Upgrade issues for Infrastructure Management

Restoring the earlier version of TrueSight Infrastructure Management Server if the upgrade fails

When you upgrade the TrueSight Infrastructure Management Server to 11.0, the installer automatically backs up the installation and configuration files of the earlier TrueSight Infrastructure Management installation in the <installation directory>10.x folder, where x is the version number from which you are upgrading. For example, TrueSight10.7.

Similarly, when you upgrade from ProactiveNet 9.6.00.002 to TrueSight Infrastructure Management 11.0, the installer automatically backs up the installation and configuration files in the <installation directory>9.6.x folder. For example, ProactiveNet9.6.00.002.

However, the SAP SQL Anywhere and Cell database files are not copied into the folder. BMC recommends that you back up the SAP SQL Anywhere and Cell database files before upgrading the TrueSight Infrastructure Management Server. For information about backing up the SAP SQL Anywhere database and the cell configuration files, see Prerequisites-for-Infrastructure-Management-upgrade.

If the upgrade fails, complete the following tasks to restore the last working version of TrueSight Management Server:

1. Make sure all the processes are stopped.
2. Rename the current <installation directory> folder to <installation directory>_bad.
3. Rename <installation directory>10.x folder to <installation directory>.
   For example, rename TrueSight10.7 to TrueSight.
4. Perform the following tasks:
   • Copy the SAP SQL Anywhere and cell database files that you backed up earlier.
   • Ensure that system environment variables such as BMC_PROACTIVENET_HOME, MCELL_HOME, and point to the renamed folder.
   • In the renamed folder, copy the BPPMServerInstalledConfiguration.xml file that you backed up before starting the upgrade process.
   • Remove the upgrade entry from the ProductRegistry.xml file. Ensure that the file contains the correct entry of the previously installed version.
   • For Windows, ensure that the registry entry points to the renamed folder, and it contains the correct entry of the previously installed version
     HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\Microsoft\Windows\CurrentVersion\Uninstall\BPPMServer

A java.lang.NumberFormatException appears after upgrade

Explanation:
The pw\custom\pronet.conf file does not end with a new line.

User Response:
Open the pw\custom\pronet.conf file, and add a new line at the end. (Press Enter.)

Reusing customized local actions after upgrading

Explanation:
You can reuse customized local actions after upgrading TrueSight Infrastructure Management Server.

User Response:
You can reuse customized local actions by completing one of the following procedures:
Procedure 1

1. Extract the localactions.jar file from the location to which you copied it during the upgrade.
2. Open the LocalAction.xml file and copy all customized local actions to installationDirectory\tomcat\webapps\localaction\localactions.jar. When copying customized local actions, replace all the <Role> tags with the <UserGroup> tag. Change the values of roles to groups because from this version of TrueSight Infrastructure Management onwards, permission to access any local action depends on user groups.
3. Run the pw system start command to restart TrueSight Infrastructure Management server.
   
   Procedure 2
    1. Backup the localactions.jar file from the installationDirectory\tomcat\webapps\localaction.
   
   2. Copy the localactions.jar file from the location to which you copied it during the upgrade, and save it in the inatallationDirectory\tomcat\webapps\localaction location.
   
   3. Open installationDirectory\custom\conf\pronet.conf and add the following property: collectorPermissionBasedOn=2
   
    4. Run the following command on the client system to delete the .jar files related to the earlier versions of local actions:
   javaws -viewer

Validation error message as 'More swap space required'

Explanation:

During upgrade, the installer validates the swap space and if there is less swap space as compared to the recommended space, the installer throws the validation error message as 'More swap space required'.

User Response:
Increase the swap space and run the upgrade again. For more information, refer the following links:

• (Windows): Changing the size of virtual memory on Windows
• (Linux): Increasing swap space on Linux

Integration Service is stopped after TrueSight Infrastructure Management installation

When the TrueSight Infrastructure Management Server and Integration Service is installed on a same Linux computer, after you upgrade the TrueSight Infrastructure Management Server, the Integration Service is stopped. You must start the Integration Service manually.

While upgrading TrueSight Infrastructure Management Server on Linux, the console displays an exception

On a Linux computer, when you upgrade to TrueSight Infrastructure Management Server 11.0, the console displays the following exception:

Exception in thread "AWT-EventQueue-0" java.lang.ClassCastException: sun.awt.image.BufImgSurfaceData cannot be cast to sun.java2d.xr.XRSurfaceData
at sun.java2d.xr.XRPMBlitLoops.cacheToTmpSurface(XRPMBlitLoops.java:145)
at sun.java2d.xr.XrSwToPMBlit.Blit(XRPMBlitLoops.java:353)

This exception is displayed due to a known JRE issue. You can ignore it.

Troubleshooting problems faced while migrating data from ProactiveNet to TrueSight Presentation Server

After running the migration utility, it is possible that you find some data, such as policies, users, and deployable packages were not migrated. In this scenario, you can detect the exact problem in the log files located at <Infrastructure-Management-Installation-Directory>\pw\pronto\lib\Datamigrator\logs.

Additionally, you can use the suggested solutions described in following table.

JBoss service does not start

Explanation

When you upgrade the TrueSight Infrastructure Management server to version 11.0 or later, the JBoss service does not start.

User response

After upgrading TrueSight Infrastructure Management server, ping the local host and make sure that it is available. For example, if you are using the IPv6 protocol on Linux, use the ping6 command to make sure that the local host is available.

Silent upgrade of TrueSight Infrastructure Management stops abruptly on a Windows computer

On a Windows computer, upgrading TrueSight Infrastructure Management without stopping the custom cells in your environment can result in some installation files getting locked. As a result, the upgrade process stops and you are unable to rerun the upgrade.

Solution: Perform the following steps:

1. Click Start > Run, type regedit, and click OK.
2. Locate and click the HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\Microsoft\Windows\CurrentVersion\Uninstall\BPPMServer registry key.

3. In the right pane, locate all the values that contain 11.0 as part of the installation directory path (in the data) and modify them to remove the 11.0 reference.

   Example

   Suppose the InstallLocation value displays the data as C:\Program Files\BMC Software\TrueSight11.0. In this scenario. modify the InstallLocation value to display the data as C:\Program Files\BMC Software\TrueSight.

4. Run the pw system stop command to stop the TrueSight Infrastructure Management Server.
5. Rerun the silent upgrade.

Cannot log in to TrueSight Infrastructure Management Server after upgrade

After you upgrade the TrueSight Infrastructure Management server, you cannot log in. This might happen because after the upgrade, the default realm in the Solution Administrator authorization profile or any other authorization profile is not updated. The default realm remains BMCRealm.

Solution

Update the authorization profile with the default realm, and add the Administrators user group of the default Realm in Remedy Single-Sign On (RSSO).

Cannot upgrade TrueSight Infrastructure Management Server silently

During a silent upgrade of the TrueSight Infrastructure Management Server, the following message appears, and the upgrade process stops:

Solution

Do the following:

1. Open a different instance of the command line interface, and run the following command:
   pw sys stop
2. Continue with the upgrade process.

Upgrade to a fix pack or a feature pack fails because of failed Cell compilation

Upgrade to a fix pack or a feature pack might fail because of Cell compilation issues. Perform the following steps to roll back the Fix Pack changes if the upgrade process fails.

1. Ensure that no other process is running that uses any files in the pw directory. Do the following:
   1. Run the pw p l command.
   2. Ensure that the status of all the process listed is Not running.
2. If any processes are running, stop TrueSight Infrastructure Management by using the pw sys stop command.
3. Restore the contents of the pw directory. Perform one of the following tasks:
   • Do the following on a Windows operating system:
     1. Go to the <install_location>\TSIM_BACKUP_FOLDER\pw directory.
     2. Copy all the contents of the pw directory.
     3. Paste the contents in the <install_location>\pw directory.
   • Do the following on a Linux operating system:
     1. Run the following command:
        cp -rf /<install_location>/TSIM_BACKUP_FOLDER/pw  /usr/pw 
        
     2. When you get the cannot overwrite non-directory `/<install_location>/pw/tomcat/webapps/pronto/jsp' with directory `tomcat/webapps/pronto/jsp' message, copy the JSP files from the <install_location>/TSIM_BACKUP_FOLDER/pw/tomcat/webapps/pronto/jsp directory and copy them to the /usr/pronto/web/html/jsp directory.
4. Delete the following files from the <install_location>/pw/pronto/conf directory:
   • federatedchacheserver-transport-tcp.xml
   • clustermanager.xml
   • pnagentcntl.conf
   • pnrate.conf
   • pnjserver.conf
   • pronet.conf

5. Make the following changes:
6. Restore the product_registry.xml file from the location where you backed up the file while preparing to apply the fix pack.
7. If you selected to migrate the KB, do the following:
   1. Delete <install_location>/pw/server/etc/<cell>.
   2. Rename <install_location>/TSIM_BACKUP_FOLDER/pw/server/etc/<cell>.org to <install_location>/pw/server/etc/<cell>.
8. (Windows) If you do not plan to upgrade to version 11.0.00.003 immediately after the rollback, make changes to the Windows Registry to reflect the correct component version. Do the following to make changes to the Windows Registry:
   1. Run the regedit command from the Start menu to open the Registry Editor.
   2. Go to HKEY_LOCAL_MACHINE > SYSTEM > ControlSet001 > services > mcell_<cell>, and change the version number against the Impact Manager service description.
   3. Go to HKEY_LOCAL_MACHINE > SYSTEM > CurrentControlSet > services > mcell_<cell>, and change the version number against the Impact Manager service description.

   4. Go to HKEY_LOCAL_MACHINE > SOFTWARE > Wow6432Node > Microsoft\Windows > CurrentVersion > Uninstall > BPPMServer, and change the version number against the DisplayName and DisplayVersion values.
      
      

      Note

      If you do not want to modify the Windows registries at this time, you will see the Fix Pack version in the Impact Manager service description even if you have rolled back the Fix Pack installation. You will also see the Fix Pack version in the Control Panel.

9. Start the TrueSight Infrastructure Management server by using the pw sys start command.
10. If you have any custom Cells, start them.
11. If you have a BMC Service Resolution integration, restart it.

Upgrade to a fix pack displays the null pointer exception

When you upgrade to TrueSight Infrastructure Management Server 11.0 Fix Pack 3, the upgrade installer displays the following exception:

Workaround for the standalone Infrastructure Management Server deployment:

1. Log in to the Infrastructure Management Server host computer.
2. Stop the Infrastructure Management Server by running pw server stop command.
3. Go to the <TrueSight Infrastructure Management Installation Directory>\pw\pronto\conf directory and locate the federatedcacheserver-transport-tcp.xml file.
4. Take a back up of the federatedcacheserver-transport-tcp.xml file.
5. Open and edit the federatedcacheserver-transport-tcp.xml file using a text editor.

6. Replace the contents of the file with the code snippet shown in the following code block, and save the file:

   Replace the federated-transport.xml file with the content shown in the code block

   <config xmlns="urn:org:jgroups" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="urn:org:jgroups http://www.jgroups.org/schema/JGroups-3.6.xsd">
     <TCP bind_addr="${jgroups.tcp.address:127.0.0.1}" bind_port="${jgroups.tcp.port:10590}" bundler_type="old" enable_diagnostics="false" internal_thread_pool.enabled="true" internal_thread_pool.keep_alive_time="60000" internal_thread_pool.max_threads="10" internal_thread_pool.min_threads="1" internal_thread_pool.queue_enabled="true" internal_thread_pool.queue_max_size="100" internal_thread_pool.rejection_policy="Discard" loopback="true" max_bundle_size="31k" oob_thread_pool.enabled="true" oob_thread_pool.keep_alive_time="60000" oob_thread_pool.max_threads="30" oob_thread_pool.min_threads="2" oob_thread_pool.queue_enabled="false" oob_thread_pool.queue_max_size="100" oob_thread_pool.rejection_policy="Discard" port_range="0" recv_buf_size="20m" send_buf_size="640k" thread_naming_pattern="pl" thread_pool.enabled="true" thread_pool.keep_alive_time="60000" thread_pool.max_threads="30" thread_pool.min_threads="2" thread_pool.queue_enabled="true" thread_pool.queue_max_size="100" thread_pool.rejection_policy="Discard" use_send_queues="true"/>
     <TCPPING ergonomics="false" initial_hosts="localhost[10590]" port_range="0" timeout="30000"/>
     <MERGE3 max_interval="30000" min_interval="10000"/>
     <FD_SOCK/>
     <FD max_tries="5" timeout="30000"/>
     <VERIFY_SUSPECT timeout="30000"/>
     <pbcast.NAKACK2 max_msg_batch_size="100" use_mcast_xmit="false" xmit_interval="1000" xmit_table_max_compaction_time="10000" xmit_table_msgs_per_row="10000" xmit_table_num_rows="100"/>
     
     <UNICAST3 conn_expiry_timeout="0" max_msg_batch_size="100" xmit_interval="500" xmit_table_max_compaction_time="30000" xmit_table_msgs_per_row="10000" xmit_table_num_rows="20"/>
     
     <pbcast.STABLE desired_avg_gossip="5000" max_bytes="1M" stability_delay="500"/>
     <pbcast.GMS join_timeout="3000" print_local_addr="true"/>
     <MFC max_credits="2m" min_threshold="0.40"/>
     <FRAG2 frag_size="30k"/>
     <RSVP ack_on_delivery="false" resend_interval="500" timeout="60000"/>  
   </config>
7. Start the Infrastructure Management Server by running the pw server start command.

Workaround for the high-availability Infrastructure Management Server deployment:

1. Stop the standby node.
2. Stop the active node.
3. Take a backup of the federatedcacheserver-transport-tcp.xml file present in the <TrueSight Infrastructure Management Installation Directory>\pw\pronto\conf directory both on active and standby nodes.
4. Using a text editor open and edit the federatedcacheserver-transport-tcp.xml file in the active node. 

5. Update the properties in the federatedcacheserver-transport-tcp.xml file as explained in the following example, and save the file:

   • Locate the TCP element and modify the bind_addr property. Replace the tsim-win01.bmc.com with the FQDN of the active node.
   • Locate the TCP element and modify the bind_port property. Replace the 10590 port number with the port number that is configured for the active node. 
   • Locate the TCPPING element and modify the initial_hosts property. Replace the tsim-win01.bmc.com[10590] value with the FQDN of the active node and the corresponding port number. Replace the tsim-win02.bmc.com[10590] value with the FQDN of the standby node and the corresponding port number.

   Edit the properties of the federated-transport.xml file on the active node

   <config xmlns="urn:org:jgroups" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="urn:org:jgroups http://www.jgroups.org/schema/JGroups-3.6.xsd">
     <TCP bind_addr="${jgroups.tcp.address:tsim-win01.bmc.com}" bind_port="${jgroups.tcp.port:10590}" bundler_type="old" enable_diagnostics="false" internal_thread_pool.enabled="true" internal_thread_pool.keep_alive_time="60000" internal_thread_pool.max_threads="10" internal_thread_pool.min_threads="1" internal_thread_pool.queue_enabled="true" internal_thread_pool.queue_max_size="100" internal_thread_pool.rejection_policy="Discard" loopback="true" max_bundle_size="31k" oob_thread_pool.enabled="true" oob_thread_pool.keep_alive_time="60000" oob_thread_pool.max_threads="30" oob_thread_pool.min_threads="2" oob_thread_pool.queue_enabled="false" oob_thread_pool.queue_max_size="100" oob_thread_pool.rejection_policy="Discard" port_range="0" recv_buf_size="20m" send_buf_size="640k" thread_naming_pattern="pl" thread_pool.enabled="true" thread_pool.keep_alive_time="60000" thread_pool.max_threads="30" thread_pool.min_threads="2" thread_pool.queue_enabled="true" thread_pool.queue_max_size="100" thread_pool.rejection_policy="Discard" use_send_queues="true"/>
     <TCPPING ergonomics="false" initial_hosts="tsim-win01.bmc.com[10590],tsim-win02.bmc.com[10590]" port_range="0" timeout="30000"/>
     <MERGE3 max_interval="30000" min_interval="10000"/>
     <FD_SOCK/>
     <FD max_tries="5" timeout="30000"/>
     <VERIFY_SUSPECT timeout="30000"/>
     <pbcast.NAKACK2 max_msg_batch_size="100" use_mcast_xmit="false" xmit_interval="1000" xmit_table_max_compaction_time="10000" xmit_table_msgs_per_row="10000" xmit_table_num_rows="100"/>
   
     <UNICAST3 conn_expiry_timeout="0" max_msg_batch_size="100" xmit_interval="500" xmit_table_max_compaction_time="30000" xmit_table_msgs_per_row="10000" xmit_table_num_rows="20"/>
   
     <pbcast.STABLE desired_avg_gossip="5000" max_bytes="1M" stability_delay="500"/>
     <pbcast.GMS join_timeout="3000" print_local_addr="true"/>
     <MFC max_credits="2m" min_threshold="0.40"/>
     <FRAG2 frag_size="30k"/>
     <RSVP ack_on_delivery="false" resend_interval="500" timeout="60000"/>  
   </config>
6. Log in to the Infrastructure Management Server host computer that is running as the standby node.
7. Using a text editor open and edit the federatedcacheserver-transport-tcp.xml file in the standby node. 

8. Update the properties in the federatedcacheserver-transport-tcp.xml file as explained in the following example, and save the file:

   • Locate the TCP element and modify the bind_addr property. Replace the tsim-visa-win02.bmc.com with the FQDN of the standby node.
   • Locate the TCP element and modify the bind_port property. Replace the 10590 port number with the port number that is configured for the standby node. 
   • Locate the TCPPING element and modify the initial_hosts property. Replace the tsim-win02.bmc.com[10590] value with the FQDN of the standby node and the corresponding port number. Replace the tsim-win01.bmc.com[10590] value with the FQDN of the active node and the corresponding port number.

   Edit the properties of the federated-transport.xml file on the standby node

   <config xmlns="urn:org:jgroups" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="urn:org:jgroups http://www.jgroups.org/schema/JGroups-3.6.xsd">
     <TCP bind_addr="${jgroups.tcp.address:tsim-visa-win02.bmc.com}" bind_port="${jgroups.tcp.port:10590}" bundler_type="old" enable_diagnostics="false" internal_thread_pool.enabled="true" internal_thread_pool.keep_alive_time="60000" internal_thread_pool.max_threads="10" internal_thread_pool.min_threads="1" internal_thread_pool.queue_enabled="true" internal_thread_pool.queue_max_size="100" internal_thread_pool.rejection_policy="Discard" loopback="true" max_bundle_size="31k" oob_thread_pool.enabled="true" oob_thread_pool.keep_alive_time="60000" oob_thread_pool.max_threads="30" oob_thread_pool.min_threads="2" oob_thread_pool.queue_enabled="false" oob_thread_pool.queue_max_size="100" oob_thread_pool.rejection_policy="Discard" port_range="0" recv_buf_size="20m" send_buf_size="640k" thread_naming_pattern="pl" thread_pool.enabled="true" thread_pool.keep_alive_time="60000" thread_pool.max_threads="30" thread_pool.min_threads="2" thread_pool.queue_enabled="true" thread_pool.queue_max_size="100" thread_pool.rejection_policy="Discard" use_send_queues="true"/>
     <TCPPING ergonomics="false" initial_hosts="tsim-visa-win02.bmc.com[10590],tsim-visa-win01.bmc.com[10590]" port_range="0" timeout="30000"/>
     <MERGE3 max_interval="30000" min_interval="10000"/>
     <FD_SOCK/>
     <FD max_tries="5" timeout="30000"/>
     <VERIFY_SUSPECT timeout="30000"/>
     <pbcast.NAKACK2 max_msg_batch_size="100" use_mcast_xmit="false" xmit_interval="1000" xmit_table_max_compaction_time="10000" xmit_table_msgs_per_row="10000" xmit_table_num_rows="100"/>
     
     <UNICAST3 conn_expiry_timeout="0" max_msg_batch_size="100" xmit_interval="500" xmit_table_max_compaction_time="30000" xmit_table_msgs_per_row="10000" xmit_table_num_rows="20"/>
     
     <pbcast.STABLE desired_avg_gossip="5000" max_bytes="1M" stability_delay="500"/>
     <pbcast.GMS join_timeout="3000" print_local_addr="true"/>
     <MFC max_credits="2m" min_threshold="0.40"/>
     <FRAG2 frag_size="30k"/>
     <RSVP ack_on_delivery="false" resend_interval="500" timeout="60000"/>  
   </config>
9. Start the active node and wait till the active node is up and running.
10. Start the standby node.