Version 4.0.0

Disclaimer All versions of Pulse are supported for maintenance for a period of 18 months following their release. This maintenance support includes updates for security vulnerabilities, bug fixes, Common Vulnerabilities and Exposures (CVEs), and the introduction of new features, among other improvements. After the 18-month support period, we strongly recommend transitioning to the most recent version or to another version that continues to receive support. This ensures an enhanced user experience.

Release Date: 18th November, 2025

New Features

This section consists of new features introduced in this release.

Refreshed Pulse UI and Admin UI

• Refreshed Pulse UI: Pulse introduces a refreshed, redesigned UI with a cleaner and more intuitive layout that enhances usability and optimizes screen space for better visibility of insights.

  • Key Highlights:

    • Redesigned pages such as Home, YARN Optimizer, Nodes, Admin, Alerts, and more
    • Enhanced workflows for simpler, more efficient operations
    • Faster performance and improved responsiveness across the platform

  • Access Details:

    • Docker-based deployment: The new Pulse UI is enabled by default. To access the legacy UI:

      • Run the accelo deploy addons command on the Pulse node.
      • Select the service Old UI from the list of services to deploy
      • Use port 4100 (https://<pulse-ip>:4100).

    • Kubernetes-based deployment: The new Pulse UI is enabled by default, and the legacy UI is not available.

    • If you are accessing both new UI and old UI, ensure to use the same protocol, either HTTP or HTTPS, for proper functionality.

• Pulse Admin UI: Pulse introduces an Admin UI. This is a web-based application for managing and monitoring the Pulse environment on Kubernetes. It provides a centralized interface to configure clusters, deploy services, upload certificates, manage configurations, and perform administrative actions on Pulse actions. For more information, see Pulse Admin UI.

New Service Observability

• ClickHouse Observability: Pulse now provides real-time observability for ClickHouse, allowing you to monitor component health, query performance, and resource utilization. You can analyze logs, detect performance issues, and build custom visualizations for continuous optimization. For more information, see ClickHouse.
• Flink Observability: Pulse now provides observability for Flink, enabling real-time monitoring of component health, applications, jobs, and vertex-level performance. You can analyze logs, identify performance bottlenecks, and create custom visualizations to enable continuous optimization and faster troubleshooting. For more information, see Flink.
• MLflow Observability: Pulse now provides observability for MLflow, offering real-time monitoring of MLflow Server health, log analysis for troubleshooting, and alerting for service availability. For more information, see MLflow.
• Standalone Kafka 3 cluster: Pulse now supports monitoring of standalone Kafka 3 deployments with Kraft and Zookper modes, enabling real-time visibility, health monitoring, and observability for Kafka environments. For more information, see Configure Pulse to Monitor Standalone Kafka 3 and Kafka and Kafka 3.

Anomaly Detection and Forecasting

• Anomaly Detection and Forecasting: Pulse introduces advanced capabilities to enhance proactive monitoring and insights.
  • Anomaly Detection: Automatically identifies and highlights abnormal data points in time-series charts based on historical trends.
  • Forecasting Alerts: Predicts future metric behavior using trained models and triggers alerts in advance when forecasted values exceed defined thresholds.
  • For more information, see Anomaly Detection and Forecasting.

Other Features

• Introduced Service Profiler: The Service Profiler in Pulse lets you capture and analyze time-based Java service performance using Java Flight Recorder (JFR). It enables profiling of JVM-based applications to identify CPU bottlenecks, review JFR files for offline analysis, and monitor profiling status directly from the Pulse UI. For more information, see Collect Service Profiling Data.
• Introduced Canary Test for HDFS, Impala, and Hive: Pulse now automates canary tests for HDFS, Impala, and Hive to monitor real-time performance metrics, including data transfer rates and query execution times. These automated tests help detect performance issues early and ensure cluster reliability. For details on setup and visualization, see Set up and Manage Canary Tests.
• Introduced Metric Search Capability: Pulse now introduces the Metric Search capability that enables you to quickly find and understand metrics and their purpose within the platform. By simplifying metric discovery and providing semantic search, Metric Search improves observability workflows and helps users make faster, data-driven decisions. For details, see Find and Understand Metrics.
• HDFS Snapshot Support: Pulse now provides visibility into HDFS snapshots, allowing you to monitor and review all directories where snapshot capability is enabled. You can view snapshot-enabled directories, track snapshot counts, quotas, permissions, and ownership details, and see all snapshots associated with each directory for better data backup and recovery, auditing, and change tracking. For details, see View HDFS Snapshots.
• Track Activity Logs on Audit: The Pulse Audit capability now tracks user activity across Dashplot Dashboards, Dashplot Visualizations, Alerts, and Cleared Incidents. You can search, filter, group, and download activity logs for improved visibility, compliance, and troubleshooting. For details, see Track Activity Logs.

• Redesigned YARN Optimizer: Pulse now features a redesigned YARN Optimizer with enhanced visualizations that display resource usage before and after optimization. You can configure application fingerprinting, YARN Agent metrics collection, and optimization settings directly from the UI for improved control and efficiency. For details, see YARN Optimizier.

New Alerts

This section consists of the new alerts introduced in this release.

• NiFi Processor Stale Alert: Pulse introduces a new NiFi Processor Stale Alert. NiFi maintains process versions in the NiFi Registry (for example, V1, V2, and so on). When a process is updated in the registry, some users might still be running an older version. When enabled, this stock alert detects and triggers whenever a user continues to use an outdated process version while a newer version is available in the registry. For details, see NiFi Alerts.
• Hive Metastore Table Lock Alert: Pulse allows you to create a custom alert for Hive Metastore Table Locks. The alert triggers when enabled, and the Hive table lock properties reach a defined threshold condition. For example, when the sum or max of lock properties (such as Locks Acquired At, Database Name, or Host Name) meets a specified operator (=, >, <, ≥, ≤) and threshold value. For more information, see Hive Locks.

Enhancements

Observability and Insights

This section consists of the observability and insights enhancements introduced in this release.

• Enhancements in Trino Observability: Pulse now collects additional metrics such as killed queries, failed queries, total available processors, etc., for creating custom Dashplots and Alerts. A query trend chart has been added to the Trino Query Details page, and a new sortable tabular view displays query stages and tasks (similar to Spark), improving visibility into long-running and failed stages. For more information, see Query Trends.
• Zookeeper Metrics Collection: Pulse now collects additional ZooKeeper metrics, including each node’s operational mode and active client connections. These metrics can be used to create custom visualizations and dashboards to monitor cluster health and identify leader or follower nodes.
• Introduced Table Lock Details: Pulse now provides detailed Table Lock Monitoring on the Databases page. You can view the total number of active locks per table and access lock summaries, including lock state, type, timestamp, and user information, through a popover. For more information, see Monitor Databases.
• Enhancement in Kafka Consumer Dashboard: In this release, the Kafka consumer dashboard has been updated to display consumer group information even when some topics have no records. Previously, if any topic in a consumer group contained null offset or lag data, the dashboard did not display that group. The updated behavior now:
  • Displays lag and offset details for active topics, even when other topics in the same group are empty.
  • Handles null or “–” values for offset and lag without causing errors or missing data.
  • Generates debug log messages instead of exceptions when offset or lag information is temporarily unavailable.
  • This enhancement improves visibility into consumer group performance and simplifies monitoring in environments where some topics may not always contain data.

Visualization and User Interface

This section consists of the visualization and UI enhancements introduced in this release.

• Added Nodes Dashboard: The Nodes > Dashboard page now provides visual charts for cluster and node performance, including disk, HDFS, and network metrics—helping you monitor health, analyze trends, and identify bottlenecks. For details, see Visualize Nodes Dashboard Metrics.

• HDFS File Explorer: The HDFS > File Explorer page in the UI has been enhanced with new capabilities for improved usability and visibility:

  • Search and Filter: Added a new Search option to locate and filter files and folders for faster access.
  • New Columns: Introduced two new columns, Replicated Size and File Replication, to provide better insights into storage utilization and replication behavior.
  • For details, see HDFS File Explorer.

• Export and Maximize Charts: Pulse now introduces two new features for charts. These enhancements enhance data accessibility and flexibility in visualization within the Pulse UI.

  • Export: Download chart data in XLSX or CCS format for reporting and analysis.
  • Maximize: Expand charts to a full-screen view for detailed inspection.

Actions

This section consists of the actions-related enhancement introduced in this release.

• Introduced YARN Kill Applications action: Pulse introduces a new YARN Kill Applications action runbook that automatically terminates YARN applications based on defined criteria. You can schedule this action using a Cron Expression to specify when Pulse should run the task and kill matching applications automatically. For more information, see Create an Action Plan > Kill YARN Applications and Schedule Action Plan.

Dashplots

This section consists of the dashplot-related enhancements introduced in this release.

• Introduced Interaction in Dashplots: This release adds a new Interaction option in Dashplots that allows you to create drilldown redirections between dashboards or to custom URLs for deeper data exploration. For detailed steps, see Dashboard Chart Options.

• Added Analytics Dashboards: The Dashplots Studio > Dashboards page now includes analytics dashboards to track and identify which users access specific data (tables, columns, and fields) across Hive, Tez, and Impala components. These dashboards are available for detailed user-level insights; User-Analytics-Hive, User-Analytics-Tez, and User-Analytics-Impala.

• Custom Dashboard Filters: Pulse now includes a “Select All” option in the Custom Dashboard filters. Previously, when grouping data by filters such as User, only the top five entries were selected by default, and users had to manually select the remaining options. With this enhancement, you can now use “Select All” to quickly include all available filter values at once, making it easier to perform broader analyses without manual selection.

Deployment

This section consists of deployment-related enhancements introduced in this release.

• Pulse Hydra Installer Support for LDAP/AD Users: The Hydra installation script in Pulse now supports reading service accounts directly from LDAP/Active Directory (AD). This update allows the installer to detect and use LDAP/AD-managed users, such as ad-pulse, without requiring them to be created locally on the host.
• New Agent Mpack version: This release introduces a new Hydra Agent Mpack version 2.0.1 that is compatible with ODP 3.2.2.0-2 (Python 2.7) and ODP 3.2.3.0-3 (Python 3.11). The Mpack versions 2.0.0 and 2.0.1 have the capability to limit CPU utilization of Hydra agents.
• Support for JMXTrans-Agent to prevent JMX port conflicts: Pulse introduces support for using JMXTrans-Agent as an alternative to PulseJMX to prevent JMX port conflicts in environments where multiple JVM processes run concurrently. This improvement enhances the stability and performance of Hive jobs during Pulse operation. For more details, see Enable JMXTrans Agent.
• Support for Azul OpenJDK 17: Pulse now supports Azul OpenJDK 17, ensuring full compatibility across Pulse containers and agents. This enables seamless deployment in environments standardized on Java 17, including enterprise Hadoop platforms.
• Introduced a New Hook: Introduced ad-hive-hook_odp_3.3.6_java11 for Java 11 compatibility and support for the updated ODP components, such as Hive 4.0.1 and Tez 0.10.4. If you are using Pulse 3.8.x or later with ODP 3.3.6.x.x, ensure to use this hook. For details, see Configure ODP Hive and Tez.

Resolved Issues

This section consists of issues that have been fixed in this release.