Red Hat Training

A Red Hat training course is available for Red Hat JBoss Operations Network

3.3 Release Notes

Red Hat JBoss Operations Network 3.3

Important Release Information for Red Hat JBoss Operations Network

Jared Morgan

Zach Rhoads

Ella Deon Ballard

Abstract

These release notes contain important information about new features, known issues, and other technical notes available at the time that Red Hat JBoss Operations Network 3.3 was released.

Chapter 1. Red Hat JBoss Operations Network Prerequisites and Basic Installation

The list of supported platforms for Red Hat JBoss Operations Network are listed at https://access.redhat.com/site/articles/112523.

1.1. Supported Configuration Information

1.1.1. Changes in Required Java Versions

JBoss ON supports two versions of Java:
  • Java 6 (deprecated)
  • Java 7
For Windows users, the 64-bit JDK and JRE are also now supported in addition to 32-bit JDK.
For a full list of supported configurations, see JBoss ON Supported Configurations page.

1.1.2. Changes in Supported Web Browsers

JBoss ON has added these browsers that are used for testing and are supported for version 3.3.
  • Firefox 17 ESR (new)
  • Internet Explorer 9 (new)

1.2. Installing or Upgrading JBoss ON

The Installation Guide provides full installation and upgrade instructions. See Installing the JBoss ON Server for comprehensive server installation instructions and Upgrading JBoss ON Servers and Storage Nodes for information relating to upgrading to the latest version of JBoss ON.

Chapter 2. New Features in Red Hat JBoss Operations Network 3.3

This version of JBoss ON introduces both new features and feature enhancements that improve JBoss ON's performance for managing resources.

2.1. Feature Enhancement: Filter Alerts by Name on the Dashboard

JBoss ON 3.3 users can now filter alerts on their dashboard by Name in addition to Time/Date and Priority Flag. This functionality is present in the Recent Alerts, Resource Alerts, and Group Alerts portlets. To use this new functionality, click the gear icon on the top of an alerts portlet. For additional details on using the alert portlets, consult the Viewing Alerts in the Dashboard section of the User Guide.

2.2. New Feature: Pre-defined DynaGroup Expressions for JBoss Enterprise Application Platform Server Groups and Domains

The JBoss ON 3.3 JBoss Enterprise Application Platform 6 Plug-in now ships with new DynaGroup Expressions that map to existing JBoss Enterprise Application Platform 6 Server Groups and Domains. The new DynaGroup Expressions also stay in sync as new domain controllers or server groups are added. These new DynaGroup Expressions are useful when managing configurations across an entire domain containing various host controllers and server groups. They are also useful when managing multiple JBoss Enterprise Application Platform 6 Domains.
Note
These DynaGroup Expressions are editable, but any changes made to them will be overwritten if the plug-in is reinstalled or updated.

2.3. New Feature: Recent Events Portlet

Users may now see Recent Events directly on the Dashboard using the Recent Events Portlet.

2.4. New Feature: Red Hat Access Integration in the JBoss ON UI

Red Hat Access is now integrated into the JBoss ON UI, allowing subscribers to open new support cases, view existing support cases, and search the Red Hat Customer Portal for exclusive knowledge and solutions. Subscribers can also use Red Hat Access to open cases against JBoss Enterprise Application Platform 6 and other products integrating with the JBoss Enterprise Application Platform 6 plug-in, as well as attach JDR reports (where available) with additional JBoss ON-specific information directly from the JBoss ON UI. For more details see the Red Hat Access section in the User Guide document.

2.5. New Feature: JBoss Enterprise Application Platform 6 Plug-in uses Host Controller Configuration for Naming Newly Discovered Resources

The JBoss Enterprise Application Platform 6 plug-in now supports using the host.xml when naming newly discovered resources. When a discovery scan finds new host controller resources, the resources will be named based on the "name" parameter in the host.xml. Resources which have already been imported are unaffected.

2.6. New Feature: JBoss ON now Supports JBoss Diagnostic Reporter

JBoss ON 3.3 Sever and JBoss ON JBoss Enterprise Application Platform 6 plug-in now supports and integrates with JBoss Diagnostic Reporter (JDR). This allows JBoss ON to collect and log details and diagnostic information on managed instances of JBoss Enterprise Application Platform 6. This enables better troubleshooting and allows subscribers to provide more information to Red Hat support when opening support cases.

2.7. New Feature: JBoss ON JBoss Enterprise Application Platform 6 Plug-in Now Supports Versioned Deployments

JBoss ON 3.3 and the JBoss ON JBoss Enterprise Application Platform 6 plug-in now support Versioned Deployments, or deployments which contain the version number in the artifact name. This enables JBoss ON to recognize and track versioned artifacts (e.g. myapp-1.0.war and myapp-1.1.war) as different versions of the same resource. JBoss ON 3.3 also supports Versioned Subdeployments (e.g. EJBs, SARs, WARs, datasources, etc within EARs) within artifacts. For more details, see Versioned Deployements and Subdeployments section in the User Guide.

2.8. New Feature: Support for Patching with JBoss Enterprise Application Platform 6.2 and Above

JBoss ON 3.3 added support for patching JBoss Enterprise Application Platform 6.2 and above from the JBoss ON UI. This allows users to apply individual as well as cumulative patches to JBoss Enterprise Application Platform 6.2 and above servers managed by JBoss ON. For more details please see the Patching JBoss Enterprise Application Platform 6.2 and Above section of the User Guide.

2.9. New Feature: JBoss ON 3.3 Now Supports Connecting to JBoss Enterprise Application Platform 6 Using SSL

JBoss ON 3.3 adds support for connecting to the JBoss Enterprise Application Platform 6 management interface using one-way and two-way SSL using the JBoss ON JBoss Enterprise Application Platform 6 plug-in. This provides additional security when using JBoss ON to monitor and manage both JBoss Enterprise Application Platform standalone servers and domains. For more details see the Changing the Secure Connection Settings for an JBoss Enterprise Application Platform 6 Server section of the User Guide.

2.10. New Feature: Bundle Content Handover in Ant Bundle Recipes

JBoss ON 3.3 now supports Bundle Content Handover in Ant Recipes using the rhq:handover tag, which allows the bundle target resource to participate in the bundle deployment. This enables functionality such as deployments in JBoss Enterprise Application Platform 6 domain mode. For more details see the rhq:handover section in the User Guide.

2.11. New Feature: Drift Configuration Monitoring in JBoss Enterprise Application Platform 6

JBoss ON 3.3 now supports drift configuration monitoring in JBoss Enterprise Application Platform 6 standalone and domain mode. This allows JBoss ON to monitor and track any changes made or file updates to the JBoss Enterprise Application Platform configuration directory including those made by JBoss ON as well the JBoss Enterprise Application Platform 6 CLI and the JBoss Enterprise Application Platform Administration Console. This also allows for JBoss ON to monitor and track deployments made to JBoss Enterprise Application Platform 6 such as applications and datasources. For more details, please see the Drift Configuration Monitoring on JBoss Enterprise Application Platform 6 Resources section of the User Guide.

2.12. New Feature: JBoss ON 3.3 Now Supports Remote Agent Install Over SSH

JBoss ON 3.3 now has the ability to perform an agent installation over SSH using the JBoss ON UI. This allows users to remotely install agents without having to login to the server directly. For more details, please see the Installing an Agent from the JBoss ON UI section of the Install Guide.

2.13. New Feature: JBoss ON 3.3 Now Supports Protection of Properties in Server and Agent Configuration Files

JBoss ON 3.3 has added support for encoding properties in the server and agent configuration files. This enables sensitive information to be hidden when viewing the server or agent configuration files directly. For more information, see the Protecting Sensitive Information in the Server Configuration and Protecting Sensitive Information in the Agent Configuration sections of the Administration and Configuration Guide.

Chapter 3. Structural Changes in JBoss ON 3.3

There are some structural changes between Red Hat JBoss Operations Network 3.3 and 3.2 which are called out because they affect the usage and experience of JBoss ON.

3.1. Deprecation of JDK 6

JBoss ON 3.3 marks the final release that supports Java Developers' Kit (JDK) 6. Future releases of JBoss ON will support JDK 7 or later versions.

3.2. Integration with JBoss Enterprise Application Platform 6.3

JBoss ON 3.3 has been built upon JBoss Enterprise Application Platform 6.3, which includes security, and bug fixes implemented in this version.

3.3. UI Improvements in JBoss Enterprise Application Platform 6 Plug-in

JBoss ON 3.3 now includes improvements in the visual representation of subsystems in the JBoss Enterprise Application Platform 6 plug-in. Subsystems are now organized in subcategories for easy reference and interaction.

3.4. Support for Oracle 12c

JBoss ON 3.3 now supports Oracle 12c for backend database services. Administrators can now configure JBoss ON servers using Oracle 10g (though support with JBoss ON is deprecated), 11i, and 12c. Previous releases are not supported.
Important
When configuring Oracle 12c as a backend for JBoss ON, note that the RESOURCE role no longer grants UNLIMITED TABLESPACE to the rhqadmin user by default. Granting this system privilege to a user must be done manually.
See the Configuring Oracle section of the Red Hat JBoss Operations Network Installation Guide for more guidance.

3.5. Updated Look and Feel

JBoss ON 3.3 has an updated look and feel throughout the entire user interface. This improves usability and enables JBoss ON to provide a more consistent user experience that aligns more closely with Red Hat's products.

3.6. JBoss ON JBoss Enterprise Application Platform 6 Plug-in Now Persists JVM Heap Arguments to JBoss Enterprise Application Platform Standalone Configuration Files

The JBoss ON JBoss Enterprise Application Platform 6 plug-in now persists JVM Heap Argument changes directly to the JBoss Enterprise Application Platform 6 startup shell scripts. Persisting directly to the startup shell scripts enables those changes to be saved in a central location which can be used consistently within and outside of JBoss ON. This is useful for maintaining consistent Java environments in scenarios where JBoss Enterprise Application Platform 6 restarts are initiated from within as well as outside of JBoss ON. For more details see "Changing JVM Heap Arguments in Standalone Mode" section of the User Guide Document.

3.7. Increased Metric Graph Limit for Groups

The graphing limit for maximum number of resources shown on a graph has increased from 10 to 50.

3.8. Agent Upgrade and Update Processes Now Preserves Custom agent-configuration.xml and log4j.xml Files in Place

In JBoss ON 3.3, the agent upgrade and update process now retain custom agent-configuration.xml and log4j.xml files in place. Previously, the update process would copy these files over as agent-configuration.xml.custom and log4j.xml.custom. This change allows users to restart the agent using the --cleanconfig option and retain their custom settings.

3.9. Ant Bundle Recipe Elements rhq:url-archive and rhq:archive Now Support the "destinationDir" Attribute

JBoss ON 3.3 has added support for the destinationDir attribute to the rhq:url-archive and rhq:archive Ant elements. Setting destinationDir allows users to specify where the archive will be stored. For more details, see the A Reference of JBoss ON Ant Recipe Elements" section of the User Guide document.

3.10. JBoss ON Agent Upgrade Now Preserves Custom Scripts

While performing an Agent upgrade, any custom scripts (.sh or .bat files) in the $RHQ-AGENT/bin directory will now be preserved. This allows users keep custom developed scripts in places across JBoss ON Agent upgrades without having to manually re-add them.

3.11. Setting Added for Enable Login Without Roles

JBoss ON 3.3 now has a global setting to control whether existing user accounts without a role assigned are allowed to login. This functionally is especially useful when integrating JBoss ON with LDAP directories. By default, user accounts without a role assigned are able to login.
This setting can be found by clicking on Administration in the top menu, going to the System Settings section in the Configuration table, and reviewing the Enable Login Without Roles setting in the General Configuration Properties section.

3.12. Manage Application Deployments to Server Groups from DomainDeployments Menu

JBoss ON 3.3 and the JBoss ON JBoss Enterprise Application Platform 6 plug-in have added the ability to deploy applications to server groups and adjust which groups those applications belong to directly from the the DomainDeployments menu for JBoss Enterprise Application Platform 6 domains. This provides users with a single, unified location to view and update where an application is deployed.

3.13. JBoss ON 3.3 Now Supports Configurable Snapshots and Retention Strategies for Storage Node Snapshots

JBoss ON now supports the ability to schedule snapshots at desired intervals for storage nodes as well as configure the snapshot retention strategy right from the JBoss ON UI . For more information on how to schedule snapshots and configure retention strategies, please see the Storage Node Snapshots section of the Administration and Configuration Guide.

3.14. Documentation for JBoss ON 3.3 Beta

For the beta release of JBoss ON, there is a limited set of updated documentation available:
  • These release notes
  • An updated Installation Guide with new JBoss ON server installation and upgrade procedures and details on the rhqctl script
  • Updated monitoring documentation with new information on adding and managing metrics storage nodes

3.15. Support Levels and Bug Reporting for JBoss ON 3.3 Beta

JBoss ON 3.3 Beta is available for evaluating new features and for testing performance.
This release is intended as a preview for new functionality in JBoss ON, including structural changes such as the new metrics storage node to store collected metrics.
This is not a final release of JBoss ON, and it is not subject to the same level of support as a general availability (GA) release of JBoss ON.
Note
While there is no service-level agreement for responding to JBoss ON 3.3 Beta issues, support will make a commercially reasonable effort to respond to reported issues within two business days.
Defects and other issues can still be reported with JBoss ON 3.3 Beta, and we encourage you to provide us with any feedback you can offer.
  • Issues are logged in the Customer Portal in the Create a New Case link under the Support tab.
    Select the Red Hat JBoss Operations Network product from the product details area, and then select the 3.3 Beta version.
  • Any issue reported for JBoss ON 3.3 Beta must be given a severity value of 4.
  • You must already have a current, active subscription for JBoss ON or a managed JBoss product to receive support for JBoss ON 3.3 Beta.
  • Issues identified in JBoss ON 3.3 Beta are not guaranteed to be resolved in JBoss ON 3.3 GA.
When installing JBoss ON 3.3 Beta for evaluation, it is critical to remember that there is no upgrade path from JBoss ON 3.3 Beta to JBoss ON 3.3 GA. Any JBoss ON 3.3 Beta servers must be removed, and new JBoss ON 3.3 GA servers installed to use the final version of the product.

Chapter 4. Other Changes in this Release

4.1. Bug Fixes

Plugin -- Other

BZ#1127875 - Process service resources continues to monitor old process even after process no longer exists or after piql/pid file has changed

A bug in the JON "Platform Plug-in" resulted in connection setting changes to the "Process" resource to be ignored. When the user changed the Connection Settings of a Process service (for example, changed the PIQL query or the PID file path), the plug-in kept monitoring the same process as it was before the change, until this process terminated or the agent was restarted. The ProcessComponent class is now fixed to take Connection Settings into account immediately after changes. The plug-in now immediately monitors the process denoted by the Connection Settings after they were changed.

BZ#1127876 - Platform process service CPU Percentage metric returns values that are inconsistent with actual CPU load measurements

When the process being monitored was no longer present, or the piql or pid file properties or values were changed, the native library the Platform Management plug-in used to retrieve CPU usage information returned unpredictable values. Improvements to the Platform Management plug-in better handle instances where a process is removed. ProcessComponent no longer adds data to the measurement report if the ProcessInfo instance shows that the process is down.

Documentation

BZ#1073483 - Incorrect URL in the Configuring JBoss ON Servers, Agents, and Storage Nodes document

An issue with incorrect URLs in the "Managing Databases Associated with JBoss ON" section of the Installation Guide did not mirror the new URL structure for the Admin Tools page. Users could not access the URLs specified in the docs, which resulted in unnecessary confusion for the user. All instances of the URL in this user guide are now changed to the correct URL pattern, which fixes the originally reported issue.

BZ#1118224 - Server can be shut down using CLI script notification invoked by fired alert

The server allowed a CLI script notification to be executed as a response to a fired alert. Users who had permission to execute scripts on the server could shut down the server using this method. To prevent this from happening, older versions of the product had the Java security manager turned on. Unfortunately this created a larger performance impact than originally estimated. The security manager is now turned off by default for this reason. To turn the security manager back on again, remove the 3 'X' in -DXXXjava.security.manager in the file bin/internal/rhq-server.{sh,bat} and restart the server.

BZ#1164362 - Installation guide section Installing the Server on Linux incorrectly refers to rhqctl.sh instead of rhqctl

An issue with how the rhqctl script was documented in the Installation Guide caused issues with executing the script correctly on Linux. A fix to the "Basic Setup: Installing the Server on Linux" section now removes the trailing .sh at the end of the command, which fixes the originally reported issue.

BZ#1138686 - JBoss ON Installation Guide should be updated to include preferred directory structure when installing JBoss ON components

The JBoss ON Agent must be installed in a writable directory, however this was not described in the product docs. A new section in the Installation Guide describes the preferred directory structure, and is available from access.redhat.com/documentation/en-US/Red_Hat_JBoss_Operations_Network/3.3/html-single/Installation_Guide/index.html#before-agent-install

BZ#1135623 - Update instructions fail to mention how old metric data can be purged after migration/upgrade

The installation and upgrade instructions for upgrading from JBoss ON <3.2 to JBoss ON >=3.2 did not provide information on how to remove the legacy metric data from the relational database. If users did not perform this step, legacy data persisted in the system, and continued to use many GB of database space. In some instances, database performance was impacted. Multiple changes were made to the Installation Guide to fix this issue. The "About the Upgrade Script" section now has two extra database management parameters described in the "Options for Upgrading" subsection. The "Upgrading the JON Server and Components" procedure has a specific call-out in Step 8 that prompts users to check the "Options for Upgrading" section so they are aware of the new parameters. This content is available for review at access.redhat.com/documentation/en-US/Red_Hat_JBoss_Operations_Network/3.3/html-single/Installation_Guide/index.html#upgrade-script

BZ#1118072 - Storage restore procedure is necessary for all rhq tables

It was discovered that missing tables in the storage restore procedure were preventing all relevant data from being correctly restored. Improvements to the procedure in https://access.redhat.com/documentation/en-US/Red_Hat_JBoss_Operations_Network/3.3/html/Admin_and_Config/storage-restore.html now include all affected tables, and fixes the originally-reported issue.

BZ#1128770 - Missing option in 7.3.2. Agent Start Options (Configuring JON Servers and Agents)

An option was missing from the list of Agent Start Options in the Configuring JON Servers and Agents documentation. Users were unable to view the information associated with the --purgeplugins command. The new documentation contains a reference to the --purgeplugins command. Users can view the command and its description.

BZ#1095805 - Changing Heapsize on JBoss ON Server doesn't work per the doc steps

An incorrect environment variable in the 'Tuning the Server JVM' section of the Configuring JBoss ON Servers, Agents, and Storage Nodes guide caused issues with changing the heapsize on the JBoss ON server. The incorrect parameter RHQ_CONTROL_JAVA_OPTS is now corrected with RHQ_SERVER_JAVA_OPTS, which now allows the heapsize to be changed as expected.

Plugin -- JBoss EAP 6

BZ#1015334 - EAP 6 host controller domain host update results in managed servers no longer being "manageable"

It was discovered that if the domainHost attribute of an EAP 6 host controller changed, the managed servers connected through the host controller became unavailable. This prevented any new managed servers from being discovered. The host controller was still partially functioning in inventory, however its managed servers were unreachable due to the domain host property value being read-only with no method to update or correct the issue. A fix to the AS7 plug-in now attempts to read domainHost from host.xml when needed. The 'domainHost' read-only plug-in property is deprecated, and a new trait is introduced to better handle the scenario.

BZ#1004977 - Datasource configuration changes SHOULD NOT require a complete reload/restart of EAP

The agent plug-in component supporting Datasource and XA Datasource resource types, like most JON plug-in components, used to re-apply every configuration property when a user changed resource configuration through the GUI or the CLI. After saving a datasource configuration change, the managed EAP 6 standalone or domain servers would put themselves in a "reload-required" state, even if the user only changed a property which does not require it. The plug-in component supporting Datasource and XA Datasource resource types now sends changed configuration properties only to the EAP 6 management interface. Additionally, the datasource "Disable" operation now has a "Allow service restart" parameter, which immediately stops the datasource service before changing a property which requires the datasource to be in a "Disabled" state. JON users can now manage datasources and XA-datasources following the workflows they are used to with the EAP6 CLI or Admin Console.

BZ#1150667 - Availability of host controller is down due to ArrayIndexOutOfBoundsException

A bug in the Application Platform Plug-in Pack caused issues with properly discovering the availability of the slave host controller. The slave host controller was discovered, however it’s availability check failed with an ArrayIndexOutOfBounds Exception. The fix implements improvements to BaseServerComponent#findASHostName(), which now uses the EAP 6 management API to fetch the local-host-name attribute from the host controller. The host or domain controller state is recorded correctly, which fixes the originally reported bug.

BZ#1080833 - Enabling/Disabling EJB3 statistics cause redeployment of Application in EAP 6

Enabling or Disabling EJB3 statistics in the EAP 6 JON Plug-in caused the configuration changes to be saved correctly, however the web application in EAP 6 was redeployed. A fix to the EJB3 subsystem now only applies configuration properties that differ from the current configuration. The read-write config update delegate reads resource config prior to applying the update, and applies changes only. This feature currently supports simple probe only. EJBs are no longer completely re-deployed when a EJB3 monitoring setting is changed.

BZ#998673 - Thread Pool keepalive-time-unit needs to define allowed values so the user doesn't have to guess

A deficiency in the JON UI concerning the allowed values of the keepalive-time unit caused confusion for customers when trying to configure the keepalive-time correctly. The keepalive-time unit field now replaces the previous free-text field with predefined, supported units of measure drop down. Customers can now confidently set the keepalive-time precision easily, without trying to convert a number to the correct unit of measure.

BZ#1034218 - EAR subdeployments are not discovered for DomainDeployment

It was discovered that when an EAR was deployed to a managed server group, it was missing its children (subdeployments). If same EAR was deployed to a standalone server, the subdeployments were present. EAR subdeployments (and their runtime child subsystems) could not be correctly discovered. The EAP management plug-in now handles subdeployments, such as those found within an EAR, which are deployed to either standalone or managed server deployments.

BZ#1072543 - Unable to collect metrics from EAP 6 datasources when a system properties is used in configuration

Collecting metrics from an EAP 6 datasource failed due to improper handling of returned data types from the DMR, specifically the minimum and maximum pool size settings of a datasource configuration. If a user declared a system property as the value of min-pool-size or max-pool-size, the DMR could not parse the properties. The fix now handles system properties in the datasource configuration file, which fixes the originally reported issue.

BZ#1077943 - [AS7] plugin's connection cleaner leaks classloader

The ASConnection of the EAP plug-in has a static thread pool which runs connection "cleaners" to prevent connection leakage.

When the plug-in container was restarted, the classes of the plug-in were loaded again. However the static thread pool was never shut down which left a thread running with a context classloader from the previous plug-in container.

The stale thread lead to an eventual out-of-memory exception because of depleted permgen memory needed to hold all the classes from all the plug-in container runs in the agent’s lifetime.

The AS7 plug-in now has a plug-in lifecycle listener that shuts down the aforementioned thread pool preventing the leak from happening.

BZ#1091235 - max-connection configuration property set to wrong definition, stating AJP instead of APR

An incorrect definition for max-connection in the JBoss EAP 6 Plug-in documentation made it difficult for users to correctly set the maximum connections for APR because the docs suggested it was for AJP connections. The javadoc source is now fixed and contains the correct text in the parameter description.

BZ#1093370 - JON plugin for EAP 6 cannot create an admin user in a moved configuration directory

A bug in HostConfiguration.java of the EAP 6 Plug-in prevented the Add User operation from succeeding in cases where EAP was started with the "configuration" directory set to a non-default location. The Add User operation was only ever intended for development using a default JBoss EAP standalone or domain installation. That being said, moving the directory should not have broken the functionality. A fix to the plug-in now correctly resolves the path to the "configuration" directory, which allows the Add User operation to succeed.

BZ#1119240 - Adding http-interface in the host controller's host-slave.xml is not recognised by JBoss ON Agent causing Host Controller's availability to stay DOWN

If the management interface in the host-slave.xml file was changed, the JBoss ON Agent threw an "Invalid port: 0" error while trying to discover the JON server. Restarting the JON server and agent did not clear the error. JBoss ON discovered the Host Controller but it was marked as DOWN and a WARN log event was added to the agent.log file. The Plug-in UI did not adequately show an indication of the issue. The fix now displays a warning icon next to the availability icon on the resource. When clicked, a window displays with the following explanation: "Unable to detect management port. Please enable management HTTP interface on <server> and then set the correct port number in Connection Settings of this resource". The same warning is logged in the agent.log file as a WARN message when the host controller is started.

BZ#1144146 - Messaging queue and subscriber runtime is not exposed

The agent plug-in component did not support monitoring runtime queues and topic subscribers on the HornetQ resource. New resource types for runtime-queue are introduced, supporting all operations and monitoring all metrics exposed by EAP. Because runtime queues can frequently appear and disappear, the Missing Resource Policy now makes it possible to auto-uninventory non-existant runtime-queues as required. These fixes now allow users to monitor and manage topic subscribers and runtime queues more efficiently.

BZ#1042797 - It is not possible to change "Inet Address" for management network interface from JBoss ON UI

A bug in the JON UI that prevented IPv4 and IPv6 inet address values from being specified as unset failed to persist the configuration change and failed to marshall configuration. A fix to the JON UI now allows changes to the inet address and fixes the originally-reported issue.

BZ#1095000 - EAP 6 WebConnectorComponent computes an incorrect default max-connections property value

The JBossWeb Connector resource component attempted to determine the value for "Max Connections" if the underlying JBoss EAP server indicated the value was undefined. This caused an incorrect value to be set for the property if the underlying connector implementation was not the default JIO connector type. The JBossWeb Connector resource no longer sets a calculated "Max Connections" value, and defers to JBoss EAP and its connector subsystem to determine the default based on the connector implementation. When "Max Connections" is "unset", it no longer overrides the connector’s default implementation for maximum connections.

BZ#955816 - Bundle deployment to EAP 6 standalone server fails due to deployDir trait not being available

When a user attempted to deploy content to EAP 6 standalone using a provisioning bundle, deployment failed if the agent was restarted since the last time the deployDir trait was retrieved from the managed resource, and the managed resource had since been shut down or became unavailable for any other reason. This resulted in bundle deployment being inconsistent. Manual intervention was required to verify the deployment was actually successful. The most recent value of the trait is now retrieved from the server instead of cached trait values. Deployment no longer fails with bundle deployments.

BZ#1128141 - HostController gets DOWN when name attribute in host.xml is empty

The AS7 plugin did not recognise the HostController if the name attribute in the host.xml file was left blank. The HostController would not deploy if there was not a name assigned to it in the host.xml file. The HostController name is now provided by the API instead of needing to be specified in the host.xml file. The HostController name is automatically detected.

BZ#1021018 - EAP6 plugin ignores rhq.agent.plugins.availability-scan.timeout setting and defaults to 10 seconds

It was discovered that the EAP 6 plug-in’s method execution timeout ignored the rhq.agent.plugins.availability-scan.timeout setting. It persisted the EAP 6 scan timeout setting, which made the behavior of the scan timeout unpredictable. The fix now uses the container’s availability timeout for sync availability checks and improves the usage plug-in configuration settings for asynchronous availability of server components.

Content

BZ#1020874 - Add subscriptions button on platform->Content->Subscriptions tab displays HTTP Status 404 in UI

Issues with the Subscriptions form code caused attempts to add a subscription to the JBoss Patches repository in the Platform > Content > Subscriptions view to fail with a HTTP Status 404 error. Improvements to the form code are now implemented that fix the originally reported issue.

BZ#952665 - Create new 'Local Disk Storage Content Source' displays IllegalStateException in UI

A bug in the JON UI caused and IllegalStateException when a user navigated to Administration > Content > Content Sources > Create New, and then clicked on the 'Local Disk Storage Content Source' link. A fix to the 'Local Disk Storage Content Source' link behavior now allows users to click the link without causing an IllegalStateException.

CLI

BZ#1038317 - Use of remote client API results in error/warning messages regarding indirectly referenced JPA annotations not being resolvable

The JON CLI did not ship with the Hibernate Core and Hibernate JPA jars because they were not used when running the JON CLI. When JON remote Java client developers used the content of the CLI "lib" directory as their list of dependencies they would see errors/warnings such as "The type javax.persistence.GenerationType cannot be resolved. It is indirectly referenced from required .class files". The JON CLI now ships with the Hibernate Core and Hibernate JPA JARs. JON remote Java clients now compile without the previously reported warnings and errors.

BZ#1135473 - Implicit Variables for CLI are not available if the script is run as notification on an alert

When used as alert notifications, CLI scripts could not find Remote API entities unless they were declared with their fully qualified name. This was due to the script engine being initialized with an empty list of entity packages. In a standalone CLI environment, the list of packages was determined by scanning the "lib" directory of the CLI installation, which was not possible with an embedded CLI. As a consequence, the alert notification script execution failed with an error message similar to "javax.script.ScriptException:org.mozilla.javascript.EcmaError: ReferenceError: "ResourceCriteria" is not defined". Modifications to the script engine now initialize the embedded CLI with a list of packages determined by scanning the classpath at runtime. Remote API entities can now be declared with short names in CLI alert notification scripts.

BZ#1131494 - CLI command to remove storage node gives transaction error

The rhq-cli command was supposed to remove a storage node with an operation mode of INSTALLED but failed with a transaction error. The storage node was not removed. Transaction management has been rectified to accommodate storage node changes. The rhq-cli command executes without error and removes the INSTALLED storage node.

BZ#1001383 - getResourceConfiguration does not return the current resource configuration

The Resource configuration returned by a call to getResourceConfiguration did not reflect the existing configuration on the resource. Instead, it reflected the latest configuration returned by a configuration-discovery scan. The client using the remote API or a CLI script was unable to retrieve the resource configuration for a given resource using an intuitive method. Enhancements to the ConfigurationManager.getLatestResourceConfigurationUpdate Remote API Javadoc now explain the retrieval features more intuitively.

CLI,Documentation

BZ#1143918 - Addition of new Log Event Source using JBoss ON CLI will break the UI if all properties (even unset ones) are not added

It was discovered that a "Globally uncaught exception" was generated in the GUI when a new Log Event Source was added using JBoss ON CLI with certain properties (even unset ones) not defined. This issue occurred even when the script added a seemingly correct Log Event Source. Any attempt to make a change (for example, disabling the Event Source) would generate the error. The fix implements improvements to the GUI ConfigurationEditor, which ensures unset fields' default values are added as required.

Installer

BZ#1051299 - Installer fails with NullPointerException when rhq.storage.nodes is undefined

When a user attempted to install a JBoss ON server with rhq.storage.nodes commented out or removed from rhq-server.properties, the installation failed if the installer was not performing a storage node installation as well. Property checking is now improved to produce a more useful error output. The message is identical to the message displayed when the cluster is not available, and points the user to several possible root causes to assist with their investigations.

BZ#1089757 - Install fails if shutdown process takes more then 10 seconds

During Installer execution, a shutdown is performed on the newly installed component(s).

The shutdown delay was hard-coded to ten seconds. While ten seconds was adequate for development environments with no load and the database running locally, a production environment may take anywhere from a few seconds to upwards of a minute to shut down. If the target component took longer than 10 seconds to shutdown, the entire install was rolled back and aborted.

The fix implements a number of improvements to the Installer shutdown process. The timeout has been increased to 30 seconds, from 10 seconds. Additionally, a system property can be set to customize the timeout value through RHQ_CONTROL_ADDITIONAL_JAVA_OPTS using the -D java option. The sysprop name is "rhqctl.wait-for-process-to-stop-timeout-secs". Additional logging is also available to assist with tracking down issues in this area.

BZ#1096923 - UNDO process on rhqctl install failure leads to missing rhq.installed marker file on re-install

If "rhqctl install" failed, re-executing the command after repairing the failure did not trigger the creation of the rhq.installed marker file. This prevented rhqctl from controlling the newly installed server. The fix enhances the vault detection code to avoid a timing issue on install for the first read, and reverts the standalone-full.xml file if any errors occur during install allowing the reinstall attempt to succeed.

Inventory

BZ#1056562 - Inventory between JON Agents and Server get unsync after the server backend database backup

An issue with synchronization between JON Agents and Servers when a database remained DOWN for longer than normal caused all JBoss ON Agents connected to the database to incorrectly show a DOWN status, when the agents were still running correctly. The only way the issue could be fixed was to restart all agents so their availability status was refreshed. A fix to the JON UI now shows JON Agents as UP after communication is restored to the database. This fixes the originally-reported issue.

Installer,Storage Node

BZ#1080158 - Do not use empty string as values for properties in rhq-storage.properties

If the rhq.storage.hostname or rhq.storage.seeds properties were uncommented and left unset in rhq-storage.properties, the storage installer created an invalid configuration. The listen_address and seeds properties in cassandra.yaml were left blank, which caused the storage node startup to fail. The fix ignores empty values in rhq-storage.properties which ensures that invalid configuration is not introduced into cassandra.yaml, and fixes the originally-reported issue.

BZ#1092707 - If re-install is attempted using a different storage node host name, install fails due to JON installer still trying to use old host name

If a user performed a JBoss ON installation but then realized the default host name used for the storage node would prevent it from being accessed from other machines in the network, there was no way to fix this without dumping the relational database and starting over. The fix now allows a re-install/re-upgrade to replace existing storage node definitions as long as the storage nodes have not been used. If none of the storage nodes are yet linked to a resource, they are eligible for replacement.

UI

BZ#1070981 - Auto fit columns used in platform utilization report results in unreadable column data

A bug with how the "Auto Fit All Columns" or "Auto Fit" view options on graphic columns rendered in the UI caused issues with displaying all data. Users who attempted to access the Reports > Platform Utilization report and selected Auto Fit observed that the CPU, Memory, and Swap columns became unreadable. The fix to the column view behavior now ensures the Auto Fit options resize the columns to fit in the value, the percent sign and the graph.

BZ#1084586 - Horizontal scrollbars are missing from Dashboard

An issue with the Resource and ResourceGroup metric D3 portlets caused chart data that exceeded the boundaries of the portlet to be cut off. There was no way to resize the view to see all the data on the dashboard. The portlet had to be resized manually so all data could be viewed. A fix implemented for the Resource and ResourceGroup metric D3 portlets now sets a fixed width for the graph container div, and provides horizontal scroll-bars when required.

BZ#1134445 - Globally uncaught exception when trying to view graph for the time period when there is no data collected

When trying to generate the metric graph for a time period when there was no collected data, a Globally Uncaught Exception was shown in the JBoss ON UI and "lowestValue is null" message was shown in the Message Center. The fix now assigns the measurement definition UOM when no valid values are available, which fixes the reported issue.

BZ#1083894 - Storage node's cluster status "NORMAL" written in RED cannot be changed

If a job ran on a storage node that timed out but finished properly, the cluster status of the node was displayed as NORMAL but written using red font which was atypical of this UX pattern. While the red text correctly indicated an underlying issue with the job, there was no way to clear the warning. The fix adds functionality to the storage node details page, which provide more information about an issue and allow the user to re-run the operation if circumstances require this action or acknowledge the error in order to remove the red font color.

BZ#994267 - LDAP login causes Invalid path was requested error to appear in server log

When logging in as an LDAP user the server appended a message to the log that suggested an invalid path was requested. The user was correctly presented the user authentication page and could subsequently log on correctly, however the erroneous log message may have confused customers who discovered this in the logs. The fix ignores the log messages coming from the JSF layer. The incorrect error message no longer displays.

BZ#1146266 - Metric graph is empty if high, low, and avg values are all the same for every time bucket within the selected time range

The JBoss ON metric graph did not display as expected if there were missing values, or if all values were the same within a selected time range that included 60 or more collected values. The minimum and maximum values allowed have been modified. Metric graphs now display the expected data.

BZ#1085956 - Topology Agents screen flickers on refresh

A bug with viewing the list of more than 50 active agents in the Topology Agents screen (Administration > Topology > Agents) caused the list to flicker while the remaining records were being loaded into the view. While this flickering state was occurring, the total agent count flicked between two row count values. The screen did not show all agents when it was initially loaded, which caused incorrect information to be displayed to users. A fix to the Topology Agents screen fixes the load issue, and displays all rows when initially loaded.

BZ#1127871 - Suspect metrics report, column "Metric" sort doesn't sort until you click several times

An issue with the sort order behaviour of the Suspect Metrics Report’s "Metric" column caused the sorting icon to indicate a sort order change but the UI would not display the records correctly. For large record results (over 1000), if the user scrolled to the end of list and then scrolled to the start of the list, the sort order began to work correctly until the user left the page. The fix changes the report behaviour to allow server-side sorting on the Resource, Metric, Date/Time and Outlier fields which improves performance. The remaining fields are not sortable server-side and have client-side sorting only allowed (they are sortable after all the data has been loaded to the client).

Storage Node

BZ#1075575 - Remove OpenJDK warning during startup

An incorrect WARN level log entry produced by CassandraDaemon.java suggested that OpenJDK was not a supported JVM when used in a Red Hat Supported Configuration. The fix removes the erroneous warning from the C* driver, and directs the C* server to use the correct C* driver version compatible with the release. The incorrect WARN level no longer displays in the logs.

BZ#1082805 - rhqctl upgrade is unable to perform data migration if estimate is executed first

If a user executed rhqctl upgrade and passed the --run-data-migrator estimate parameter, the data could no longer be upgraded using the expected tool. Attempting to execute the --run-data-migrator do-it command fails indicating that the installation/upgrade is already finished. The fix removes all options for the --run-data-migration argument. Users can no longer use none, estimate, or do-it following the argument. If the argument is used when invoking the upgrade comment, rhqctl will migrate the data right away as part of the upgrade process.

BZ#1084056 - Storage node has internal server metrics "Anti Entropy Sessions" marked as unavailable

It was discovered that the "Anti Entropy Sessions" Internal Server Metric statistic was being monitored when it should have logically been excluded. The Anti Entropy Sessions metric was being discovered only after a Repair operation was initiated, and was marked as DOWN when the JON server was restarted until another Repair operation was run. The fix adds a missing policy to the "Anty Entropy Sessions" type (a single bean), which now allows users to configure the behaviour of a missing resource after a Cassandra or Storage Node server restart. This added policy allows users to change the default (DOWN) convention of MISSING to something more suited to their use case. This bean is particularly useful for long running repair jobs, because it provides important telemetry for the progress of the repair job. Even if the bean disappears after a C* restart, it will be visible again to JON as soon as the repair job is invoked. Users can now collect Anti Entropy metrics correctly between server restarts.

BZ#1127868 - Storage node snapshots are not generated when read repair storage cluster job times out

An issue with the read repair storage cluster job caused storage node snapshots to time out and the job associated with the job to be cancelled. The fix introduces improvements to code that now prevents a locking issue, which corrects the issue.

Launch Scripts

BZ#1088032 - Linux/Unix shell scripts attempt to use readlink even if it isn't installed/exist causing "readlink: not found" to be displayed

When executing launcher scripts such as rhqctl, the readlink command is used to resolve a scripts' real location in case it is a symbolic link. Users running operating systems where readlink was not installed by default (such as Solaris), or not present on the path, experienced a message similar to: "./rhqctl: readlink: not found". The fix to rhqctl now shows a warning message if readlink does not exist, which allows users to install the appropriate packages and continue with the operation.

BZ#1112240 - rhqctl not able to find new defined RHQ_AGENT_PIDFILE_DIR in rhq-agent-env.sh

An issue in the rhqctl tool with pidFile detection caused the RHQ_AGENT_PIDFILE_DIR and RHQ_SERVER_PIDFILE_DIR environment variables to be ignored. The tool was unable to find the pidfile directory specified by the environment variable, and therefore could not stop the server and agent.

The fix removes pidfile detection from rhqctl and uses rhq-agent-wrapper.sh / rhq-server.sh to shut down servers and agents if necessary.

Database

BZ#1093265 - Database deadlock errors when attempting to apply alert definitions updates

If a user made a change to an alert template, saved it, realized that something was missed, made a second change to the same alert template and then resaved it, a database deadlock error could occur if the updates followed in quick succession. The fix adds a longer button unavailability timeout to prevent concurrent updates.

BZ#1079303 - Excessive database size when using bundles causing UI to fail

The bundle view was retrieving all audit messages for all versions of the bundle when the bundle details were requested. Additionally the relationship between the data tables that contained the bundle deployment information and the bundle’s audit messages were missing an index, which resulted in the bundle view page taking a very long time to display. In some cases, it would not even load. If the user attempted to delete the old bundle information, the deletion would very often fail due to a transaction timeout. A relationship index is now added to the bundle deployment and audit message data tables. Audit messages are only loaded for the bundle version that is currently being accessed. Additionally, when a bundle version delete operation is performed, its corresponding audit messages are deleted asynchronously. The bundle view now loads much faster as it is not retrieving unnecessary data, and the delete operation no longer requires the bundle’s audit messages to be deleted at the same time as the bundle version.

Provisioning,UI

BZ#1107338 - Bundle Files page does not display all required files due to a missing page control

A missing page control on the Bundle Files page prevented all files in a bundle from being displayed correctly in the UI. The fix adds the page control to the Bundle Files page, which corrects the originally-reported issue.

Core Server

BZ#1108833 - Enabling LDAP on one JBoss ON Server in HA Configuration is not propagated to another server(s) until their restart so the users cannot log in

When LDAP was enabled on one JBoss ON Server in HA Configuration, the change was not propagated to other servers in the group until the servers were restarted. This prevented users from logging onto the other servers in the group. The fix reinstalls the JAAS login modules when the HA nodes detect a change in system settings. The check is performed every 60 seconds. If LDAP configuration is enabled or disabled in a HA group, the other servers are now aware of the change in 60 seconds.

BZ#1113591 - Blank space around value of rhq.server.email.smtp-host property in rhq-server.properties is not removed and causes java.net.UnknownHostException when starting JBoss ON

Whitespace present in the rhq.erver.email.smtp-host property value in rhq-server.properties caused a java.net.UnknownHostException when JBoss ON was started. The fix adds a check to the rhqctl start and install commands that print warnings in case the tool finds lines with trailing spaces in rhq-server.properties file.

BZ#1115813 - [Enhancement] An email sent from JBoss ON using custom java mail server is not delivered because setFrom() is not set in JON MimeMessage

An issue with the JBoss ON server not including setFrom() in the MimeMessage from messages sent through the custom Java mail server caused emails not to be delivered. The only way to workaround the issue was to relax the SMTP policy in the custom mail server in cases when email comes from the JBoss ON server. The fix now sets the From address correctly using setFrom() before the message is sent through the custom Java mail server. Emails can now be correctly sent.

BZ#1119927 - When using JVM 1.7.0_40 or later, server logs SAXParserImpl parsing errors indicating property is not recognized.

The Xerces library included in the JON enterprise container did not support the accessExternalDTD or entityExpansionLimit properties, which caused ERROR messages to appear on server start and in server.log. The Xerces library has been removed from the product and replaced with a version where the bug was patched upstream, fixing the originally-reported issue.

BZ#1108912 - Server log warnings from CriteriaQueryRunner: could not initialize recoveryAlertDefinition when viewing recent alerts portlet

When the Recent Alerts dashboard portlet loaded the alert history, it also attempted to load the alerts corresponding recovery alert. If the alert did not have a recovery alert, a "could not initialize recoveryAlertDefinition" WARN level log entry was appended to server.log. This message was incorrectly classified as a WARN log message, and customers were unsure what action needed to be taken to correct the issue. The fix to the Recent Alerts dashboard portlet catches and ignores any EntityNotFoundException instances, which fixes the originally-reported issue.

Upgrade

BZ#1080508 - Server fails to start if rhq.server.log-level is not explicitly defined in rhq-server.properties

The JON Server failed to start when the "rhqctl install" command was run without the rhq.server.log-level property specified in the rhq-server.properties file. Customers who recently upgraded from a JON version not containing this parameter received ERROR log messages: "Cannot resolve expression 'expression "${rhq.server.log-level}". A fix to the rhqctl install parameter logic ensures servers now start when this parameter is not present. The ERROR log message has also been downgraded to an INFO log message, to realign the log message severity accordingly. Customers no longer experience the originally-reported issue.

Content,Performance

BZ#1073691 - Table deadlock issue when processing content report due to missing index and very large transactions in ContentServerServiceImpl.mergeDiscoveredPackages

Legacy issues with ContentServerServiceImpl.mergeDiscoveredPackages caused Content reports to occasionally fail to complete due to a concurrency/deadlock issue. The fix introduces various enhancements to ensure that the process is thread safe, and fixes overall performance issues.

Monitoring - Alerts

BZ#1119331 - Job Misfire handler fails: AlertAvailabilityDurationJob' job:Job class must implement the Job interface

A hotfix change to how availability duration jobs were scheduled (from using the Quartz job service to an EJB timer) caused jobs scheduled before the hotfix was applied—​but not evaluated or expired when the server was shut down—​to produce Job Misfire Handler errors. The fix now takes into account availability duration jobs left behind and changes the job to meet the new scheduling mechanism.

Monitoring -- Other,UI

BZ#1126495 - Graphs fail to render with 'java.lang.IllegalArgumentException: lowValue' in log

An issue with late measurement reports caused no graph to be displayed and metrics were unable to be displayed. A lowValue IllegalArgumentException was also generated. A fix to how invalid metrics are handled is now implemented, which fixes the originally-reported issue.

Agent

BZ#1128210 - Two platforms created for a single agent; one Java one Linux

When an agent’s native libraries were disabled or re-enabled. One of the servers was created with two associated platform instances. It was not possible to monitor both resources after importing the second discovered platform. The fix adds logging to NativeSystemInfo to detect why the Java platform was selected.

BZ#1124619 - Agent Update prompt-command is executed 5 minutes after RHQ Agent operation is invoked

Using the Agent prompt --update command resulted in a delay. The update action did not occur until several minutes after the --update command was executed. The --update command has been replaced with an Update Agent operation. The update operation is executed immediately on the remote agent.

BZ#1127873 - The agent's config export command produces an invalid XML file.

The JON agent’s 'config export' command generated an invalid XML file due to an introduced regression in upstream code that erroneously stripped the DOCTYPE and DTD declaration from the file. The absence of a DOCTYPE and DTD results in an invalid XML file and the file subsequently failed validation by the Java Preference system when parsed. The fix corrects the introduced regression, and resolves the issue of exporting valid configuration files.

Core Server,Performance

BZ#1133112 - Inventory Merge is slow and consumes a lot of memory when a parent has many children

If a monitored server had thousands of EJBs deployed, the server created a situation where it had a huge child cardinality for a single parent. Inventory merge was slow and consumed too much memory under these conditions. The fix ensures child sets are not pulled unnecessarily during inventory merge. The inventory merge process is now more efficient for large EJB deployments.

Core Server,Documentation,Monitoring - Alerts,UI

BZ#1146096 - Alert definition/notifications show with unknown configuration

It was discovered that when an alert notification definition becomes invalid, the GUI displays all notification definitions as unknown. The fix implements improvements to the server-side alert notification code. Error messages now display information that describes the issue for the invalid alert notification. Other notification definitions remain unaffected and their alert notifications display normally.

Plugin -- JBoss EAP 5

BZ#790101 - as5 plugin: "Client JAR [file:/foo.jar] does not exist or is not readable (note, this JAR may not be required for some app server versions)." warnings should not be logged for AS/EAP versions that do not require the corresponding jar

If a version of JON did not require a particular JAR, a "Client JAR does not exist or is not readable (note, this JAR may not be required for some app server versions)" warning would be logged whenever a server was discovered, or rediscovered. This caused unnecessary information in the logs, and caused customers to be concerned about the errors due to the severity of the log level. The fix changes the WARN-level message to DEBUG level. While it is not possible to distinguish between non-existent and non-readable JARs, activating the DEBUG log level will identify the originally-reported behavior, and the issue can then be investigated further.

BZ#885848 - Changing WAR/EAR deployment from exploded to unexploded (or vice versa) results in resource always being DOWN/Unavailable

An issue with the Deployment Name being read-only after initial discovery meant that changing a WAR or EAR deployment from exploded to unexploded (or vice versa) after it was initially deployed resulted in the resource always being DOWN or Unavailable. It was not possible to monitor or manage the resource. The fix removes connection setting’s "deploymentName" attribute and adds the "deploymentKey attribute. The attribute is only loaded when the component is started, or when it goes down because of NoSuchDeployment. Improvements to the Refresh button on the Monitoring/Calltime page are also implemented as part of this fix.

BZ#1077412 - Restart operation on EAP 5 resource doesn't return failure message when restart fails

The restart operation for the AS5 plug-in did not fail when a failure occurred for the shutdown or start portion of the command. The Restart operation reported success regardless of the fact the existing process was still running and was not restarted (shutdown failed and start failed). A fix to the Server component class takes stop/start subtasks results into account. If there is an error message in the subtask result, the "restart" operation is failed and reports this message.

Resource Grouping

BZ#1118091 - RHQ Agent resources are not being auto-clustered in resource group due to their unique resource keys

The way RHQ Agents used their managing agent’s name as their resource key caused each agent to appear as a distinct resource in the resource group navigation tree. Agent discovery now assigns the value RHQ Agent to a newly discovered agent’s resource key if the managing agent has the same name as the platform resource key. Agents have the same resource key by default, regardless of which platform they are from. This allows agents to be auto-clustered and to appear as a single node in the resource group navigation true.

Inventory,Plugin -- JBoss EAP 6

BZ#1112733 - Adding EAP 6 server with additional outbound socket binding ref to the inventory will incorrectly add the same binding to the mail configuration of the JBoss ON Server resource (RHQ Server)

It was discovered that adding an additional EAP server to the inventory that had an additional outbound socket binding reference added the same binding option to the mail configuration for every EAP server. The Outbound Socket Binding Reference showed radio buttons for all available values in any EAP instance added to the inventory. If the incorrect option from another EAP was selected, the configuration saved successfully and EAP server restarted, it would throw a "missing dependents" Exception and mail could no longer be sent from the server. Query used to select available socket binding options was limited to the instance that is being configured.

Agent,Launch Scripts

BZ#1148186 - rhq-agent-wrapper's config and cleanconfig commands source RHQ_AGENT_START_COMMAND instead of executing it resulting in failures

Users were unable to use the config and cleanconfig commands with the JBoss ON rhq-agent-wrapper if they were using the su or sudo commands in the RHQ_AGENT_START_COMMAND value. JBoss ON encountered errors when attempting to execute these commands with the rhq-agent-wrapper. An environment variable field has been added to the RHQ_AGENT_START_COMMAND value. The RHQ_AGENT_START_COMMAND value can be used with the su or sudo commands to start JBoss ON.

Monitoring -- Other

BZ#1125439 - MeasurementScheduleManagerBean.findSchedulesByCriteria fails due to duplicate baselines for single schedule_id

Attempting to access resource schedules threw an exception due to duplicate baselines present in the schedule_id variable. Schedules would not load and JBoss ON was unable to retrieve the metric data. A unique identifier index has been added to the baseline.schedule value to detect duplicate baselines for a schedule. Schedules now load as expected.

BZ#1064506 - Group Metric Graph on Dashboard empty

Issues with the maximizing the view of the Group Metric Graph on the Dashboard caused missing data when viewing graphs for servers in a Compatible Group, but not on individual resources. Improvements to the Group Metric Chart portlet fixes the originally-reported issue. Users can now correctly see data for servers in compatible groups.

Plugin -- JBoss EAP 6,UI

BZ#1127869 - DataSource and XADataSource standalone and profile components can not be configured when JARs are deployed to EAP server

An issue with the Driver_Name property of a managed deployment containing invalid values prevented users from updating the DataSource and XADataSource component configuration. If there were three or less JARs installed, and the current configuration value did not equal the name of one of the JARs, navigating to a DataSource resource’s configuration page generated a banner error. If there were four or more JARs installed, the property changed from a radio-button to a drop-down list. In this case, the error was not displayed. If the target driver was installed as a module or had a corresponding driver entry in the JBoss EAP server’s configuration, it was not listed in the drop-down. This also prevented configuration from being updated. The fix now only allows selecting a proper JDBC driver in the datasource configuration, which corrects the reported issue.

Usability

BZ#1127872 - Exporting Suspect Metrics report gives blanks for the "Out of Range Factor (%)" column.

A bug in the CSV export code for the Suspect Metrics Report caused exported report data to be missing the "Out of Range Factor (%)" column data. The underlying issue with the CSV export code was fixed, and the column data is now included when exported.

Plugin -- Apache,Usability

BZ#1127874 - Apache httpd VirtualHost ServerName value should be used for resource's name in inventory

JBoss ON was not reflecting the Apache Virtual Host ServerName specified in the httpd.conf virtual host definition. This caused JBoss ON to not display Apache Virtual Hosts correctly: all virtual hosts showed the same name or IP Address. The Virtual Host resource name was shown as the machine name rather than the specified ServerName. The fix changes the behaviour of the Apache Plug-in to select the ServerName and VirtualHost combination as the first priority for resourceName, and fall back to ServerName if required.

Plugin -- Apache

BZ#844235 - JON smnp modules generate a kernel warning

It was discovered that #ifdef macros were incorrectly included in C code for customers compiling SNMP modules. Unnecessary kernel warnings were randomly thrown, and appeared in the logs when running SNMP modules. The fix updates the C code provided for customers compiling SNMP modules to remove erroneous #ifdef lines. This prevents spurious kernel warnings, as originally reported.

4.2. Known Issues

Documentation

BZ#1141734 - JON agent native support is not supported for Solaris 10 and Solaris 11 x86

Agent native support is not supported on Solaris 10 and Solaris 11 x86 systems. While EAP 5 is supported on Solaris 10 (x86, x86_64. sparc64) and EAP 6 is supported on both Solaris 10 and Solaris 11 (x86, x86_64. sparc64), the JON agent does not provide native support for these platforms. For more information about supported platforms, see https://access.redhat.com/articles/112523

BZ#1133590 - Unable to login in to 'Red Hat Access' in IE

Known bugs with Microsoft Internet Explorer Cross Origin Resource Sharing (CORS) support prevented users logging onto JON through a HTTP connection, and then following HTTPS links in the iFrame containing the Red Hat Access plug-in. There is no workaround to this issue other than not using Microsoft Internet Explorer to access Red Hat Access through JBoss ON.

Upgrade

BZ#1139765 - metrics_index and Anti Entropy Sessions resources are down after upgrade to jon3.3.er2

Four resources (metrics_index, one_hour_metrics, six_hour_metrics, twenty_four_hour_metrics) are marked as down on the dashboard after upgrade. Some resources may no longer be present (on purpose) during the update and now show as missing. The resource' recorded data is still present, and can be viewed through the missing resources page.

UI

BZ#1138688 - [windows] unable to log in to JON from FF

It was discovered that if the Mozilla Firefox browser clock was out of sync with the server clock by a large amount of time, the user was not able to log in because the session timeout immediately logged the user out of the UI. The fix now correctly synchronizes the computers clocks correctly, which allows users to log in correctly.

BZ#1166324 - Broken Doc Link in JON 3.3 CR03

An issue with URL redirection on the Documentation Set link in the /#Help/Section1/Section1Item2 area of the Help page caused the link to 404. Customers are not currently able to easily access the full set of JBoss ON 3.3 documentation. To workaround the issue, open the correct URL (https://access.redhat.com/documentation/en-US/Red_Hat_JBoss_Operations_Network/3.3/) manually. All other docs links in the Help tab are functioning correctly. A fix for the redirect issue is currently being promulgated, and will be implemented by the end of November 2014.

4.3. Enhancements

Storage Node

BZ#1074633 - RFE: Manage storage node snapshots

Snapshots are generated weekly during scheduled maintenance, and when nodes are (un)deployed. A snapshot consists of hard links to SSTable files, which in themselves consume very little disk space.

When an SSTable was deleted during compaction, space was not reclaimed if the SSTable was included in a snapshot. This behavior caused SSTable data to build-up over time to unacceptable levels. This issue revealed that there was no mechanism to manage snapshots through the UI.

Multiples fixes are now included in the product. To address issues with servers running out of disk space on storage nodes, snapshots are no longer part of the weekly scheduled job, and are disabled by default. When enabled, snapshots can now be copied to a different location with more available storage space. Server-side capability now exists to manage snapshots for the storage cluster. System settings have been introduced, which can be updated through the Storage Administration pages. Users can also enable and disable snapshot management of storage clusters, and set cron expressions to run management tasks regularly.

Database,Operations

BZ#1115725 - Operation history continues to grow unbound with no easy way to purge old history

It is now possible to purge operation history to provide better database performance and reduce table space issues. This feature adds an operation history purge to the data purge job. A new system setting called "Delete Operation History Older Than" is added to Administration > Configuration > System Settings >> Data Manager Configuration Properties. The default for this system setting is 0 days, which means disabled. The db-upgrade adds the new system setting (also set to disabled) to prevent upgrades from automatically forcing an unexpected purge of operation history. Auto-purge and retention configuration was available for alert history and history. However operation history was excluded from these operations. The existing option of going into each resource individually to track down operation history was unacceptable. Customers can now better manage operation history in JBoss ON.

Database

BZ#1069165 - Add support for Oracle 12c

JBoss ON now also supports Oracle 12c as a backend database. Oracle 11c is becoming end of life.

Launch Scripts

BZ#1061311 - rhqctl contains user modifiable values that result in an unsupported rhqctl script

The way documentation was worded in the rhqctl script suggested that environment variables could be modified directly in the script, rather than in a sourced file. Users who modified the script directly could have lost their changes if this script was updated in subsequent product releases. The script now adds support for a /bin/rhq-server-env.sh|bat file. The file is not mandatory, and is sourced by the tool only if it is present. The rhqctl script also includes error handling relating to this file to verify whether environment variables are required but not supplied by the user. The improved features of rhqctl remove any confusion about where to specify environment variables.

Document History

Revision History
Revision 3.3-652014-11-26Jared Morgan
Prepared for JBoss ON 3.3 GA
Corrected some link issues in the New Features section.
Removed Install and Upgrade Instructions, and linked to the Installation Guide for the full procedures.

Legal Notice

Copyright © 2015 Red Hat.
The text of and illustrations in this document are licensed by Red Hat under a Creative Commons Attribution–Share Alike 3.0 Unported license ("CC-BY-SA"). An explanation of CC-BY-SA is available at http://creativecommons.org/licenses/by-sa/3.0/. In accordance with CC-BY-SA, if you distribute this document or an adaptation of it, you must provide the URL for the original version.
Red Hat, as the licensor of this document, waives the right to enforce, and agrees not to assert, Section 4d of CC-BY-SA to the fullest extent permitted by applicable law.
Red Hat, Red Hat Enterprise Linux, the Shadowman logo, JBoss, OpenShift, Fedora, the Infinity logo, and RHCE are trademarks of Red Hat, Inc., registered in the United States and other countries.
Linux® is the registered trademark of Linus Torvalds in the United States and other countries.
Java® is a registered trademark of Oracle and/or its affiliates.
XFS® is a trademark of Silicon Graphics International Corp. or its subsidiaries in the United States and/or other countries.
MySQL® is a registered trademark of MySQL AB in the United States, the European Union and other countries.
Node.js® is an official trademark of Joyent. Red Hat Software Collections is not formally related to or endorsed by the official Joyent Node.js open source or commercial project.
The OpenStack® Word Mark and OpenStack logo are either registered trademarks/service marks or trademarks/service marks of the OpenStack Foundation, in the United States and other countries and are used with the OpenStack Foundation's permission. We are not affiliated with, endorsed or sponsored by the OpenStack Foundation, or the OpenStack community.
All other trademarks are the property of their respective owners.