Renaming replicated table

Learn how to rename a Kudu table involved in a replication job by performing coordinated changes on both the source and sink clusters.

Renaming a replicated table requires coordinated changes on both clusters and a controlled job restart. You must stop the job cleanly by ensuring no writes are in flight before you rename any table.

Follow the Kudu replication Pre-stop checklist to drain the pipeline.
Stop the replication job by using a savepoint.

You must confirm that the pendingCount, unassignedCount, and pendingRemovalCount metrics are all zero before you stop the job. This ensures the savepoint contains no in-flight scan splits.

Rename the source Kudu table by running the following command:

kudu table rename_table <source-master-addresses> old_name new_name

Rename the sink Kudu table.
If the job.tableSuffix property is configured, the sink table name follows the <tableName><suffix> format. You must rename it to new_name<suffix> to match the new source name.

Without a suffix:
```
kudu table rename_table <sink-master-addresses> old_name new_name
```
With the job.tableSuffix=_dr property (for example):
```
kudu table rename_table <sink-master-addresses> old_name_dr new_name_dr
```
Restart the replication job from the savepoint.

For more information, see Restarting the replication job and finding the job ID. You must pass the new table name by using the --job.tableName new_name argument. The job.tableSuffix value does not change.