| source_service | ApiServiceRef | | [optional] |
| source_path | str | The path to replicate. | [optional] |
| destination_path | str | The destination to replicate to. | [optional] |
| mapreduce_service_name | str | The mapreduce service to use for the replication job. | [optional] |
| scheduler_pool_name | str | Name of the scheduler pool to use when submitting the MapReduce job. Currently supports the capacity and fair schedulers. The option is ignored if a different scheduler is configured. | [optional] |
| user_name | str | The user which will execute the MapReduce job. Required if running with Kerberos enabled. | [optional] |
| source_user | str | The user which will perform operations on source cluster. Required if running with Kerberos enabled. | [optional] |
| num_maps | int | The number of mappers to use for the mapreduce replication job. | [optional] |
| dry_run | bool | Whether to perform a dry run. Defaults to false. | [optional] |
| bandwidth_per_map | int | The maximum bandwidth (in MB) per mapper in the mapreduce replication job. | [optional] |
| abort_on_error | bool | Whether to abort on a replication failure. Defaults to false. | [optional] |
| remove_missing_files | bool | Whether to delete destination files that are missing in source. Defaults to false. | [optional] |
| preserve_replication_count | bool | Whether to preserve the HDFS replication count. Defaults to false. | [optional] |
| preserve_block_size | bool | Whether to preserve the HDFS block size. Defaults to false. | [optional] |
| preserve_permissions | bool | Whether to preserve the HDFS owner, group and permissions. Defaults to false. Starting from V10, it also preserves ACLs. Defaults to null (no preserve). ACLs is preserved if both clusters enable ACL support, and replication ignores any ACL related failures. | [optional] |
| log_path | str | The HDFS path where the replication log files should be written to. | [optional] |
| skip_checksum_checks | bool | Whether to skip checksum based file validation during replication. Defaults to false. | [optional] |
| skip_listing_checksum_checks | bool | Whether to skip checksum based file comparison during replication. Defaults to false. | [optional] |
| skip_trash | bool | Whether to permanently delete destination files that are missing in source. Defaults to null. | [optional] |
| replication_strategy | ReplicationStrategy | | [optional] |
| preserve_x_attrs | bool | Whether to preserve XAttrs, default to false This is introduced in V10. To preserve XAttrs, both CDH versions should be >= 5.2. Replication fails if either cluster does not support XAttrs. | [optional] |
| exclusion_filters | list[str] | Specify regular expression strings to match full paths of files and directories matching source paths and exclude them from the replication. Optional. Available since V11. | [optional] |
| raise_snapshot_diff_failures | bool | Flag indicating if failures during snapshotDiff should be ignored or not. When it is set to false then, replication will fallback to full copy listing in case of any error in snapshot diff handling and it will ignore snapshot delete/rename failures at the end of a replication. The flag is by default set to false in distcp tool which means it will ignore snapshot diff failures and mark replication as success for snapshot delete/rename failures. In UI, the flag is set to true by default when source CM Version is greater than 5.14. | [optional] |
| delete_latest_source_snapshot_on_job_failure | bool | A flag configuring distcp behaviour for the case when the distcp mapreduce job fails. This failure is sometimes due to some issues with the snapshot. By default, in case of mapreduce job failure the latest source side snapshot is deleted. Set this flag to false to make this replication not delete the last successfully replicated old snapshot of the source dataset in case of job failure. The default value of this flag is true. Not setting this flag is equivalent to the value true. | [optional] |
| num_fetch_threads | int | The number of threads to use for fetching the file statuses from HDFS during source file listing. The value 0 means don't use parallel fetching. The null value means to use distcp defaults. | [optional] |
| destination_cloud_account | str | The cloud account name which is used in direct hive cloud replication, if specified. | [optional] |