Previous | Next


Migrate nodes


Step 16: Scenario 2: Migrating cells across operating systems using the command-line tools

These steps apply to cross operating system migrations only. If you are not completing a cross operating system migration of a node, see Scenario 1: Migrating a cell using the command-line tools. Ensure that the v8.0 dmgr is running. For each node that you plan to migrate to v8.0, perform the following steps:

For the migration to be successful, use the same source node name and cell name for each node that you migrate to v8.0 or later.


Procedure

  1. Install WAS v8.0.

  2. Create the target node managed profile. 
    cd v8_install_root/bin/
./manageprofiles.sh -create
                  -profileName node1
                  -templatePath /opt/WASv8/profileTemplates/managed
                  -nodeName currentNode1Name
                  -cellName currentCellName
                  -hostName mynode1host.company.com

  3. Use the updated supplemental CD you created to save the current node X configuration information to a migration backup directory. Choose a new directory for the backup files. For example: 
    cd /supplemental/migration/bin/
./WASPreUpgrade.sh /mybackup_old_host/v61tov8node1 /opt/WASv61
                  -oldProfile 61node1
                  -machineChange true

  4. Check the WASPreUpgrade console output for error and warning messages. You might find the following messages: "Failed with errors" or "Completed with warnings". Also, look in the following log files for error or warning messages:

    • myback_old_host/v61tov8node1/logs/WASPreMigrationSummary.log
    • myback_old_host/v61tov8node1/logs/WASPreUpgrade.timestamp.log
    • myback_old_host/v61tov8node1/logs/WASPreUpgrade.trace

    If the WASPreUpgrade command is successful, you do not need to check the log files for error or warning messages.

  5. Use the archive tool of your choice to create a compressed file of the backup directory created by the WASPreUpgrade command. For example: 

    cd /mybackup_old_host
/opt/WASv61/java/bin/jar -cf v61tov8node1.jar v61tov8node1/

  6. Move the archived file to the target machine.

  7. Create a directory on the target machine and extract the archived file to the new directory. For example: 
    mkdir /mybackup_new_host
cd /mybackup_new_host
/opt/WASv8/java/bin/jar -xf v61v8dmgr01.jar

    ...where mybackup_new_host is the directory to which you are migrating your files.

  8. Stop the application servers on the old node, then stop the node agent on the old node.

  9. Stop and disable the dmgr on the old host. Ensure that you do not use the dmgr on the old host.

    To disable this dmgr, rename the associated serverindex.xml file as indicated in the following information:

    Old name

    $PROFILE_ROOT/config/cells/cell_name/nodes/node_X/serverindex.xml

    New name

    $PROFILE_ROOT/config/cells/cell_name/nodes/node_X/serverindex.xml_disabled

  10. Run the WASPostUpgrade command to restore the saved node X configuration into the new v8.0 managed profile. For example: 
    cd v8_install_root/bin/
./WASPostUpgrade.sh mybackup_new_host/v61tov8node1
                    -profileName v61tov8node1
                    -oldProfile 61node1
                    -replacePorts TRUE
                    -backupConfig TRUE
                    -includeApps TRUE
                    -scriptCompatibility TRUE
                    -username myuser
                    -password mypass

  11. Check the WASPostUpgrade console output for the following messages. You might find the following messages: "Failed with errors" or "Completed with warnings". Also, look in the following log files for errors or warning messages:

    • mybackup_new_host/v61tov8node1/logs/WASPostMigrationSummary.log
    • mybackup_new_host/v61tov8node1/logs/WASPostUpgrade.target_profile.timestamp.log
    • mybackup_new_host/v61tov8node1/logs/WASPostUpgrade.target_profile name.trace

    If the WASPostUpgrade command fails, you might need to restore the v8.0 dmgr from the backup configuration file. If the WASPostUpgrade command processing ran the syncNode command, then the dmgr is aware that the node X has been migrated. The node X cannot be migrated again until the dmgr has been restored to the state before the node X migration.

  12. Check the v8.0 dmgr SystemOut.log file for error or warning messages.

  13. Start the migrated v8.0 node X agent.

  14. Check the v8.0 dmgr and node X SystemOut.log for error or warning messages.

  15. Synchronize the cell if the auto synchronization process is not enabled.

  16. Start the appropriate application servers on v8.0 migrated node X.

  17. Run the backupConfig command and save the v8.0 profile configuration to a file. For example: 
    cd  v8_PROFILE_ROOT/v61tov8node1/bin/
./backupConfig.sh /mybackup_new_host/v61tov8node1.zip
                  -username myuser
                  -password mypass
                  -nostop

    Each time you run the backupConfig command on a specific node, use a new backup file name.

  18. Save the dmgr configuration using the backupConfig command. On the v8.0 dmgr host... 
    cd v8_PROFILE_ROOT/v61v8dmgr01/bin/
./backupConfig.sh /mybackup_new_host/v61v8dmgr01backupMigratedDmgrPlusNodeX.zip
                  -username myuser
                  -password mypass

    For each node that you migrate, backup the v8.0 dmgr to a new backup file.

  19. Repeat the previous steps for additional nodes.


Support information (generally applicable)


What to do next

Troubleshoot




+

Search Tips   |   Advanced Search