Preparing to Join an Active Cluster
To expedite the join process, make sure that the TDV Server that is to join the cluster contains minimal metadata beforehand. Because all existing resources on the joining node are overwritten with the cluster data at join time, it is best to prepare the joining node with a “clean” repository. For example, the repository has only what is present the TDV installer is run.
If the joining node has resources that you want to save and potentially restore, be sure to follow the process prior to joining the cluster.
To prepare a TDV Server with existing resources prior to joining a cluster
1. Run a backup_export of the system, and validate the CAR file by importing it into another system.
Validation is important, so you are sure to have a full backup file in case you want to restore data later.
2. Use the repo_util.sh -DropSchema command.
3. Restart the server to recreate the schema automatically.
4. Run backup_import of the CAR file that was previously been exported, if you want to restore the original data or resources.
See the TDV Administration Guide for more information about using the backup_export, backup_import, and repo_util utilities.