Skip to main content

Release Notes (2026)

V2.4.2 (June 26)​

Guanyun Platform​

Fixes​

  1. Fixed an issue where the API returned errors when editing a notification policy whose recipient type is "Receiver Group".

  2. Fixed an issue where the Infra Agent could not be properly disabled due to an abnormal module switch state synchronization when the apm-dc-frontend service delivered configurations to UniAgent.

Lightweight Applications​

Tracing​

Fixes​
  1. Fixed an issue where the frontend component became unresponsive when clicking "Zoom In on Selected Area" after selecting a region on the heatmap.

V2.4.0 (May 10)​

Guanyun Platform​

New Features​
  1. Platform-Level Preference Settings: Supports unified configuration of platform appearance and display styles by super administrators, including custom Logo, browser title (Title), website icon (favicon), copyright information (Copyright), login landing page, and watermark. Also supports switching the left-side navigation mode between “Dock” and “Menu Bar” styles.
  2. metric：Added GenAI-OpenClaw series indicators.
Improvements​

  1. Platform Time Control: When the URL parameters contain second-level timestamps startTime and endTime, the system automatically converts and aligns them to minute-level granularity.

  2. Global Filtering: Supports hierarchical linkage filtering across four dimensions: business system, application, instance, and request.

  3. Performance Optimization: Optimized lightweight application UI loading performance to reduce white screen time during page initialization.

AI Assistant​

New Features​
  1. Quick View for Entities:
    • Quick-link navigation for business systems, applications, instances, requests, and hosts
    • Quick-link navigation for traces
  2. Multi-step Tasks in One Question:
    • Supports handling multi-step tasks within a single user question
  3. Content Export:
    • All AI replies can be exported and downloaded as Markdown files
  4. Business Analysis:
    • One-click analysis for database instances, database operations, and hosts
    • AI analysis for eBPF diagnostics
Improvements​
  1. Refactored the AI analysis workflow and simplified the structure of answers
  2. Optimized the conversation handling flow for a smarter experience
  3. Enhanced ServiceFlow analysis for applications and requests

Lightweight Applications​

Applications, Business Systems, Requests​

New Features​
  1. Security Data Display: Added security-related data display in the application overview page to enhance security observability for application runtime status.
Fixes​

  1. The response time breakdown list for ExternalTransaction now supports displaying request names and directly opening the request overview page.

  2. Added filtering capability to the response time breakdown view, allowing quick filtering of request data by conditions.

  3. Added default Web request filtering rules for applications, business systems, and global distributed tracing.

  4. Added SQL collection length configuration capability for global, business system, and application dimensions.

  5. Optimized the user experience of jumping from reports to help documentation, improving guidance efficiency.

  6. Added a “Settings” operation column to the first-level lists of business systems, applications, and instances for quick access to corresponding configuration pages.

  7. Fixed incorrect host and container parameters in the relationship graph on overview pages (affecting business systems, applications, and requests).

Improvements​

  1. Optimized list data query performance for business systems, applications, and requests.

  2. Optimized filtering capabilities for business system and application-level request lists, supporting filtering by entity name, entity attributes, and entity tags.

  3. Optimized query parameters for request overview and relationship data, standardizing queries to use entity IDs.

  4. Optimized error analysis capabilities for lightweight applications such as business systems, applications, and requests: supports directly locating error code nodes in trace details from the error trace list.

Database​

Improvements​
  1. Added a “Distributed Tracing” tab to the Database overview page, allowing users to view related call chain data from the database perspective.

Distributed Tracing​

Improvements​

  1. Added a status code column to the distributed tracing list and added status code tags in the trace detail header area.

  2. Added log viewing capability in trace details, supporting viewing application logs related to the call chain.

Error Analysis​

Improvements​

  1. Added analysis tabs such as StackTrace, RootCause, Referring Pages, Client IPs, and Caller Applications to the error analysis details page, and added application tag display in the header area.

  2. Added time range zooming by drag-selection in error analysis charts, supporting refined analysis of error trends, error type breakdowns, and error counts.

  3. Automatically fills business system IDs and application IDs into the URL during error analysis page initialization, and synchronizes URL parameters when switching dropdown selections.

Host, Kubernetes​

Improvements​
  1. Added log display capability to Host and Pod overview pages, supporting viewing related log data.

Real User Monitoring​

Improvements​
  1. Aggregation Optimization:
    • Backend / HTTP: Automatically detect and aggregate similar URIs/backend services to avoid metric data fragmentation.
    • Web / Mini-programs: Also support URL aggregation on the Web side.
  2. Client Applications — Mobile:
    • Performance: Optimized rum-dc performance to reduce CPU usage.
    • Symbol-table parsing enhancement: Improved handling logic for very large iOS symbol tables (>500 MB) to address occasional timeouts when parsing very large projects.
  3. Client Applications — Web:
    • Click events: By default, the Web client will no longer automatically collect user click actions captured by probes.
Fixes​
  1. Client Applications — Mobile:
    • Network: Fixed an issue where, after enabling national cryptography (SM) encryption, the client would by default perform a second compression because it did not detect the backend load‑balancer state, resulting in reduced transmission efficiency or parsing failures.

Alarm​

New features​
  1. Entity-based alert rules support grouping; grouping is hidden for VIP accounts.
  2. Alert short links and opening alerts via short links without a password.
  3. On alert details, notifyees now include the notification content.
  4. Suspected issues list: added notification policy name; click the right panel to view details.
  5. Suspected issue details: added notification display.
  6. Record and display successful alert notification API invocations.
  7. Alert notification templates: added variable and dictionary descriptions.
  8. Alert details: adjusted parameters passed when navigating to multi-dimensional exploration (request name is included in the conditions).
  9. After drilling down from the alert list into details and returning, the previous list filters and state are preserved.
  10. Suspected issues list: added notification policy name; click the right panel to view details.
  11. Alert configuration: the notification policies shown must match those on the detail page.
  12. Suspected issues list: show alias first; fall back to the entity name when no alias exists.
  13. Configuration file: added per-product switches to disable “receiving” reported events from agents.
  14. Alert notification templates: added an “entity ownership” field variable.
  15. Alert configuration (entity-based): added “grouping”.
  16. When the “Invoke HTTP service” entity is selected, grouping automatically adds the “URI” dimension. When “Configure by metric model” is selected, metrics related to HTTP service calls also automatically include the “URI” dimension in grouping.
Improvements​
  1. Updated notification channel icons for notifyees on alert details.
  2. Addressed cases where, before an alert cleared, restarting Flink in standalone mode or an abnormal Flink restart in distributed mode prevented the alert from clearing, or data above the threshold triggered the alert again.
Fixes​
  1. Notification policies (Guanyun): when an alert rule referenced in filter conditions was deleted, residual IDs could still appear.
  2. New notification policy configuration: filter dropdown values could not scroll.
  3. Alert configuration ordering was unstable when customers edited rules (order appeared to follow last modified time); ordering now follows creation time.
  4. After modifying an alert rule, the alert incorrectly appeared as deleted in the alert list.
  5. Alert configuration with the intelligent AI algorithm: data exceeded the threshold but no alert event was raised.
  6. Alert details: external service entities should not support navigation.

V2.3.0 (January 06)​

Lite App​

Real User Monitoring​

Improvements​
  1. RUM - Global Settings:
    • Request settings - key request settings: Added English language support for request templates.
    • Request settings - key request settings: Added English language support for the export list.
  2. RUM - Mobile:
    • Creating application page: Added English language support for in-page help content.
    • Single sample details: Added English language support for region and device information.
    • Crash details: Added English language support for region information.
    • OOM details: Added English language support for device and exception information.
    • Network detection: Added English language support for IP identification in MTR task details.
    • User analytics: Added English language support for the active users by region chart.
  3. RUM - Web:
    • Nacos configuration: Added a switch for pushing data to Guanyun, controlling whether enable/disable operation data is pushed.
    • Creating application page: Added English language support for help content.
    • Application settings: Added English language support for NPM prompt messages.
  4. RUM - MP:
    • Page names now follow the aggregated naming rule.
    • Requests without Content-Type are categorized as API in resource_type.
    • Request overview: Added resource interface to the relationship topology.
    • Creating application page: Added English language support for help content.
  5. Session Replay:
    • User analytics: Added English language support for region and device information.
    • User analytics - session details: Added English language support for user information.
  6. RUM / Session Replay:
    • Improved the English online help manual.

Observe and explore​

Improvements​
  1. Dashboards:
    • When the language of Guanyun platform is switched to English, dashboards display English templates by default.
    • In dashboards' Chinamap, “Nanhai Zhudao, the South China Sea Islands” is displayed in English.
  2. Alerts:
    • In exported Excel files for alert rules, the “Enable/Disable” field is displayed in English.

APM​

Improvements​
  1. Improved the English online help manual.

Technical Support​