Back up tables

You can use the KuduBackup Spark job to backup one or more Kudu tables.g

When you first run the job for a table, a full backup is run. Additional runs will perform incremental backups which will only contain the rows that have changed since the initial full backup. A new set of full backups can be forced at anytime by passing the --forceFull flag to the backup job.

Following are some of the common flags that you can use while taking a backup:
  • --kuduMasterAddresses: This is used to specify a comma-separated addresses of Kudu masters. The default value is localhost.
  • --rootPath: The root path is used to output backup data. It accepts any Spark-compatible path.
The following code run a KuduBackup job which backs up the tables foo and bar to an HDFS directory:
spark-submit --class org.apache.kudu.backup.KuduBackup [***FULL PATH TO kudu-backup2_2.11-1.12.0.jar***] \
  --kuduMasterAddresses [***KUDU MASTER HOSTNAME 1***]:7051,[***KUDU MASTER HOSTNAME 2***]:7051 \
  --rootPath hdfs:///[***DIRECTORY TO USE FOR BACKUP***] \
   impala::[***DATABASE NAME***].foo impala::[***DATABASE NAME***].bar