Summary of enhancements, fixes, and open issues.
There is no supported upgrade path from CA UIM 8.5 beta to CA UIM 8.5 GA.
CA UIM 8.5 uses hub and robot version 7.80. If you have applied post-7.80 patches to your primary hub, they will be overwritten during the 8.5 upgrade. You will need to redeploy your hotfix on the primary hub to have your current expected hub and robot version.
This article explains the new features in this release. It also details the international support, product accessibility, and third-party software agreements for CA UIM.
This release of CA UIM includes the following new features:
Predefined CA Business Intelligence Dashboards
New dashboards are available for the following applications:
- Summary and default dashboards- You can use these dashboards to view alarms for devices, probes, and groups in a CA UIM environment. These dashboards work with any probe in your CA UIM environment.
- Azure- Requires the azure probe version 3.00 or later.
For more information, see the CA Business Intelligence Dashboards Release Notes.
This release of CA UIM adds support for monitoring SystemEDGE-enabled devices using the snmpcollector probe (3.3 and later) and CA UIM (8.5 and later). For more information, see the new how-to article on the CA UIM Probes DocOps space titled, Monitor SystemEDGE-enabled Devices with the snmpcollector Probe. The following probe updates were made to support this feature:
discovery_agent Includes CA SystemEDGE Devices
The discovery_agent probe now identifies devices running CA SystemEDGE.
discovery_server Supports CA System EDGE
get_devicescallback now includes the Roles parameter in order to identify CA SystemEDGE devices.
New MCS Profile Types Available
This release includes the following new profile types:
- Email Gateway
- SNMP Gateway
For more information, refer to your probe documentation.
German and Korean Languages Reintroduced
German and Korean language support is once again available.
discovery_agent Allows Custom SNMP Mapping
<snmp>/<models>configuration section has been added to the discovery_agent probe to allow custom mapping from an SNMP sysObjectID to device model and roles.
udm_manager Database Connection Retries Extended
During periods of high database use, such as maintenance periods, udm_manager previously could exit with a no-restart error code, initiating a probe shutdown before it was ready. The udm_manager probe now uses an increased number of retries to establish connection to the database. On unsuccessful completion of the retries, the probe now initiates a restart.
This release includes the following software support updates.
Solaris Support Update
As of this release, Solaris is no longer a supported installation option for the following UIM components:
- UIM Server (primary hub)
- UMP Server
- CABI Server
- Unified Reporter
Solaris robots and secondary hubs are still supported.
For more information regarding supported operating systems, see the Compatibility Support Matrix.
Updated Probes and Packages List
The following probes and packages are installed as part of this CA UIM release.
You can locate the version number for individual probes on both the
Installed Packagestabs in Admin Console.
The correct version number for the trellis probe is located on the
Installed Packagestab. The version number for trellis that appears on the
Probestab displays the version number of the latest installed service.
2.01 on the
Installed Packagestab and in Infrastructure Manager
8.50 on the
You can confirm your CA UIM package versions by navigating to the
Installed Packagestab in Admin Console.
das (data access services)
The following optional probes are downloaded to the archive during CA UIM installation.
The following issues were resolved in this release:
Admin Console Archive View Refresh Issue Corrected (Support Case 105246)
This issue only impacts the Admin Console archive view. The other views in Admin Console do not experience this issue.
In cases where Admin Console is installed but does not have web archive connectivity (either due to lack of an Internet connection or a firewall), the Admin Console archive view could become unavailable due to constant refresh attempts. This issue no longer occurs in CA UIM 8.5.
The discovery_agent Probe Now Correctly Identifies Devices with Multiple SNMPv3 Profiles (Support Case 324554)
Inconsistent SNMPv3 discovery results when there are two or more SNMPv3 profiles being applied to the same IP address with the same user name but different settings are fixed.
The Processes Probe No Longer Saves the Incorrect Name and Password (Support Case 00335972)
In Admin Console 8.47, the processes probe would incorrectly store a user's CA UIM user name and password in the following circumstances:
- The user was attempting to configure the processes probe using Admin Console.
- The user's browser had the enable passwords feature on.
- The process probe configuration file contained empty user name and password fields.
This no longer occurs in the latest release of CA UIM.
USM Close Alarm Performance Improved (Support Case 00582161)
An update to the nas_api_service in trellis has resolved an issue in which USM was slow to load after acknowledging multiple alarms in USM simultaneously.
USM Intermittent Session Error Corrected (Support Case 00523231)
An update to the nas_api_service in trellis has resolved an issue in which USM would intermittently display the following error message:
An unknown error has occurred.Refreshing your browser may resolve the issue.
In these cases, the following error would appear in the portal.log file:
ERROR [PortalInitAction:111] (80) Session error, Unable to open a client session: Connection timed out ERROR [PortalInitAction:111] Caused by: java.net.ConnectException: Connection timed out
USM javax.persistence.PersistenceException Error Corrected (Support Case 00582161)
An update to the nas_api_service in trellis has resolved the following error in USM:
javax.persistence.PersistenceException: org.hibernate.exception.LockAcquisitionException: could not execute queryDetails:com.firehunter.ump.exceptions.DataFactoryException : nullStack Trace:javax.persistence.PersistenceException: org.hibernate.exception.LockAcquisitionException: could not execute queryCaused by: org.hibernate.exception.LockAcquisitionException: could not execute queryCaused by: com.microsoft.sqlserver.jdbc.SQLServerException: Transaction (Process ID 372) was deadlocked on lock | communication buffer resources with another process and has been chosen as the deadlock victim. Rerun the transaction.
wasp Probe Tomcat Update (Support Case 00512587)
The wasp probe has been updated to use Tomcat 7.0.72.
The following items are the current known issues with this release:
Assigned ToDropdown Does Not Sort Alarms Correctly
Sort Bydrop-down to sort by the
Assigned Toalarm parameter results in incorrect sorting of the alarm table list. To correct this, you can sort by clicking the
Assigned Tocolumn in the alarm table.
At a Glance Reports Do Not Support SQL Database QoS Metrics
The At A Glance report will only return the CDM QoS for the SQL Server if CDM/RSP is configured for it. Currently, none of the SQL Database QoS metrics are returned for this report.
Automatic Groups Using Enriched Origins Might Require Reconfiguration
After upgrading to CA UIM 8.4 or later, if you are using Automatic Groups that:
- Use theOriginfield that is configured usingAdministration > Configure Automatic Groups from USM.
- Contain enriched origins
You might see new groups with no members for each enriched origin. To correct this issue, you can either:
- Edit one of the new groups and explicitly configure the proper enriched origin filter. The group will then display the correct group members.
- Create standard dynamic groups using your desired filter criteria.
baseline_engine and prediction_engine Calculations Are Inaccurate After Error Conditions
If prediction_engine and baseline_engine are calculating baselines or new metrics and an error condition occurs, the calculations will be inaccurate from the time the error condition began. After prediction_engine and baseline_engine are restarted, baselines generated by baseline_engine will be accurate after sufficient samples have been collected, and the predictive alarm metrics will start again at the top of the next hour.
Changes to Alarm Filters Are Not Reflected Immediately in USM
If you use Account Admin to modify the alarm filter of an ACL, and then immediately view alarms in USM with that ACL, the alarms that are shown might not reflect the change that is made to the alarm filter. This occurs because the default refresh interval for the alarm cache is set to 5 minutes for performance reasons.
Change Password Portlet Is Not Available After Upgrade
The Change Password portlet is no longer available after upgrading to CA UIM 8.4 or later. Administrators can change a user's password with the Account Admin portlet.
Custom URL Frame Does Not Refresh
If you create a dashboard and add a custom URL, you have the option of selecting an auto-refresh interval for the custom URL frame. Due to a known limitation with JasperReports, if you select an auto-refresh interval, the frame will attempt to refresh at the interval you specify but will not successfully refresh. In addition, whether you select an auto-refresh interval, or accept the default
Manual Onlyrefresh, the custom URL frame will not successfully refresh when you click the refresh button.
Therefore, if you use a custom URL in a dashboard, it is recommended that you:
- Donotselect an auto-refresh interval
- Avoid clicking the refresh button
If you do click the refresh button, you can reload the dashboard and then the custom URL frame displays correctly again.
Dashboard Design Permission Not Honored for Account Contact Users
In the Account Admin portlet, you can add the Dashboard Design permission to an ACL. However, if you assign that ACL to an account contact user, the Dashboard Design permission is not honored. The Dashboard Design permission is only honored when it applies to a bus user.
Existing Unified Reporter Schema Name Issues During Upgrade
Using an existing schema or catalog name on the Database Connection panel during Unified Reporter upgrade can generate an error that the entered name is already in use. To fix this error, use a different schema or catalog name.
Interface Groups Not Supported with Mobile App
Currently, the application does not support interface groups with either Android or iOS devices.
Java-Based Probes Might Not Start on Secondary Robot Deployment
If java-based probes are deployed to a robot that uses a version of Java older than version 7 (java_jre1.7), they might not start. To resolve this, redeploy java_jre7 to the robot and then restart the robot.
Java SDK Environment Variable Requirement
If you use the Java SDK to send messages to UIM, set the following environment variable to designate the loopback IP address for contacting the local robot. For example:
Liferay Issue Might Cause Permissions Error in UMP Portlets
Due to a known issue with Liferay Enterprise Edition (UMP 7.5 and later), users might encounter a permissions error in UMP portlets. This issue can occur in UMP instances that were created as follows: an administrator exported pages as a LAR file, created a site template and imported the LAR file, created a site that is based on the template, and then deleted one or more pages from the template.
This issue can be avoided if you export and import a LAR file with the
Permissionscheckbox selected. When you export pages as a LAR file, select the
Permissionscheckbox under the heading Other. When you create the site template and import the LAR file, ensure that you again select the
If you encounter this issue, the work-around is to select the gear icon in the upper right-hand corner of each UMP portlet. In the Configuration pane, select the
Permissionstab, and then select the
Viewcheckbox for each user in the Role column.
Monitoring Configuration Service Profile Template Field Names
While MCS templates and the associated probe configuration GUIs (either in Admin Console or Infrastructure Manager) for a particular probe will have the same configuration options, individual field names might not always match between them. For more information about the available fields for each probe GUI, see the specific documentation for your probe.
Name Resolution Conflicts on Debian v6 with the automated_deployment_engine Probe
By default Debian v6 uses the address 127.0.1.1 as the name resolution address. When a robot is deployed to a Debian 6 system using automated_deployment_engine, after system restart the robot attempts to bind to 127.0.1.1 as the address it is available at. Use the following work-around to avoid contention for 127.0.1.1 on your Debian 6 system:
- When installing the robot manually or with automated_deployment_engine, you must go to the target system after installation and must add the following line to the robot.cfg file:
robotip = ip_address
ip_addressis the desired IP address that the robot should bind to on the target system.
- When deploying to Debian 6.0.5 using XML, you must define the<robotip>ip_address</robotip>option, whereip_addressis the IP address that the robot should bind to on the target system.
Probe Administration Returns an Unknown Error in a Windows Cluster Environment
When you attempt to configure a probe, probe administration might return an
unknown errormessage in a Windows Cluster environment. Specifically, this problem has occurred with the ems probe, but can happen with other probes. To work around the issue, follow these steps:
- Log in to Infrastructure Manager as the administrator.
- On the Security tab, selectProbe Administration.
- ClickNew Probeto add an entry for the ems probe.
- Select ems in theProbedrop-down list.
- Select Admin in theAccessdrop-down list.
- Enter an asterisk ( * ) in theIP Maskfield, and clickOKto save your changes.
Robot Communication Failure on Unix When Using Nimldr
If you are using Nimldr, robot communication over the network can fail due to an invalid /etc/hosts file. If this occurs, ensure that the /etc/hosts file on any system hosting a UIM robot, hub, server, or UMP instance contains a valid entry for the local system. This must be a hostname (does not need to be fully qualified) and IP address. If only loopback is defined (for example, 127.0.0.1 localhost), then the controller probe on that system will be unaware of its own IP address, resulting in network communication failure.
Other methods of robot or hub deployment do not require /etc/hosts configuration.
Because the associated probes are currently unavailable, you cannot deploy the following MCS templates in CA UIM 8.5:
- Log Forwarding
- Log Forwarding Apache
- Log Forwarding Catalina
- Log Forwarding Log4j
- Log Forwarding Oracle Alert
- Log Monitoring
- Log Monitoring Service
- Setup axa_log_gateway
- Setup log_forwarder
- Setup log_monitoring_service
Unknown Error Occurs During Some Alarm Operations
You might see the following error when using the
Set invisible, or
Assignicons in the USM Alarm View:
An unknown error has occurred.Refreshing your browser may resolve the issue.Details:com.firehunter.ump.exceptions.DataFactoryException : nullStack Trace:org.springframework.orm.jpa.JpaSystemException: org.hibernate.jdbc.BatchedTooManyRowsAffectedException: Batch update returned unexpected row count from update ; actual row count: 2; expected: 1; nested exception is javax.persistence.PersistenceException: org.hibernate.jdbc.BatchedTooManyRowsAffectedException: Batch update returned unexpected row count from update ; actual row count: 2; expected: 1Caused by: javax.persistence.PersistenceException: org.hibernate.jdbc.BatchedTooManyRowsAffectedException: Batch update returned unexpected row count from update ; actual row count: 2; expected: 1Caused by: org.hibernate.jdbc.BatchedTooManyRowsAffectedException: Batch update returned unexpected row count from update ; actual row count: 2; expected: 1
You can safely ignore this error.
wasp Fails to Start Without mpse
If mpse is not deployed and running on the same robot as wasp, or if wasp does not indicate in its configuration file where mpse is running, then wasp fails to start. This results in UMP and any other webapps failing to load. The work-around for this issue is to stop the wasp probe and edit the wasp configuration to point to the hub address (for example, /domain/hub/robot/mpse) where mpse is deployed. Ensure that mpse is running when you restart wasp.
WebSphere and Vblock Unified Dashboard Helps Link Are Incorrect
The help links for the Websphere and Vblock Unified Dashboards return a
No Topics Founderror. The correct links are:
Include localization details for your release.
The user interface is available in the following languages:
- Brazilian Portuguese
- Simplified Chinese
The user documentation is available in the following languages:
- Brazilian Portuguese
Third-Party Software Agreements
List the TPSRs by name and version. Ensure that users have access to the licenses. For example, attach a zip of the licenses to the page and update the download link.
- aspectjrt 1.8.2
- aspectjrt 1.8.5
- Commons IO 2.4
- commons-compress 1.10
- commons-compress 1.12
- Groovy 2.4.5
- Guava 18.0
- infinispan_commons 7.2.5.final
- infinispan-core 7.2.5.final
- infinispan-spring4 7.2.5.final
- jackson-annotations 2.6.3
- jackson-core 2.6.3
- jackson-databind 2.6.3
- jackson-dataformat-csv 2.5.4
- jackson-datatype-hibernate4 2.6.3
- jackson-datatype-json-org 2.6.3
- jackson-module-jaxb-annotations 2.6.3
- jersey client 2.22.1
- Jettison 1.3.7
- Log4j 1.3.2
- mapdb 1.0.8
- Quartz 2.2.2
- spring-framework 4.2.2.RELEASE
- swagger 2.2.2
- Tomcat 7.0.69
- zip4j 1.3.2