General troubleshooting

If you are having problems with your OMi installation and need to contact HPE Software Support, use the opr-checker tool to generate a summary of your installation.

1. Run the opr-checker tool from the following location and specify a target location for the output file:

   <OMi_HOME>/opr/support/opr-checker.[bat|pl] -a > <TempDir>/opr-checker.xml

2. Send the resulting XML file to Software Support for analysis.

Verify that all ports required by OMi servers are not in use by other applications on the same machine. To do so, open a command prompt or terminal window, and run netstat (or use any utility that enables you to view port information). Search for the required ports.

You can also check the <OMi_HOME>/log/jboss7_boot.log for ports in use. If the jboss7_boot.log reports "Port <> in use" but you do not see that this port is in use when you run netstat utility, restart the server and then start OMi.

For details on the ports required by OMi, see Port Usage.

The OMi logging and tracing capabilities for Apache Flex-based UIs help simplify on-site troubleshooting.

Logging is designed to be run on the client system and uses JavaScript to send logging and tracing output to a browser window. You can enable logging and tracing on demand and save the output to aid troubleshooting.

To enable logging, do the following:

1. Log on to OMi:

   https://<server_name>.<domain_name>/omi

2. Open the following URL to open the logging configuration application.

   https://<server_name>.<domain_name>/opr-admin-server/logging/logging.html

   The Logging Settings dialog box opens.

3. To enable logging, select the Enable logging for this session check box.

4. Select the required logging level. Each log level includes the higher levels:

   Critical: Shows only problems of level Critical.

   Error: Shows problems of level Error and Critical.

   Warning: Shows problems of level Warning, Error, and Critical.

   Info: Shows problems of level Info, Warning, Error, and Critical. The resulting information provides a brief overview over the flow of the executed steps.

   Debug: Shows problems of level Info, Warning, Error, and Critical. The resulting information provides a detailed, verbose view over the flow of the executed steps.

5. Optional. Set a filter value to find instances of the specified text strings. By default, all messages that have a category starting with com.hp. are logged. You can restrict this to certain categories to filter the output or extend it, for example, to show messages from the Flex framework.

6. Click Apply.

   The settings become effective for the current browser session.

7. Restart the user interface that you want to trace.

   As soon as the first log output is sent, a browser window opens that shows the log messages.

   Note Each time you change the logging settings you must restart the user interface that you want to trace.

   The settings are for applications that are served from the same domain as that of the current browsing session. If you restart the web browser, you must reapply the logging settings.

To save the log information, do the following:

1. In the browser window that shows the log messages, click Export to log4j.

2. In the Opening GuiTrace.log dialog box, select whether you want to open or save the GuiTrace.log file.

If a high number of events with ETIs is forwarded to OMi over a long period of time (30 minutes or longer), the correlation engine only considers up to the specified number of most recent events. If this limit is exceeded, the oldest events are removed from the queue and no longer considered for correlation.

This limit is set using the Max Waiting Queue Size attribute and is located in:

Administration > Setup and Maintenance > Infrastructure Settings > Operations Management > Topology-Based Event Correlation Settings

The default value is 5000. The valid range is 100 to 20000.

If you are experiencing this problem, lower the incoming event rate or increase the Max Waiting Queue size limit. If the limit is increased, you should also monitor the memory consumption and, if necessary, increase the memory setting (parameter -Xmx) for the opr-backend process.

• No rule is selected in the Correlation Rules pane

• No view is active in the Rules Topology pane

• No configuration item type is selected in the Rules Topology pane

• No indicator is defined for the selected CI type

The rule is invalid or incomplete, for example:

• The rule does not have at least one symptom event

• The rule does not have a cause event

• The topology path is invalid

• The rule refers to a configuration item that is not resolvable

• This issue is related to the Security Settings of Internet Explorer.

  To display the title in place of the URL, go to:

  Internet Options> Security> Internet Zone> Custom Level

  Enable Allow web sites to open windows without address or status bars.

If the links from RTSM Administration do not work, this may be caused by one of the following:

• Make sure that the OMi gateway server is able to access the Default Virtual Server for Application Users URL. This URL can be found in Infrastructure Settings:

  Administration > Setup and Maintenance > Infrastructure Settings

  Alternatively, click Infrastructure Settings.

  In the Foundations field, specify Platform Administration. The URL is located in the Host Configuration table.

• If you are using a reverse proxy or load balancer, make sure you log in through the URL specified above.

The Recipients page does not open, or the recipients page opens, but recipients cannot be added or modified.

This error occurs if the default templates for recipients were not loaded when OMi was installed. Do the following to fix this issue:

1. Access Administration > RTSM Administration > Package Manager.

2. Click Deploy packages to server (from local disk).

3. Click Add, select the BSMAlerts.zip file and click Open. Deploy the package.

   You may have to copy the BSMAlerts.zip file from the OMi system to your local system.

Open Control Panel > Java > Temporary Internet Files > Settings and make sure Keep temporary files on my computer is selected. If the problem persists, clear the Java cache by clicking Delete Files in the same location.

Possible Cause:

Old java files on your client computer.

Solution:

Clear the Java cache by following this procedure:

1. Open Control Panel > Java > Temporary Internet Files > Settings.

2. In the Temporary Internet Files section, click Settings.

3. In the Temporary File Settings dialog box, click Delete Files.

This is due to Java security enhancements. Before Java applets and Java Web Start applications run, the signing certificates are checked to ensure that a signing certificate was not revoked.

If there is a proxy problem or any other network issue, you should disable this Java feature.

To disable Java from performing certificate revocation checks:

1. Open the Control Panel.
2. Click the Java icon to open the Java Control Panel.
3. Click the Advanced tab.
4. Locate and deselect the option to perform certificate revocation checks.

After the reconfiguration of OMi in a distributed environment, the workspaces menu is empty.

Possible Cause:

This error might occur if a new database is created on the data processing server, with the uimashup files being on the gateway server.

Solution:

If the database was recreated during the reconfiguration, run the following commands to copy the uimashup files to the correct location:

Windows: xcopy /S /Y "<OMi_Home>\conf\uimashup\import\loaded\*" "<OMi_Home>\conf\uimashup\import\toload"

Linux: -rf /opt/HP/BSM/conf/uimashup/import/loaded/* /opt/HP/BSM/conf/uimashup/import/toload

After reloading the Event Browser (HTML), Japanese view names are displayed as question marks, and no events are listed.

Solution:

This issue occurs if the Microsoft SQL Server database is not installed on a Japanese operating system. To resolve the issue, install the MS SQL database on a Japanese operating system.

OMi requires sticky sessions for users. Make sure the persistency settings are set to stickiness by session enabled or Destination Address Affinity (depending on the load balancer).

• Check the KeepAlive URIs.

• Virtual hosts and load balancer must be configured with a fully qualified domain name (and not an IP) for LW-SSO to work.

Connectivity problems include the inability to log into OMi.

Possible Cause:

This can happen if the TopazInfra.ini file is empty or corrupt.

To verify that this is the problem:

1. In a browser on the gateway server, type https://localhost:29000 to connect to the JMX console.

If prompted, enter the JMX console authentication credentials (if you do not have these credentials, contact your system administrator).

2. Under Topaz, select Topaz:service=Connection Pool Information.

3. Click the showConfigurationSummaryInvoke button toward the bottom of the page. If the Operation Result page is blank, the TopazInfra.ini file is empty or corrupt.

Solution:

To solve this problem, rerun the OMi configuration wizard and either reconnect to your existing management database or define a new management database. If you did not discover a problem with the TopazInfra.ini file, contact HPE Software Support.

Possible Cause:

The page file size is too small.

Solution:

Configure the page file size to be at least 150% of RAM size. Restart the server.

You can track who attempted to log into OMi:

1. Open the following file:

2. <OMi_HOME_GW>/log/jboss/UserActions.servlets.log

You can display a list of the users that are currently logged in:

1. Open the JMX console on the machine that is running OMi. (For detailed instructions, see JMX Console.)

2. Under the Topaz section, select service=Active Topaz Sessions.

3. Invoke the java.lang.String showActiveSessions() operation.

If the High Availability Controller's Automatic Failover mode is enabled and the management database has been down for some time, some processes may be stopped and will not resume automatically when the management database returns to normal operation. These processes will have the status STARTING on the OMi serverStatus.jsp tool (Windows: <OMi_HOME>\topaz\serverStatus.jsp, Linux: /opt/HP/BSM/topaz/serverStatus.jsp).

Solution:

Restart these processes once the management database is available again.

Possible Cause:

The handling of the high availability information has been changed with RTSM 10.22, and due to this the following messages might be logged in /log/odb_boot.log after an update to 10.62 or higher:

SEVERE: <db>_rtsm: failed removing row 1: java.io.EOFException; please delete it manually

Solution:

The messages can be ignored, but they will make it difficult to find real problems in the log file. Stop the messages by emptying the following two tables in the RTSM database:

MS SQL databases:

DELETE FROM [TABLE_NAME].[dbo].[HA_WRITER]

DELETE FROM [TABLE_NAME].[dbo].[JGROUPSPING]

Oracle databases:

DELETE FROM "TABLE_NAME"."HA_WRITER"

DELETE FROM "TABLE_NAME"."JGROUPSPING"

PostgreSQL databases:

DELETE FROM HA_WRITER;

DELETE FROM JGROUPSPING;