WAS v8.0 > Migration and coexistence > Distributed operating systems > Migrate product configurations > Rolling back environments
Rolling back a federated node
We can use the restoreConfig and wsadmin commands to roll back a migrated WAS v8.0 federated node to the state that it was in before migration. For each federated node to roll back, roll back the federated node itself and the corresponding changes made to the primary repository located on the dmgr.
Supported configurations:
This topic is about configuration migration, such as migrating dmgrs and federated nodes in a network deployment environment. The Application Migration Toolkit for WAS provides support for migrating applications from previous versions of WAS to the latest product version. For information about migrating applications, read more about the Application Migration Toolkit.
Best practice: When migrating a v6.x or 7.x federated node, the best practice is to perform the following actions to be able to roll it back to its previous state after migration:
- Back up your existing configuration.
- Run the backupConfig command or your own preferred
utility to back up the v8.0 dmgr configuration.
Make sure that you note the exact name and location of this backed-up configuration.
Read the "backupConfig command" article in the information center for more information.
- Run the backupConfig command or your own preferred
utility to back up the v6.x or 7.x federated node configuration.
Make sure that you note the exact name and location of this backed-up configuration.
Read the "backupConfig command" article in the information center for more information.
- Run the backupConfig command or your own preferred
utility to back up the v8.0 dmgr configuration.
- Migrate the federated node.
- If necessary, you can now roll back the federated node that you just migrated.
If you do not have a backup copy of your v8.0 dmgr configuration as it was before you migrated the v6.x or 7.x federated node to roll back, you cannot use the procedure described in this article and roll back your whole cell.
We must perform all of the backup and rollback actions for each migrated federated node before you proceed to migrate another federated node.
Procedure
- Run the backupConfig command or your own preferred utility to back up the v8.0 dmgr
configuration at the current state.
Make sure that you note the exact name and location of this backed-up configuration.
Read the "backupConfig command" article in the information center for more information.
- Stop all servers and the node agent on the v8.0 federated node that to roll back.
- Restore your previous configuration.
- Run the restoreConfig command or your own preferred utility to restore the previous v8.0 dmgr
configuration.
- Make sure that you restore the same backed-up configuration that you created just before you migrated the federated node.
- If we have made changes to the environment (application or configuration changes for example), these changes are rolled back at the same time
and cause the other nodes to force synchronization with the dmgr.
Read the "restoreConfig command" article in the information center for more information.
- Make sure that you restore the same backed-up configuration that you created just before you migrated the federated node.
- Perform one of the following actions to restore the Version 6.x or 7.x configuration for the federated node.
- Run the restoreConfig command or your own preferred
utility to restore the v6.x or 7.x configuration.
Make sure that you restore the same backed-up configuration that you created just before you migrated this federated node.
Read the "restoreConfig command" article in the information center for more information.
- Use the wsadmin command to run the migrationDisablementReversal.jacl script from the v6.x or 7.x PROFILE_ROOT/bin directory of the federated node.
In a Linux environment,
for example, use the following parameters:
./wsadmin.sh -f migrationDisablementReversal.jacl -conntype NONE
Tip: If we have trouble running the migrationDisablementReversal.jacl script, try to manually perform the steps in the script.
- Go to the following directory:
PROFILE_ROOT/config/cells/cell_name/nodes/node_name
where node_name is the name of the federated node to roll back. - If you see a serverindex.xml_disabled file in this directory, perform the following actions:
- Delete or rename the serverindex.xml file.
- Rename the serverindex.xml_disabled file to serverindex.xml.
- Go to the following directory:
- Run the restoreConfig command or your own preferred
utility to restore the v6.x or 7.x configuration.
- Run the restoreConfig command or your own preferred utility to restore the previous v8.0 dmgr
configuration.
- Start the v8.0 dmgr.
- Perform any application maintenance that is required.
- Synchronize the v6.x or 7.x federated node with the dmgr.
Read the "Synchronizing nodes with wsadmin.sh" article in the information center for more information.
- If you chose to keep the installed applications in the same location as the prior release during migration to v8.0 and any of the v8.0 applications are not compatible with the prior release, install applications that are compatible.
- Start the rolled-back v6.x or 7.x federated node and servers.
- Validate that the configuration is satisfactory.
This is the last chance to undo the rollback action by restoring the deployment-manager configuration that you backed up in the first step.
- Delete the v8.0 profile for the federated node that you rolled back to v6.x or 7.x.
Read the "Deleting a profile" article in the information center for more information.
Results
The configuration should now be returned to the state that it was in before migration.
What to do next
We can now restart the migration process if to do so.