Release Notes 3.6.X

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.

Version 3.6.8

Date: 1st October, 2025

To use the latest Pulse 3.6.8 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.8.

For details about the supported upgrade path, see Supported Pulse Upgrade Path.

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

Fixed an issue that prevented users from adding visualizations to an existing dashboard on the Dashplots page after the first save. Users can now add and save multiple visualizations to the same dashboard after the first save without errors.

Version 3.6.7

Date: 24th July, 2025

To use the latest Pulse 3.6.7 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.7.

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

Resolved an issue where the Pulse UI page failed to reload when using regex filters in the custom dashboard search.
Resolved an issue with the Impala query details missing in the Pulse UI for queries where the query plan is larger.

Version 3.6.6

Date: 10 June, 2025

To use the latest Pulse 3.6.6 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.6.

This section consists of enhancements introduced in this release.

HDFS

Support for Large Quota Values: Pulse now supports processing larger DS and NS quota values up to exabytes (19 digits), enhancing its ability to handle high-volume data workloads.

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

Sr.No	Description
1	Resolved an issue where downloaded query results from Spark, Impala, and YARN pages did not match applied filters. The download now correctly reflects the selected filter criteria.
2	Resolved an issue where visualization charts with a logarithmic Y-axis scale were not saving correctly.

Version 3.6.5

Date: 31 May, 2025

To use the latest Pulse 3.6.5 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.5.

This section consists of new features and enhancements introduced in this release.

Alerts

New Alert Metric for NameNode RPC Calls: The namenode_topuser_opcounts_count metric is now available for alerting. It tracks the number of RPC calls to the NameNode, grouped by user and call type. You can now create alerts with this metric to notify on unusually high activity by specific users or operation types.

Dashplots

Highlight Data in Custom Dashplot: Pulse now provides you with a capability to select a column, specify the highlight type (row or column), and choose the highlight element (background or text). You can then configure values and assign color codes accordingly. The rows or columns that meet the specific conditions set are highlighted as configured, making it easier to identify and track data points in larger clusters. For more information, see Highlight Result Data with Color Codes.
Visualize HDFS DS and NS Quota Usage: The Path Analysis feature now includes HDFS quota values for the selected path, enabling visualization of usage relative to quota in charts on the Dashplots page.

Impala

Added New Metrics to Impala Queries Page: Pulse now displays key Impala metrics such as Statistics Corrupt and Statistics Missing on the Impala Queries page, enabling filtering and exporting of results based on these metrics.

These statistics metrics are also available on the Impala Query Details page in addition to the Impala Query page. For detailed information about these metrics, see Impala Queries.

Note These metrics are not visible by default; you can select them through the Settings button on the Impala Queries page.

Encryption Support

Encrypt and update credentials: Run accelo admin encrypt and provide a plaintext password or passphrase. Pulse now encrypts it, then you can manually update the relevant configuration files with the encrypted string.
- Secure LDAP connections: Encrypt the LDAP bind password, update it in the ldap.conf file, and ensure the encryptedPassword flag is set to true. For more information, see Encrypt and Update the LDAP Bind Password and Edit the LDAP configuration.
- Enable SSL for GraphQL: Encrypt the SSL passphrase used in the GraphQL layer and update it in thead-core.yml file. For more information, see Encrypt and Update Passphrase for GraphQL SSL Configuration.

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

Sr.No	Description
1	Resolved an issue causing lag in streaming YARN applications displayed on the Pulse UI.
2	Resolved an issue where newly created visualizations were not saving correctly on the Dashplots page.
3	Resolved an issue where the `Search_Hits` condition was missing from the drop-down when creating or editing an alert.
4	Resolved an issue where the number of regions and online regions for HBase tables were not displayed by adding a new variable to configure the request timeout for the hbazer.

Version 3.6.4

Date: 4th April, 2025

To use the latest Pulse 3.6.4 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.4.

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

Sr. No	Description
1	The issue where the /opt/acceldata directory was cleaned up during uninstallation, requiring hook jars to be replaced after every installation, is now fixed. This cleanup process previously caused issues during upgrades. Acceldata now ensures that the /opt/acceldata directory is not cleaned up during uninstallation.

Version 3.6.3

Date: 25th March, 2025

To use the latest Pulse 3.6.3 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.3.

This section consists of an enhancement introduced in this release.

Dashplots

Custom Dashboards: The Storage I/O Metrics and HDFS I/O Metrics dashboards are now available by default on the Dashplots page.

Storage I/O Metrics: View summary and line charts showing trends over the selected time range:
- Average Read and Write I/O bytes and time across DataNodes.
- Average Read and Write I/O bytes and time across MasterNodes.
- Average Read and Write I/O bytes and time for root mounts across MasterNodes.
HDFS I/O Metrics (Performance Tracking): View summary and line charts showing trends over the selected time range:
- Average Read and Write I/O bytes and time across DataNodes.
- Average Read and Write I/O bytes and time across the cluster.

Version 3.6.2

Date: 20th March, 2025

To use the latest Pulse 3.6.2 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.2.

Impala

Optimized performance for the Impala Connector.

Version 3.6.1

Date: 18th March, 2025

To use the latest Pulse 3.6.1 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.1.

This section consists of enhancements introduced in this release.

YARN Optimizer

Enabling Node Limit as Configured in YARN Configurations: Previously, each NodeManager was configured with the entire OS-level memory usage, increasing the risk of Out-of-Memory (OOM) exceptions. With this enhancement, only the specified memory limit configured in the Ambari UI will be considered. To enable this feature, see Enabling Node Limit as configured in YARN Configurations.

Logs

Logs Page Enhancements: The following improvements have been made to the Logs page:
- Page Navigation (Current Page Indicator): You can now click on the current page number to edit it and jump directly to a specific page.
- Pagination Controls: Added « (First Page) and » (Last Page) icons for quick navigation to the first and last pages. For detailed information, see Filtering and Viewing the Logs.

New Certification and Support

JDK-11 Certification: Pulse is now certified for JDK-11, which provides enhanced performance and new features and aligns with the latest Java standards.
Support for Spark 3.5.2: This release supports the Spark 3.5.2 version in the standalone environment.

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

Sr.No	Description
1	Fixed an issue where the Logs page was not properly redirecting from the Nodes page.
2	Fixed an issue with the Group By feature on the Logs page to ensure it accurately displays the correct information.

Version 3.6.0

Date: 21st February, 2025

To use the latest Pulse 3.6.0 version, you must upgrade your Docker version to 20.10.x or higher and perform the required migration steps. For more information, see Upgrade to Version 3.6.0.

This section consists of new features and enhancements introduced in this release.

New Pulse Alerts

Introduced the HDFS_PATH_USAGE alert: This HDFS alert is now available and it monitors the HDFS path's storage usage and triggers a notification when it reaches the specified threshold percentage. For example, if the threshold is set to 10%, the alert gets triggered when the path utilizes 10% of the total HDFS storage, ensuring proactive storage management. For more information, see HDFS Alerts. Note The HDFS paths must be present in the pathanalysis.txt file under $AcceloHome/data/fsanalytics/cluster/.
Introduced the KAFKA_READ_WRITE_SKEWNESS alert: This Kafka alert is now available and is triggered when a broker experiences a disproportionate read and write activity compared to other brokers. For more information, see Kafka Alerts.
Introduced the NiFi Connection Status Alert: Pulse now enables you to create the NiFi Connection Status alerts, alerting users of NiFi node disconnection events. This ensures proactive monitoring, enabling timely response to threshold breaches and maintaining system reliability. Note This functionality works on all the NiFi environments.
Introduced the Queued Flowfile Alert: Pulse now enables you to create and send alerts to monitor the queued flow file count for a connection based on the configured threshold.
Alerts Accept Regex in Filters: Pulse now supports regex matching in filters, allowing more flexible and dynamic alert configurations.

Reports and Dashboards

Generating Stock Reports: Pulse now enables you to generate and download a report to verify the number of applications running, their status (such as whether they are up or down), the hosts on which they are running, and other relevant details. For more information, see Generating the Stock Report.
Introduced the NiFi API Summary Dashboard: Pulse now introduces a new dashboard that distinguishes between external API requests (made by external agents such as users using the UI) and internal API requests (made internally by the nodes themselves, replicated across different machines of the cluster). For details about enabling the NiFi request metrics, see Enabling the NiFi API Summary Dashboard.

Actions and Executions

Introduced Yarn Application - Auto Killer Action: This new action enables automatic termination of Yarn applications that exceed specified conditions. Applications can be killed by entering the Application ID while also specifying queues where termination should be restricted. The termination criteria include Duration, Queue Usage Percentage, Memory Usage Threshold, and VCores Usage Threshold. For more details, see Yarn Application Auto Killer.
Enhanced Action Execution Tracking for Killed Queries: Pulse has now enhanced the Action Executions capability for actions that kill queries by categorizing them in red. By clicking on the red-categorized execution name, you can view the details of the killed queries, including the list of terminated queries, service names, application IDs, and reasons for termination, and also you can redirect to the respective components page to get more information about the queries. For more information, see Check Execution Logs and Killed Queries.
Enhanced “Move Pending App Queue” Action: You can now give multiple destination queue names (comma-separated) instead of a single queue name. This action now intelligently selects the queue with the most available resources from this list, maximizing resource utilization. You must also ensure that the provided queue list contains only queues where they have job submission permissions. For more information, see Move Pending App Queue.
Clean up HDFS and Local Files using the Access Time: You can now efficiently clean up both HDFS and local files by configuring the last access date (atime) through the respective actions, streamlining the file management process. For more information, see HDFS Log Cleanup and Local Log Cleanup.
Introduced Cron Generator on Actions: Pulse Actions service now enables you to create and manage cron expressions easily. This capability helps schedule tasks with precise time intervals, supporting various patterns such as minute, hourly, daily, weekly, and custom schedules. For more information, see Actions and Actions Service.
New Parameters Added to the HDFS Balancer Playbook: In this release, several new parameters have been introduced to the HDFS Balancer Playbook, enhancing its customizability and scalability. These additions enable fine-tuned resource management, optimizing performance, stability, and efficiency during balancing operations. For more information, see HDFS Balancer. The new parameters are:
- Mover Threads
- Dispatcher Threads
- Maximum Concurrent Moves
- Get Blocks Size
- Get Blocks Min Block Size
- Maximum Size to Move
- Block Size
- Maximum Idle Iterations
- Maximum Queries Per Second

Notification

Configuring Multiple Xmatters Groups for Notifications: Pulse now enables you to configure multiple Xmatters groups and keys using the accelo config alerts notifications command. Once configured, the group names will appear in the Alerts UI, allowing you to select the desired user group from the drop-down list to send notifications when an alert is triggered. For more information, see xMatters Notifications and Notifications.
Enhanced Quantum Notification Details: When an alert is raised, the Quantum Notification ticket now includes the environment or cluster details.

Security

Masked Spark Authentication Secret Key in Pulse: Pulse now enhances security by masking the spark.authenticate.secret in the Pulse UI > Spark > Jobs Compare Metrics section. This ensures that sensitive authentication details remain hidden, preventing unauthorized access and enhancing overall data protection.

Visualization and User Experience

Ability to Create Visualization for Killed Queries: Pulse now enables you to create dashplot or visualization for Impala Kill Queries and YARN Kill Applications, providing deeper insights into query termination patterns. These queries may have been killed due to factors such as high resource consumption, incorrect queue selection, or extended runtime. For more information, see Dashplot.
Enhanced "View in Context" Option in Logs UI: On the Logs page, when you click on three dots > View in Context to gain deeper insights into log details, the page design has been enhanced to provide a more comprehensive and seamless log viewing experience.

Other Enhancements

Optimized Ad-Alerts Component: The ad-alerts component has been optimized by implementing a caching mechanism for alerts, a thread pool mechanism for endpoint connection alerts, fixing connection leaks, and improving various Akka configurations. These enhancements have significantly improved the overall stability and performance of the ad-alerts system.
Enhanced the Data Download Capability: Pulse now supports downloading more than 10K rows of data in the CSV or XLSX format for modules such as Tez Queries, Spark Jobs, Impala Queries, and Yarn Application Explorer without RM Live. Note The maximum number of rows supported is 2 million for CSV format and 1 lakh for XLSX format.
Total Driver Runtime for Spark Jobs: Pulse now displays the total driver runtime for Spark jobs in the Spark Jobs Timeline section, providing better visibility into the driver execution runtime. For more details, see Spark Stages.
Introduced Queue-Level Job Details: On the YARN > Capacity > Queue Usage page, you can now click on a data point at specific intervals to view key details, including the following.
- Number of running queries
- User
- Job type (e.g., Spark, YARN)
- Queue Utilization by Applications
- Execution timestamp The purpose of this functionality is to display the top applications consuming the most queue capacity in the selected interval. This enhancement provides a more granular view resource utilization, improving cluster monitoring and troubleshooting. For more information, see Other Capacity Charts.

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

Sr.No	Description
1	Resolved an issue with scheduling the FSImage export action. The action now functions as expected when created and scheduled.
2	Resolved an issue where data was not displaying on the Pulse UI after importing custom dashboards by reducing the Yarn Application Interval to 5 seconds to capture the short-lived yarn application container logs.

This section consists of the security or vulnerability fixes made in this release.

Sr.No	Vulnerability	Severity	Description
1	Cross-Domain Misconfiguration	Medium	Fixed a cross-domain misconfiguration issue where improper CORS settings could allow untrusted domains to access sensitive data or functionality, posing risks such as data theft or Cross-Site Request Forgery (CSRF) attacks. The solution involved removing the custom CORS configuration, allowing the browser to enforce its default security policies for cross-origin requests.
2	Unencrypted communications	Medium	Fixed an issue where the application permitted unencrypted connections, increasing the risk of user data exposure to eavesdropping and manipulation, especially on public Wi-Fi, shared networks, or compromised infrastructure. To address this, we configured the server to enforce TLS 1.2+ with recommended cipher suites, ensuring all communications are encrypted and secure.
3	Graphql Nested Queries Denial Of Service	Medium	Fixed an issue where attackers could exploit deeply nested or overly complex GraphQL queries, leading to excessive resource consumption that could degrade performance, crash the server, or render it unavailable to legitimate users. This issue was addressed by implementing custom logic to restrict query depth to a maximum of 10 levels. Queries exceeding this depth now return a 500 status code, preventing potential abuse.
4	Cookie No HttpOnly Flag	Medium	Fixed an issue where cookies lacked the HttpOnly flag, allowing JavaScript access and increasing the risk of Cross-Site Scripting (XSS) attacks that could expose sensitive information such as session tokens. The issue was resolved by enabling the HttpOnly flag for XSRF-TOKEN and JWT cookies, preventing client-side scripts from accessing these cookies and reducing XSS risks. Additionally, client-side authentication was updated by replacing direct JWT cookie checks with two new dedicated endpoints.
5	Cookie without SameSite Attribute	Low	Fixed an issue where cookies were being set without the SameSite attribute, exposing them to potential Cross-Site Request Forgery (CSRF) attacks. Without this attribute, cookies could be sent in cross-origin requests, enabling malicious sites to exploit user sessions. This issue was resolved by initializing cookies, including those used for XSRF-TOKEN and JWT, with the SameSite=Strict attribute, ensuring they are only sent in same-site requests and mitigating CSRF risks.
6	Cookie without Secure Flag (JWT)	Low	Fixed an issue where the JWT cookie was not always set with the Secure flag, allowing it to be transmitted over unencrypted HTTP connections, which could expose it to interception and potential session hijacking. The issue was addressed by ensuring the Secure flag was applied based on the protocol. An environment variable has also been introduced as a fallback to override this condition, providing flexibility for different deployment environments.
7	X-Content-Type-Options Header Missing	Low	Fixed an issue where the server omitted the X-Content-Type-Options header, which could allow browsers to interpret files as a different MIME type, increasing the risk of content sniffing attacks. The issue was resolved by setting the X-Content-Type-Options header to nosniff, ensuring that browsers strictly adhere to the declared content types and preventing MIME type misinterpretation.
8	Information Disclosure - Debug Error Messages	Low	Fixed an issue where some API calls exposed full error stack traces, potentially revealing sensitive information about the application's internal structure. Proper error handling has been implemented now to ensure that only relevant error messages are returned, while full error details are securely logged on the server for internal debugging purposes.
9	Server Leaks Version Information via "Server" HTTP Response Header Field	Low	Fixed an issue where the web server exposed version details through the Server and X-Powered-By headers, which could aid attackers in identifying server software and exploiting known vulnerabilities. The fix involved removing the X-Powered-By header from all responses to prevent the exposure of server technology details.
10	HSTS Missing From HTTPS Server	Low	Fixed an issue where missing HSTS (HTTP Strict Transport Security) could expose users to man-in-the-middle attacks by allowing protocol downgrades. The Strict-Transport-Security header is now conditionally added for HTTPS connections, ensuring secure communication. Additionally, an environment variable, ENFORCE_HSTS, has been introduced to allow optional control over this enforcement if needed.

Last updated on

Was this page helpful?