Clone Tables #

Paimon supports cloning tables for data migration. Currently, only table files used by the latest snapshot will be cloned.

To clone a table, run the following command to submit a clone job. If the table you clone is not modified at the same time, it is recommended to submit a Flink batch job for better performance. However, if you want to clone the table while writing it at the same time, submit a Flink streaming job for automatic failure recovery.

<FLINK_HOME>/bin/flink run \
    /path/to/paimon-flink-action-0.9.0.jar \
    clone \
    --warehouse <source-warehouse-path> \
    [--database <source-database-name>] \
    [--table <source-table-name>] \
    [--catalog_conf <source-paimon-catalog-conf> [--catalog_conf <source-paimon-catalog-conf> ...]] \
    --target_warehouse <target-warehouse-path> \
    [--target_database <target-database>] \
    [--target_table <target-table-name>] \
    [--target_catalog_conf <target-paimon-catalog-conf> [--target_catalog_conf <target-paimon-catalog-conf> ...]]
    [--parallelism <parallelism>]

If database is not specified, all tables in all databases of the specified warehouse will be cloned.

If table is not specified, all tables of the specified database will be cloned.

Example: Clone test_db.test_table from source warehouse to target warehouse.

<FLINK_HOME>/bin/flink run \
    /path/to/paimon-flink-action-0.9.0.jar \
    clone \
    --warehouse s3:///path/to/warehouse_source \
    --database test_db \
    --table test_table \
    --catalog_conf s3.endpoint=https://****.com \
    --catalog_conf s3.access-key=***** \
    --catalog_conf s3.secret-key=***** \
    --target_warehouse s3:///path/to/warehouse_target \
    --target_database test_db \
    --target_table test_table \
    --target_catalog_conf s3.endpoint=https://****.com \
    --target_catalog_conf s3.access-key=***** \
    --target_catalog_conf s3.secret-key=*****

For more usage of the clone action, see

<FLINK_HOME>/bin/flink run \
    /path/to/paimon-flink-action-0.9.0.jar \
    clone --help