16774]

• Refresh Cost Impact for Evaluations: We've added a new option to "refresh" the cost impact for a particular evaluation for only supported resources. With this option, users can choose to refresh the cost impact to see updated savings. Please note, it can take up to 30 minutes to show the updated results. [CCM-16807]

• Pagination and Ignore Recommendations in Governance Overview Page: Users can now view 5 recommendations per page on the overview screen of Asset Governance. We now also show ignored recommendations in the list with the option to remove them from the ignored list in the overview page itself. [CCM-16824]

Early access features​

• Introduction of CloudAccountId and VM ResourceId in ClusterData Table: This feature is a beta release behind a feature flag. We are introducing cloudAccountId and VM resourceId in the clusterData table. This addition enhances the capabilities of the system, providing more comprehensive data for analysis and management purposes. Please note that this feature is currently in early access and may undergo further improvements based on user feedback. [CCM-15506]

Fixed issues​

• Enhanced Discount Calculation in Cloud Cost Management: To improve the accuracy of discount calculations, we've updated our data utilization strategy. Previously, discounts were calculated based on the lineitemtype column, leading to potential inaccuracies under specific filter scenarios. Now, we've integrated the use of edpdiscount, bundleddiscount, and privaterateddiscount columns from the Cost and Usage Report (CUR), ensuring discounts are accurately calculated at the service level. [CCM-20458]

• Enhanced Rendering of Timeseries Data for Shared Bucket Cost: Previously, when viewing shared bucket costs using the shared bucket filter, the timeseries data was not rendered correctly. We have implemented a fix to address this issue, ensuring that the timeseries data is now properly displayed. [CCM-16876]

• Improved Cluster Cost Calculation for Grouping by Cluster Cost Category: Previously, when grouping by cluster cost category, Others costs were calculated from pod cost, resulting in costs being displayed even if they were 0. To address this, we have enhanced the calculation process to derive costs from node cost instead of pod cost depending on the group by selected. This ensures accurate cost representation, eliminating discrepancies in cases where costs should be zero. [CCM-17050]

March 2024​

Version 1.10.2​

New features and enhancements​

• Perspectives update: Added a new query parameter called updateTotalCost in the create and update perspective call. When set to false, the total cost that is displayed on the list perspective page will not be calculated. Instead, only create/update operation for the perspective will be performed.[CCM-16724]
• New limits for cost categories: Per account, there can be up to 25 cost categories, 1000 cost buckets, and 10 shared buckets. Additionally, nested cost category hierarchy can extend up to 5 levels per cost category. [CCM-13843]

Fixed issues​

• Fixed intermittent issues which resulted in failure in S3 data sync and data ingestion for seamless and on-time updating of data. [CCM-14988]

Version 1.9.5​

Fixed issues​

• Previously the exported CSV was not reflecting the Perspective Preferences set by the user. Now it has been fixed and the rows of Perspective Grid and exported CSV should match. [CCM-16586]

February 2024​

Version 1.8.1​

Fixed issues​

• We added a default value (CURRENT_MONTH) for the overviewTimeFilter parameter in the Overview Forecasting API, ensuring consistency and simplifying usage. [CCM-16458]
• We identified and resolved a high memory and CPU utilization issue in our delegate pods, traced back to improper handling of Chronicle libraries. The fix involved ensuring the StoreTailer objects are closed after each use, significantly improving system performance and stability. [CCM-16052]

Version 1.7.3​

New features and enhancements​

• We have seamlessly integrated Azure preferences into the Account Settings. Users can conveniently configure your preferences there, and once the Azure preferences feature is launched, they will be applied across all Azure perspectives for enhanced customization and consistency. [ccm-15789]

Fixed issues​

• In the Perspective UI, readability was previously compromised as the total cluster cost information overlapped behind the tile. With the latest update on the perspective details page, formatted cost will now be clearly visible, and users can access the full value through tooltips for enhanced clarity and usability. [CCM-16413]

Version 1.6.5​

New features and enhancements​

• In the Azure connector flow, a new field named "billing type" has been incorporated to identify users' billing types. This enhancement sets the groundwork for enabling Azure cost preferences in future updates. [CCM-15978]
• We've implemented the edit flow for commitment orchestration, granting users the ability to modify their commitment orchestration details seamlessly. [CCM-11304]

Fixed issues​

• The Filter API is paginated; however, the UI filter component lacked pagination support. To address this, we have implemented a result limit of 1000. Now users will be able to see all the saved filters in the recommendation filters. [CCM-16364]
• Previously, users faced limitations when attempting to adjust constituents within budget groups, hindering adaptability to organizational changes. Now, Budget group edit flow will not block the budgets or budget groups selected in that budget group eliminating constraints and enhancing flexibility. [CCM-16196]
• After upgrading to SMP version 0.120, users encountered issues with BI Dashboards loading, prompting them to add connectors despite existing connectors at the account level. In response, for SMP Environment, we've implemented redirects to the Dashboards module to facilitate viewing BI Dashboards seamlessly. [CCM-15995]

January 2024​

Version 1.5.1​

Fixed issues​

• Previously, it was a bit difficult editing cost categories, especially on smaller monitors or when the web browser isn't maximized. To address this, we've implemented a fix in the layout of rules specifically for smaller windows. With this adjustment, users can now seamlessly edit cost categories even on smaller screens, ensuring a smoother experience across different viewing contexts. [CCM-15991]
• Upon clicking back from the shared cost screen, users encountered a non-functional continue button due to form validation issues. We have resolved this impediment from the UI side, ensuring users can proceed seamlessly without hindrance. [CCM-15990]

Version 1.4.2​

Fixed issues​

• In Commitment Orchestrator, the exclusion of resource instances was found confusing sometimes, necessitating a clearer flow. We've made enhancements to the commitment orchestration setup flow. [CCM-15844]

Version 1.3.0​

Fixed issues​

• While filtering for a GCP anomaly and clicking on the Anomaly link in perspectives redirected users to the Azure-based anomaly screen. To rectify this, we've implemented a solution to ensure that only relevant anomalies are displayed by adding an "EQUAL" case in the switch condition with the appropriate condition format for equal cases. [CCM-15649]

Version 1.2.1​

Fixed issues​

• Users without edit permissions for perspectives were still able to see the "+ New Perspective" button, which was not grayed out, leading to confusion. Now, if users lack edit access on perspectives or folders, the "+ New Perspective" button will be disabled, preventing confusion. Furthermore, error messages have been refined to provide clearer feedback. [CCM-15611]
• When creating or updating a perspective with an "invalid" cost category. If a cost category shares the exact same name as the attribute "shared cost," attempting to include it in a perspective results in a failure message: "Oops, something went wrong on our end. Please contact Harness Support." We have improved the error message with details telling that the cost category name cannot be same as shared cost bucket name. [CCM-15536]
• Horizontal scrolling was absent from all pages except perspective-details. This has now been successfully addressed and resolved. [CCM-14720]

Previous releases​

2023 releases

December 2023​

Version 81801​

Fixed issues​

• Previously, clicking on "Log Section" in the Review Screen redirected users to Compute Coverage instead of defaulting to the Events Logs Section as intended. Now, from the review section of commitment orchestration, the link for the log section correctly directs users to the Events Logs Section. This issue has been successfully resolved. [CCM-15189]
• Sometimes, changing filters in the default Perspective interface would reset the page but fail to display the table initially. This issue has now been resolved, with the table consistently appearing after filter and groupby adjustments. [CCM-14990]

November 2023​

Version 81700​

New features and enhancements​

• Pagination for perspectives has been added for faster loading times of perspectives. By default only 20 perspectives will be shown. To see all the perspectives set the pageSize as 10,000 and pageNo as 0. By default all perspectives are ordered by most recent. (CCM-15124)

Version 81601​

New features and enhancements​

• RBAC (Role-Based Access Control) support has been implemented for the Commitment Orchestrator. CCM Admins now possess the authority to configure the Commitment Orchestrator for master accounts. On the other hand, CCM Viewers are granted access to visibility screens within the Commitment Orchestrator interface. (CCM-15040)
• Earlier, we didn't support adding relevant rule filters for perspectives created through cloud providers. As a result, all anomalies were being displayed on the cloud providers' perspective, regardless of whether they were relevant to that perspective or not. In this release, we have now added rule filters for cloud providers to address this issue. (CCM-15068)

Fixed issues​

• Previously, entering connector names resulted in incorrect error message "Delegate with that name already exists." This hindered the quick connection setup. Now, the system accurately reflects the correct error message for when Delegate validation failures. (CCM-14963)

Version 81501​

New features and enhancements​

• The perspectives page has been enhanced with pagination for both the Card and List views. Each page will display a maximum of 20 perspectives, addressing the issue where some customers experienced lag during the initial rendering of the perspectives list pages. (CCM-14018)

Fixed issues​

• In the updated functionality for K8s connectors, the "View Costs" feature is now enabled based on the presence of cluster data rather than relying solely on the last events received. This enhancement ensures that users retain the ability to view historical costs. (CCM-14984)

Version 81402​

New features and enhancements​

• Previously, there was no option to export Recommendations as CSV files. Now, we have added a new feature that enables users to export Recommendations as comma-separated values (CSV) files. (CCM-14274)

Fixed issues​

• Previously, changing the project in JIRA didn't clear fields, causing potential creation failures. (CCM-14842)

However, now, the form (except Ticket Summary and Description) resets on project change, ensuring a smoother process.

• Previously, anomaly detection on K8s Services lacked a threshold, leading to excessive alerts. (CCM-14865)

Now, we have implemented a threshold of $3 for anomaly detection on K8s Services to refine the alerting process.

• Previously, incorrect entity types for Azure in anomalies caused misdirected notifications on Slack and email. (CCM-14864)

However, this issue is fixed now by changing the logic for Azure entity types.

October 26, 2023, version 81300​

Fixed issues​

• Previously, the search functionality in the perspective grid was nonfunctional when grouped by cost categories, causing inconvenience in data retrieval. (CCM-14384)

  This issue is fixed now, enabling users to efficiently search within the perspective grid, even when cost categories are applied as grouping criteria.

• Previously, Anomaly alert logic for default perspectives was not functioning correctly. (CCM-14670)

  This issue has been resolved, ensuring accurate anomaly notifications for default perspectives via Slack and email.

• Previously, the Budget explorer chart displayed lower costs due to a query issue. (CCM-14758)

  This issue is fixed by removing the "group by" from the budget time-series query, ensuring more accurate cost representation.

October 20, 2023, version 81202​

Fixed issues​

• Previously while creating and AutoStopping rule, the k8s cluster selection list only displayed top 100 records, which made it difficult to choose a cluster when the list had more than 100 entries. (CCM-14644)

  This is fixed by increasing the limit from 100 to 500 records. This will assist in retrieving every cluster that is accessible for selection.

• Previously, in autostopping rule creation flow for Azure cloud provider, there's an option to create one AutoStopping rule using AppGateway belonging to one subscription and VMs which restricted the selection of VMs belonging to another subscription. (CCM-14515)

  This is fixed by adding support for selecting multiple connectors, one for the AutoStopping rule and other for the VMs. This helps in having independent connectors for each subscription.

• To avoid any confusion, tooltips are added for Potential Monthly Cost and Potential Monthly Savings on EC2 Recommendation details page. (CCM-14613)

• A bug was identified during ingestion of AWS CUR data. The impact of the bug was causing AWS cost data from the current month to be dropped while ingesting AWS-CUR updates for the previous month. Since AWS updates CUR of previous months until the first few days of next month, this issue was hit in the initial days of the month. This bug got introduced recently while supporting ingestion of future-dated entries in any billing-period (i.e. supporting ingestion of costs which have usageDates beyond the current month). (CCM-14618)

  This is now fixed by considering billingperiodstartdate and billingperiodenddate during ingestion from AWS CUR.

• It was observed that the budget's cost was not matching with the costs in the underlying perspective, resulting in showing lesser cost in the budget explorer chart. (CCM-14758)

  This issue is fixed by making some changes in the underlying budget cost explorer query.

October 12, 2023, version 81100​

Fixed issues​

• Previously, attempting to edit a cost bucket with operands selected as "NOT NULL/NULL" led to an unexpected error, subsequently hindering the editing of other buckets. (CCM-14519)

  This issue has been fixed by introducing a custom validation to resolve the error.

• Previously, our application allowed fetching anomalies for perspectives created through labels. The queries used for fetching anomalies in these cases were based on the default groupBy field, leading to the display of numerous incorrect anomalies in the labeled perspective, which were unrelated to the labeled resources. (CCM-14242)

  As of this release, we have discontinued support for fetching anomalies in perspectives created solely through labels. This change is aimed at improving the accuracy of anomaly reporting and ensuring that only relevant anomalies are presented.

October 5, 2023, version 81000​

Fixed issues​

• Previously, For EC2 Recommendations, the monthly savings from crossFamilyRecommendation or sameFamilyRecommendation fields, based on the selected preferences, were used to display the savings amount in the widget. However, for Terminate Recommendations, as these values were not available, it resulted in showing $0 in the widget. (CCM-14544)

This is fixed by switching to totalMonthlyCost and totalMonthlySavings for the potential savings widget in EC2 Terminate Recommendations.

September 27, 2023, version 80904​

New features and enhancements​

• Previously, CCM displayed only the essential Jira or ServiceNow fields in Recommendation Workflows. However, with this enhancement, CCM introduces a new field + Fields that allows users to add optional fields as needed.

Fixed issues​

• Previously, CCM used to display all anomalies, including the new ones that were labeled as "N/A." (CCM-14275)

  However, now, anomalies will not be shown on the UI for newer entities. Newer entities are those having data for a duration of 15 days or less.

• In the AWS perspective, the cost calculation is based on the selected Groupby field, and CCM uses the SUM of awsUnblendedCost. However, when CCM detects anomalies for AWS, it is based on the SUM of awsBlendedCost. This led to a cost mismatch between what's displayed on the AWS perspective and the cost reported for anomalies.(CCM-14096)

  This issue is fixed by using SUM of awsUnblendedCost to detect AWS (Account, Service and UsageType) anomalies.

September 20, 2023, version 80804​

Fixed issues​

• Previously, on the AutoStopping details page, the dry-run flag did not invoke the savings API when turned on. (CCM-14232)

  This issue has been fixed. Now, the cost savings are correctly displayed in the dry-run mode.

September 7, 2023, version 80702​

New features and enhancements​

• Cost category enhancement (CCM-12879)

Implemented a new feature that enables users to copy cost buckets from one cost category to multiple cost categories simultaneously. You have the flexibility to choose any number of buckets for copying. Upon selection, a popup prompt appears, allowing you to specify the target cost categories for copying the selected buckets. Upon successful completion, you receive a success notification along with relevant details.

However, it's important to note that while copying you may encounter issues if the destination cost category already has a bucket with the same name as the copied one. In such cases, you can address the conflict by renaming the bucket before attempting the copy operation again.

Fixed issues​

• An issue with the budget group warning message (CCM-14150)

  Previously, there was a discrepancy between the dates displayed in the warning message and those in the budget table column. This issue has been fixed now, and the dates are consistent on both the table and the warning message.

  

• Previously, users were unable to schedule the same AutoStopping rule for the same time in different time zones, such as setting the rule to run at 8 PM - 9 PM IST and 8 PM - 9 PM EDT, which was considered an overlap. (CCM-14181)

  This issue has been resolved. Added validation to check for time zones to allow multiple schedules for a single rule.

• Multiple OverlappingFileLockException errors were occurring due to the Chronicle Queue library version used. Consequently, this issue led to Perspectives not displaying any data. (CCM-14174)

  The issue is resolved now. The Chronicle Queue library has been upgraded to a newer version, resolving the errors.

September 04, 2023, version 80606​

New features and enhancements​

• Display refunds or discounts on the graph within perspectives. (CCM-13443)

  Previously, our graph in perspectives didn't display refunds or discounts, resulting in empty spots when values were negative. This enhancement improves this by aggregating negative values into a red-colored bar chart. You can now toggle a button in General Preferences to view these previously hidden negative costs.

  

Fixed issues​

• After editing an existing cost category, previously, when attempting to create a new one, the drawer displayed the details for the last accessed cost category. (CCM-13973)

  This issue is fixed. The new cost category builder drawer now opens in the expected empty state, and the edit drawer will correctly display the details of the selected cost category. Even after editing a cost category, if you open the new cost category drawer, it correctly shows the empty state as intended.

• Previously, users were unable to select This Quarter in the Overview time range filter. Instead, the application defaulted to This Month upon selection. (CCM-13505)

  This issue has been fixed.

• An issue with onboarding an existing AutoStopping rule on Azure App Gateway version 1 (CCM-13903)

  This issue is resolved. Now the health check configurations of the AutoStopping rule are being used to detect the current backend settings associated with the rule.

• The screen went blank when deleting all characters in the AutoStopping Rule Name field while editing the rule. (CCM-13816)

  This issue has been resolved.

August 23, 2023, version 80500​

New features and enhancements​

• ServiceNow integration with Recommendations (CCM-11150)

  Introducing ServiceNow as a ticketing tool to create tickets for recommendations. You can use either Jira or ServiceNow as your ticketing tool. You need to configure this setting at the account level on the Default Settings page. For more information, go to View and apply recommendations.

• AWS perspectives enhancement (CCM-13914)

  Introducing support for the following additional Group By options in AWS perspectives:

  • Billing entity
  • Line item type

Fixed issues​

• Previously, within the budget Edit flow, the monthly breakdown values would reset to default values. However, currently, the resetting occurs only when there's a change in the Budget Type. (CCM-13763)

• Previously for ALB proxy HTTP route configuration, when it comes to redirect actions, only redirect URLs were supported. However, now, users have the flexibility to include either a redirect URL or specify a target port and protocol. (CCM-13702)

August 02, 2023, version 80301​

What's new​

• AWS AutoStopping proxy enhancement (CCM-13497)

  You can now select the subnet ID from the dropdown list for AWS AutoStopping proxy creation.

  

• Perspective Preferences enhancement (CCM-11145)

  Perspective preferences provide you the flexibility to control which cost factors are considered in your billing and CUR (Cost and Usage Report) reports within your perspective. You can now include cost factors such as discounts, taxes, and refunds. For more information, go to Perspective Preferences.

info

The current configurations for Show others and Show unallocated cost in clusters are preserved. This means that though the default settings have these preferences set to false, any existing perspective with these preferences set to true will retain their current state and not be overridden.

• Improved UI handling during the AutoStopping rule creation process (CCM-13527)

  The page on which users select either a load balancer or an AutoStopping Proxy has been enhanced to include an additional API that retrieves information about proxies created previously in shared VPCs. Now users can use a proxy created in a shared VPC across projects and connectors.

• Overview page enhancements (CCM-13326)

  • The pie chart now shows a hover state.

  • The forecast trend in the widget is removed.

  • Added forecast time period in the forecast cost widget.

  • Changed the heading of the cloud cost widget from Top AWS accounts to Top 10 AWS accounts.

    

Fixed issues​

• Previously, configuring both the redirect URL and target port for redirection while creating a redirect-based AutoStopping rule led to an error. (CCM-13475)

  This issue has been resolved by modifying the validation process. Now, if the redirect URL is defined, the validation process checks whether the target port is greater than 0. Specifying both redirect URL and target port is not allowed as it is an invalid configuration. However, for ALBs, only redirect URLs are allowed.

• Users were unable to validate their YAML files when creating a Kubernetes AutoStopping rule. (CCM-13459)

  This issue has been resolved. Users will now be able to validate the YAML successfully.

• Previously, users experienced performance delays while editing cost categories with more than 50 buckets, and every subsequent action took several seconds to trigger. (CCM-13205)

  The issue has been resolved, and the overall user experience has been enhanced by streamlining the process of managing cost categories even with a large number of buckets.

July 13, 2023, version 80102​

What's new​

Recommendations page UI enhancement (CCM-12693)

The Include dropdown on the Recommendations page has been removed. Instead, the following toggle options have been added in the Filter panel as shown in the screenshots below:

• Show Recommendations on Parent resource
• Show Recommendations on Child resource
• Show Recommendations on resources added to the IgnoreList

By default, the first two options are enabled, and you can modify the toggles to customize the list filtering.

&nbsp

Fixed issues​

• The message displayed on the UI was incorrect when there were no recommendations for the account. (CCM-13250)

  This issue is fixed now.

  

• Adding an invalid cost category in a perspective rule caused the Anomalies feature to not function as expected. (CCM-13218)

  Now, the cost category IDs are validated when added in the perspective rule.

July 07, 2023, version 80002​

What's new​

• Azure VM recommendations (CCM-13142)

  Now, the recommendations are computed based on both Memory Utilization data and the existing CPU Utilization tracking data.

• Budget Sorting Enhancement (CCM-10948)

  This enhancement allows you to conveniently sort budgets alphabetically in Harness CCM. You can now browse and navigate through budgets more efficiently.

• Recommendations enhancement (CCM-11665)

  You can now easily move recommendations from the Applied state back to the Open state. This enhancement allows you to easily rectify accidental closure of recommendations or marking Jira tickets as done by returning them to an actionable state.

Fixed issues​

• Nodepool recommendations displayed incorrect savings data. (CCM-12816)

Implemented a check to exclude nodepools that have more than one instance family from generating recommendations. This is necessary as the current calculations for such nodepools result in incorrect recommendations.

June 30, 2023, version 79906​

What's new​

• Azure inventory management (CCM-12676)

  As part of the Azure inventory management, now you can monitor the Memory Utilization data for virtual machines (VMs) along with the existing CPU Utilization tracking data.

• Clone AutoStopping rules (CCM-12337)

  You can now clone an AutoStopping rule. To clone a rule, navigate to the rule you want to replicate and select the Clone option from the more options menu. After selecting Clone, you can update the instance details according to your requirements. This allows you to create a new rule based on the existing one, saving you time and effort in setting up similar rules for different instances.

• Budget alert enhancements

  • The cost alerts for daily budgets are now triggered on an hourly basis. Previously, cost alerts for daily budgets were triggered only at specific intervals, which could potentially result in delayed notifications if the threshold was crossed outside those intervals. However, with the increased frequency, you can now receive timely alerts as soon as the threshold is exceeded, regardless of the time of day. (CCM-12028)

  • Significant improvements have been made to the Slack budget alert messages for both budgets and budget groups. Now, when receiving a budget alert, you will find detailed information related to the perspective from which the budget was created, the allocated budget amount, the current spend, and the forecasted spend. (CCM-12647)

    

Fixed issues​

• The cost data was not displayed on the Perspectives page. (CCM-12752)

  This was caused by the challenge of pushing large volumes of billing data into BigQuery. This issue has been resolved by adding support to upload and handle large datasets.

• Previously, users were unable to view and manage the ignored recommendations for EC2 instances. (CCM-13004) (ZD-46353)

The payload for adding EC2 recommendations to the Ignore List was incorrect. Now, the issue is fixed, and the functionality is working as expected.

June 21, 2023, version 79803​

What's new​

• Added a tooltip on the Cloud Integration page. (CCM-12559)

  In the Cloud Integration page, if the connector data is unavailable, the View costs link is disabled. However, as soon as the data becomes available, the link is enabled. Now, a tooltip providing a concise explanation as to why the link is disabled appears when you hover over the disabled link.

• Asset Governance filter panel enhancement. (CCM-12854)

  Previously, in the Asset Governance > Evaluations page, only the target accounts with execute permissions were included in the Target Accounts field in the filter panel. Now, this functionality is enhanced so that all target accounts with view permissions are also included in the list.

Fixed issues​

• The budget screen displayed inconsistent margins, leading to overlapping text in different columns. To address this issue, the columns in the budget list have been readjusted, ensuring that the text in each column no longer coincides with the text in adjacent columns. (CCM-10980)

• An error occurred with the HTTP AutoStopping rule. (CCM-12729)

  The detection of routing rules on the Azure Application Gateway was impacted due to the presence of an additional custom probe configuration. To address this issue, during the detection of routing rules for the specified port configuration, any custom probes are now ignored. However, the custom probe will continue to be utilized for the selected rule.

June 09, 2023, version 79701​

Early access​

Propagate force cool down (CCM-12338)

You can now propagate force cool down from primary rule to dependent rules.

Earlier, when stopping a rule from the UI, you had to stop its dependant rules one by one. With this enhancement, you can propagate the stop operation to dependant rules as well.

Propagating cool down to dependant rules is optional. You can stop the primary rule with or without propagating cool down to dependent rules.

June 06, 2023, version 79601​

What's new​

Cost Category enhancement (CCM-12585)

When building a cost category, it is now possible to incorporate another cost category as a rule. However, there are important considerations to keep in mind when using a cost category within your rule.

• You cannot include a nested cost category as a rule within another cost category if either of these cost categories contains a shared bucket.
• You cannot add the same cost category as a rule in the cost bucket.
• You cannot create cyclic nested cost categories, where a cost category is nested within each other.
• You can nest cost categories to a maximum of 20 levels.

Fixed issues​

• Budgets that contain the / character in their names were previously experiencing issues with correctly opening the budget details page. (CCM-12062)

  Previously, when the budget name appended to the URL contained the / character, it was treated as a separate route. This caused the browser to fail in loading the corresponding budget details. Now, before appending the budget name to the URL, it is properly encoded. This ensures that the browser handles the / character correctly, allowing the page to load as expected.

May 29, 2023, version 79505​

What's new​

Azure VM recommendations

Introducing Azure VM recommendations that identifies idle or under utilized VMs, ensuring efficient resource allocation and significant cost savings. For more information, go to Azure recommendations.

Fixed issues​

• The recommendations for workloads with containers that do not have properly set limits and requests displayed a value of -1 or null. (CCM-11765).

  This issue has been resolved.

• The users encountered an issue where they were unable to toggle the Hide progress page and Dry run options. Each time the toggle button was clicked, an error was thrown, preventing them from enabling or disabling these options successfully. (CCM-12438)

  The UI was relying on a deprecated API to update the rule, causing issues with toggling the values. This issue has been resolved by replacing the older API with a new v2 API. This update restored the functionality, allowing users to toggle the values.

• The ALB that was created in AWS did not appear on the Harness AutoStopping rules creation page. (CCM-12517)

  This issue has been resolved.

• The AutoStopping fixed schedule did not execute the rule as configured in the schedule settings. (CCM-12396)

  The issue was identified as missing triggers from Dkron, an external service, corresponding to the created schedule. To address this, a backup job was created. This backup job is designed to trigger the required operation in case the triggers from Dkron are missed. By implementing this solution, the system ensures that the necessary operation will still be executed even if the triggers from Dkron are not received.

• The users were not able to create an Azure Application Gateway successfully as the Azure function package was corrupt. (CCM-12550)

  Rolling back to the previous function package fixed this issue.

May 19, 2023, version 79400​

Fixed issues​

• Budget group missing from the Budget page. (CCM-12334)

  Previously, updating a budget group rendered its history irrelevant due to its dependence on child entities. However, this issue has been resolved by introducing support for modifying the budget group history during updates.

• An error occurred while attempting to save an AutoStopping rule with multiple proxy configurations and a custom domain in the GCP proxy. (CCM-12048)

  Saving the AutoStopping rule did not append custom domain providers for non-AWS cloud providers. This resulted in a validation error at the back-end. This issue has been resolved. The required field custom_domain_provider is now being set for all cloud providers.

May 05, 2023, version 79300​

Early access​

Asset Governance

The Asset Governance feature now includes support for access control through Role-Based Access Control (RBAC). This enhancement allows more granular control and management of permissions. For more information, go to Asset Governance RBAC.

This feature is behind the feature flag, CCM_ENABLE_CLOUD_ASSET_GOVERNANCE_UI.

Fixed issues​

• Previously, the budget amount in the monthly fields did not default to zero when selecting the yearly budget period. (CCM-12289)

  This issue is resolved. Now, when choosing the budget type as Specified amount, the budget amount for individual months correctly defaults to zero.

  

• Users couldn't dismiss the "How to get started with creating rules?" modal on the Cost Category page. (CCM-12278)

  The issue is resolved. Now, you can collapse the modal if you don't want to see it on the screen.

• Updated the default Budget Type as Specified amount. You could select Last period spend if you like to set up budget based on that budget type. (CCM-12254)

• A stopped AutoStopping rule displayed Scale down on the rule Details page. A stopped rule must display Scale up and a running rule must display Scale down. (CCM-11920)

  This issue is resolved now.

  

• The escape character &amp rendered incorrectly in the budget dashboard. (CCM-11683)

The issue is resolved now.

April 19, 2023, version 79104​

What's new​

• Recommendations enhancement (CCM-11769)

  A new filter has been added to recommendations, which allows the selection of the age of the recommendations. This filter allows you to specify how many days old recommendations should be included in the results.

Fixed issues​

• The Recommendations page displayed incorrect savings value. (CCM-12082)

  This issue has been resolved. The value in the grid now matches with the widgets.

• Spike in BigQuery cost. (CCM-12027)

  Limited the data queried by users with restricted access (granular RBAC enabled) by implementing a time filter of 30 days. These users can retrieve recommendations only from the past 30 days, effectively reducing the overall size of the query results.

• Modifying individual budgets within a budget group resulted in inconsistencies within the budget group as a whole. (CCM-11854)

  To fix this issue, you are allowed to modify only the budget type, budget amount, and configure alerts for individual budgets. You cannot modify other parameters.

• While configuring budget groups, you cannot add a negative integer in the Cascading > Proportionally field. The total sum of the proportions should always be 100. (CCM-11852)

  

April 05, 2023, version 79001​

What's new​

• Workload recommendations enhancement. (CCM-9161)(Zendesk Ticket ID 34658)

  Introduced support for 100th percentile in workload recommendations. Recommendations will be displayed for 100% usage of workloads.

Fixed issues​

• Updated the default perspective names from Aws to AWS and Gcp to GCP. (CCM-11770)

• Discrepancy in the number of EC2 recommendations. (CCM-11730)

  The Terminate-type recommendations were not being saved, and certain EC2 recommendations were disappearing after a specific interval of time. This issue has been resolved.

• The link to the perspective on the Anomalies page was incorrect. (CCM-11403)

  This issue has been fixed, and the link now directs to the correct perspective.

March 21, 2023, version 78903​

What's new​

• Enabled audit trail for budget groups. (CCM-11387)

  With this enhancement, you can track all CRUD operations such as Create, Delete, and Update related to budget groups.

• Display the AWS account ID and name on the Recommendations page. (CCM-11666)

  The AWS Account ID has been added to the ECS Services on the Recommendations list page.

• Cost category enhancement (CCM-10580)

  Introduced support to allocate the cost of shared cost buckets by a fixed percentage among each cost bucket. A new user interface has been developed.

  

  For more information, go to Use Cost Categories.

Fixed issues​

• The error message displayed while creating a Jira ticket to apply recommendations was not meaningful. (CCM-10822)

  A comprehensive Jira error message will be displayed in the user interface whenever it is feasible. However, there may be instances where only a generic message such as "Error creating issue" will be displayed when the Jira error is not parsed.

March 13, 2023​

What's new​

• Cost Category enhancements (CCM-10280)

  • When calculating the cost for Unattributed, the rules present in the shared cost bucket are not considered to eliminate duplicate costs.
  • If Cost Category is NOT NULL in a perspective, it means all cost buckets are considered. Unattributed is not taken into account.
  • If the Cost Category is NULL, it indicates that the cost buckets are not considered in the perspective. Unattributed is taken into account.
  • Previously, all shared cost buckets were displayed as No Groupby. Now, when you apply a GroupBy option other than the cost category, the cost of the rules present in the shared cost bucket are displayed in a separate entity based on the GroupBy selection you have made. However, it is important to note that this change will be effective only if you have incorporated cost category with shared buckets in perspective rules.

Fixed issues​

• Previously, deleting a cost category caused the perspectives that utilized the cost category in their rule or GroupBy to crash. (CCM-9902)

  This issue has been fixed. Now, before the cost category is deleted, you will receive a prompt to perform one of the following actions:

  • Delete any perspectives that depend on the cost category.
  • Remove the cost category from the perspective's rule or GroupBy.

• When you edit an AutoStopping rule in which the Health Check option is disabled, it is enabled when the client-side data is fetched. (CCM-11472)

  This issue has been fixed. Now, the Health Check option remains disabled when you edit the AutoStopping rule.

March 06, 2023​

Fixed issues​

• The ECS service billing data was missing in the Perspectives. (CCM-11464)

  This issue has been fixed, and all data is now accurately reflected on the Perspectives page without any errors.

• The ANOMALY_DETECTION_CLOUD job responsible for displaying cloud anomalies was not being executed in accounts without a cluster connector. (CCM-11228)

  This issue has been fixed, and anomalies are now computed even in the absence of a cluster connector.

• Previously, when attempting to delete a Recommendation filter, an error message would appear.(CCM-11300)

  This issue has been fixed now, and you can successfully delete a Recommendation filter without any error messages.

March 01, 2023​

What's new​

• Introducing support for adding more than one CCM GCP connector when you have two or more billing export tables with different billing account IDs in the same dataset. (CCM-11244)
• Introducing support for assigning a custom static port as the source port in the port configuration of the TCP traffic-based AutoStopping rule. (CCM-11264)

Fixed issues​

• Previously, the Start Date selected when creating a budget was not being saved and instead the date of budget creation was being displayed as the Start Date. (CCM-10952)

  This issue is fixed now, and the Start Date selected during budget creation is now being saved correctly.

• Previously, even when the Cascading option was turned off, the budget amount was being equally divided among all the budgets in the group. (CCM-10950)

  This issue is fixed now. The budget amount is no longer being distributed among individual budgets, ensuring that the budget amount of each budget remains unchanged.

• If a health check status code is not entered for the AutoStopping proxy, Harness falls back to using the default range of 200-299. (CCM-11007)

February 20, 2023​

What's new​

• AutoStopping Proxy for HTTPS and TCP connections.

Harness CCM introduces AutoStopping Proxy to support AutoStopping for HTTPS and TCP connections.

Fixed issues​

• The potential monthly savings displayed on the UI did not match with the Spot or On-Demand recommendations. (CCM-10698)

  The logic to calculate the potential monthly cost displayed on the UI has been fixed. Now, the savings match with the Spot or On-Demand recommendations.

• Added the missing instance family types for Azure node pool recommendations. (CCM-10246)

• When you create a budget with an invalid Period Starts from date with respect to the Budget Period, the error message displayed was unclear - "Invalid request: Error in create budget operation. Start time of budget is invalid." (CCM-10487)

Now, the message clarifies why the date is invalid - "Invalid request: Budget Period and Period Start date cannot add up to be in the past."

• The cost details API was returning only the AWS account ID without the account name. (CCM-10573)

Now, the API returns both account name and ID.

January 31, 2023​

Fixed issues​

• Hourly data on the Perspectives page showed an incorrect billing amount for multiple accounts. CloudFunction was unable to delete the existing records but continued ingesting a new entry in clusterDataHourly in BigQuery. (CCM-10711)

  This issue is fixed. Now, the Instance_Billing_Hourly job execution is limited to 5 times per minute to avoid CloudFunction failure.

• The total costs displayed on the Overview page and the license-util page (or API) were incorrect in accounts with at least one Azure connector. (CCM-10678)

  A bug fix in the ingestion of aggregated costs for Azure resolved this issue.

January 18, 2023​

Fixed issues​

• While creating a Jira ticket to apply EC2 recommendations, the Account Name field in the Jira description incorrectly displayed the Account ID. (CCM-10507)

  Now, the issue is fixed, and the account name is displayed correctly.

January 04, 2023​

Early access​

• Standardize your currency across Harness CCM (CCM-9280)

  This release introduces Currency Preference that enables you to view the entire CCM application in your preferred currency for different cloud providers. This feature is behind a feature flag CCM_CURRENCY_PREFERENCES.

• API implementation for the Currency Preferences feature (CCM-9632)

  You can now use the Currency Preference API to select the currency in which you want to view your entire CCM application across different cloud providers. Go to Harness API Documentation for more information.

{
"status": "SUCCESS",
"data": [
{
"dashboardName": "AWS Cost Dashboard",
"dashboardId": "226",
"cloudProvider": "AWS",
"description": "Discover and track analytical insights into your AWS cloud costs",
"serviceType": "",
"redirectionURL": "#/account/H5W8ioxxxA2MXg/dashboards/folder/shared/view/226"
},
...
{
"dashboardName": "AWS RDS Inventory Cost Dashboard",
"dashboardId": "3309",
"cloudProvider": "AWS",
"description": "Overview of cloud spend across RDS instances, breakdown by accounts, regions, instance types etc.",
"serviceType": "AWS RDS",
"redirectionURL": "#/account/H5W8ioxxxA2MXg/dashboards/folder/shared/view/0000"
}
],
"metaData": null,
"correlationId": "bc71c537-048f-4d53-80cd-8462158e1471"
}

Cloud Development Environments Release Notes

Review the notes below for details about recent changes to Harness Cloud Development Environments.

August - [2025.08.v1]​

With the last release, we introduced Self-Hosted Gitspaces. This release builds on that foundation with several key new features for Harness Gitspaces. Let’s dive into the details.

⭐️ [New Feature] AWS Support for Self-Hosted Gitspaces​

👉 Read more about this feature

We’re excited to announce AWS cloud infrastructure support for Harness Self-Hosted Gitspaces. This milestone brings greater flexibility and control, enabling you to create and host Gitspaces within your own AWS Cloud Infrastructure. This not only provides you with greater control over your Gitspaces, but also provides an extra layer of security over your infrastructure and source code data.

To get started, configure your AWS cloud infrastructure in Harness, set up the Terraform module to provision required AWS resources, and install the VM Runner and Delegate on your AWS VM instance. You’ll then be able to create and manage Gitspaces within your own AWS infrastructure.

⭐️ [New Feature] Access Control - RBAC for CDE​

👉 Read more about this feature

This release introduces RBAC for CDE (Role-Based Access Control). RBAC lets you control who can view, access, and modify resources across different scopes. By defining roles and resource groups, you ensure that only authorized users can act on CDE resources, adding an additional layer of security and compliance.

CDE RBAC supports the following resources:

⭐️ [New Feature] Gitspace Admin Settings​

👉 Read more about this feature

We’re introducing Gitspace Admin Settings, which allow you to customize Gitspaces to align with your organization’s standards and policies. These settings provide fine-grained control over:

• Git Providers – Manage which Git providers are available for creating Gitspaces.
• Code Editors – Control which code editors are available to users.
• Cloud Regions & Machine Types – Define the infrastructure regions and machine types available.
• Gitspace Images – Restrict which container images can be used to spin up Gitspaces.

⭐️ [New Feature] Reset Support in Gitspaces​

👉 Read more about this feature

This feature enables you to reset a Gitspace, reverting it to the latest configuration from the source. Resetting ensures your Gitspace reflects the most recent configuration or infrastructure changes while preserving your home directory data.

⭐️ [New Feature] Custom Machine Image Support for Self-Hosted Gitspaces​

👉 Read more about this feature

You can now configure custom machine images for Self-Hosted Gitspaces. This feature allows you to:

• Use a custom machine image for your infrastructure gateway.
• Configure a custom image for the VM Runner.
• Provide a custom machine image for your Gitspaces machine.

When configuring infrastructure, you can directly specify a custom machine image (GCP) or custom AMI ID (AWS).

⭐️ [New Feature] On-Prem Git Providers Support​

👉 Read more about this feature

Harness Gitspaces now support on-prem Git providers. With this capability, you can spin up Gitspaces using source code repositories hosted within your own infrastructure. This provides an added layer of security and control over your source code and metadata and ensures your data never leaves your server and infrastructure.

🚀 Releasing Self-Hosted Gitspaces — July [2025.07.v1]​

Introducing Harness Self-Hosted Gitspaces​

We’re thrilled and excited to announce the release of Harness Self-Hosted Gitspaces — a big step forward in improving developer experience and productivity at enterprise scale.

Self-Hosted Gitspaces are on-demand, remote cloud development environments hosted within your organization’s own infrastructure. These environments come pre-configured for instant coding and provide an extra layer of security by giving you full control over infrastructure and data. This helps reduce the risk of external data exposure and ensures your source code isn’t cached or accessed by third-party cloud services. Read more about Self-Hosted Gitspaces.

Why Self-Hosted?​

Local development environments often create more problems than they solve — inconsistent setups, endless configuration hours, security and compliance risks, and the list goes on. Harness Hosted Cloud Development Environments solve these challenges by delivering speed and simplicity to developers.

Harness-Hosted Gitspaces are fully managed by Harness and reduce setup complexity for end users. However, as organizations grow and adopt stricter security and compliance standards, Harness-Hosted Gitspaces can pose some enterprise-level challenges:

• Security & Data Sovereignty: Lack of full control over source code and developer data.
• Customization Limitations: Hosted solutions may not support deep customization of tooling and environment setup.
• Latency Issues: Fixed hosted regions can lead to performance issues for globally distributed teams.

In contrast, Self-Hosted Gitspaces solve these challenges by offering complete infrastructure and data control.

Read more about the different Use Cases of Self-Hosted Gitspaces.

Get Started​

Refer to Configuring Self Hosted Gitspaces to get started with setting up Self-Hosted Gitspaces. If you’re interested, here’s what you need to do:

1. Go Through the Prerequisites and Fundamentals: Before you begin, it’s important to understand the underlying architecture and concepts of Self-Hosted Gitspaces.

   • 🔗 Self-Hosted Gitspaces Architecture
   • 🔗 Key Concepts

2. Configure Gitspace Infrastructure via Harness UI: Start by configuring your Gitspace infrastructure in the Harness UI. This lets you input your infrastructure details, which will be referenced later during provisioning.

   • 🔗 Configure Gitspace Infrastructure via Harness UI

3. Configure and Set Up the Terraform Module: Once the infrastructure is configured in the UI, initialize and use the Harness Gitspaces Terraform Module. This module will create all required GCP infrastructure and set up VM instances for the CDE Gateway.

   • 🔗 Set Up Terraform Module

4. Set Up Runner and Install Delegate: After provisioning, install the VM Runner and Delegate on your GCP VM instance. This ensures the Harness Control Plane can establish a connection with your infrastructure to create and manage Gitspaces.

   • 🔗 Set Up Runner and Install Delegate

5. Create Machines & Gitspaces: Once everything is set up, you can start creating machines and Gitspaces directly.

   • 🔗 Create and Manage Machines

What’s Coming Next?​

We have a lot of exciting features on the way:

• AWS infrastructure support for Self-Hosted Gitspaces
• RBAC for admin functionality
• Admin-controlled features to manage Gitspaces access
• Support for custom machine images
• …and many more!

✨ Stay tuned for our next release - happening soon!

Code Repository release notes

These release notes describe recent changes to Harness Code Repository.

March 2026​

Version 1.80.x​

New features and enhancements​

• The repository summary page now displays a language breakdown showing the programming languages used in the repository. Languages are also computed automatically when importing a repository, so imported repos show language information immediately. (CODE-5093, CODE-5177)

• Forked repositories are now restricted to private visibility. Forks of private repositories can no longer be changed to public, preventing accidental exposure of private source code through forks. (CODE-5094)

• Improved error messages for invalid branch names. When creating a branch with invalid characters (for example, using a commit message containing spaces as a branch name), the error message now clearly indicates which characters are invalid and what naming rules apply. (CODE-4900)

• Added maximum length validation for summary and link fields in status check reports, preventing oversized payloads from causing issues. (CODE-4786)

• The new Code Repository UI now fully supports public repositories, including unauthenticated access to the repository summary page, file browsing, commits, branches, and pull requests. (CODE-4513)

Bug fixes and improvements​

• Fixed an issue where the compare branches and pull request commits tabs only displayed approximately 15 commits, even when many more existed between the compared branches. Pagination is now available to navigate the full commit list. (CODE-5182)

• Fixed an issue where file attachments added to a pull request description were lost after creating the PR, particularly when using AI-generated summaries or PR templates. (CODE-4677)

• Fixed an issue where the gitignore template listing API incorrectly included non-template files (such as README.md) in the response. The API now also returns HTTP 400 instead of 500 for unsupported template requests. (CODE-5162)

• Fixed an issue where branch protection rules were not properly enforced when the commit API used full branch references (refs/heads/ prefix) instead of plain branch names. (CODE-4920)

• Fixed an issue where the repository listing API returned HTTP 500 when paginating through repositories that contained forks whose source repository had been deleted. (CODE-5137)

• Fixed an issue where the repository settings UI did not allow selecting a validation pipeline in branch rules until the pipeline had been triggered at least once. (CODE-5198)

Version 1.78.x​

New Repository Settings and Rules Interaction​

As part of ongoing improvements to repository restriction enforcement, we have a long-term project to refine the interactions among Settings, Push Rules, Branch Rules, and Tag Rules.

Settings that overlap with push rules will be removed in a future release, as their functionality is being replaced by rules.

During this transition, we have identified issues with how some settings were being enforced. Some were not enforced as intended, and changes will be deployed in the coming weeks to ensure they are applied correctly. This may result in breaking behavior for configurations that relied on the previous unintended evaluation. (CODE-5067)

Customers are strongly encouraged to prefer push rules over settings for new configurations where equivalent rule-based conditions exist, as rules support overrides and will remain the supported mechanism going forward.

Customers should review existing repository configurations and remove overlapping restrictions from Settings when equivalent Push Rules exist.

After upgrading, repository behavior should be verified, since some settings that were previously not enforced may now be applied correctly and can affect push operations.

The update will be part of the 1.78.x release (tentatively week of March 30, 2026).

For more information, go to Repository Settings and Push Rules.

Version 1.77.x​

New features and enhancements​

• The fork-sync API now returns HTTP 409 status code when sync encounters merge conflicts, replacing the previous 200 response. This provides clearer conflict detection for API consumers. (CODE-5104)

• Improved Git LFS upload performance by streaming file content during OID calculation, reducing memory usage for large file uploads. (CODE-4993)

Bug Fixes and Improvements​

• Fixed an issue where listing repositories in a forked repo's space would fail if the upstream (source) repository had been deleted. (CODE-5090)

• Fixed fork-related API endpoint definitions in the OpenAPI specification to return correct status codes. (CODE-4616)

February 2026​

November 2025​

Version 1.64.x​

New features and enhancements​

• Harness Code introduces a complete UI revamp, offering a faster, cleaner, and more intuitive experience with improved navigation, search, and code review workflows. Developers can now opt in or out to explore the redesigned layout and streamlined repository management before it becomes the default. This is currently behind the following feature flags CODE_ENABLE_MFE_V2 & PL_UNIFIED_OPT_IN_ENABLED that need to be enabled for the opt-in experience.

Bug Fixes and Improvements​

• Fixed an issue where filename with a special character "%" caused the file view or the landing page of the repository to break.
• Fixed syntax highlighting issue. Users will now see raw text without syntax highlighting if the underlying text has syntax errors.

September 2025​

Version 1.56.x​

New features and enhancements​

• Added support for including user groups in Code Owners and Reviewer assignments.

August 2025​

Version 1.54.x​

New features and enhancements​

• Added target selection to Account, Org, and Project Branch/Tag rules, allowing users to include or exclude specific repos by name or pattern.
• Added audit log support for Tag Rules, Webhook CRUD operations, and bypassed tag actions.

July 2025​

Version 1.50.x​

New features and enhancements​

• Added support for tag rules in Harness Code Repository. Users can now restrict tag creation, deletion, and updates with pattern-based rules.
• Added support for including user groups in the bypass list for branch rules and tag rules.

Version 1.48.x​

New features and enhancements​

• Added a banner prompting users to Create PR when a new branch is recently pushed.
• Added repository sort and scope filters to the repo list page. Users can now sort repositories and toggle between account, org, and project scopes to control visibility.

June 2025​

Version 1.46.x​

New features and enhancements​

• Added the ability to set favorite repositories across the account.

Fixed issues​

• Resolved an issue where Markdown in Harness Code did not render images.

Version 1.45.x​

New features and enhancements​

• Added Audit logs for force push to the default branch.
• PR dashboard enhancement.

May 2025​

Version 1.43.x​

New features and enhancements​

• Added support for Git Large File Storage (LFS) to enable versioning of large binary files in repositories.
• Added Jira integration to enable visibility between code repository and issue tracking workflows.

Version 1.42.x​

New features and enhancements​

• Added support for partial and shallow Git clones to improve performance and reduce footprint when cloning large multi-service repositories.

Version 1.41.x​

New features and enhancements​

• Added support to display the merge method used when merging a pull request.

• Added support to indicate if rules were bypassed during pull request merge.

January 2025​

Version 1.28.2​

New features and enhancements​

• Added Archive repository functionality
• Added networking and API support for SSH with git repositories
• Add checkbox for comment events and review events
• Support for Harness SMP - Self Managed Platform
• Support for Jira Cloud integration to link commits with issues
• List Status Checks for account/org/project

Fixed issues​

• We need org view and edit permissions in order to create rules/labels on the account level

December 2024​

Version 1.26.2​

New features and enhancements​