Command-line options

OBLOADER allows you to import required data by specifying command-line options.

Sample command

The following code provides an example on how to import data. For more information, see the following Options table and the Scenarios topic.

If you connect OBLOADER to an OceanBase cluster through OceanBase DataBase Proxy (OBProxy), use the -h and -P options to specify the IP address and port number of the OBProxy host. Then, use the -c option to specify the name of the cluster to be connected to. If you directly connect OBLOADER to an OBServer, you do not need to use the -c option to specify the name of the cluster to be connected to.

[admin@localhost]>./obloader  -h <host IP address> -P <port number> -u <username> -p <password>  --sys-password <password of a user under the sys tenant> -c <cluster> -t <tenant name> -D <--database> [--ddl] [--csv|--sql] [--all|--table 'table name'] -f<data file or directory>

Options

Option	Required	Description
-h (--host)	Yes	The IP address of the OBProxy or OBServer to be connected to.
-P (--port)	Yes	The port number of the OBProxy or OBServer to be connected to.
-c (--cluster)	No	The cluster name of the cluster. You do not need to specify this option if OBDUMPER is directly connected to an OBServer.
-t (--tenant)	No	The tenant name of the user.
-u (--user)	Yes	The username that you use to log on to the database.
-p (--password)	No	The password that you use to log on to the database.
-D (--database)	No	The name of the database.
-f (--file-path)	Yes	The directory that stores the data file or the absolute path of the data file.
--sys-user	No	The username of a user under the sys tenant. Default value: root@sys.
--sys-password	No	The password of the user specified by the --sys-user option. By default, this option is left empty.
--public-cloud	No	Uses the limited mode to import data. You do not need to specify the --sys-user option in limited mode.
--file-suffix	No	The extension of the data file. Do not confuse a file extension and a file format. You can use regular expressions.
--file-encoding	No	The file encoding format. Default value: UTF-8.
--ctl-path	No	The directory where the control file is stored. The extension of the control file is updated to .ctrl from .ctl.
--map-path	No	The directory where the file that stores data field mappings is stored.
--log-path	No	The directory to which log files are exported.
--ddl	No	Imports files in the DDL format. Default file name extension: -schema.sql.
--sql	No	Imports files in the SQL format. Default file name extension: .sql.
--mix	No	Imports files that contain both schemas and data.
--csv	No	Imports CSV files. We recommend that you use this option. Default file name extension: .csv.
--pos	No	Imports files in the POS format. In these files, data is organized based on the byte offsets of a fixed length.
--cut	No	Imports files in the CUT format in which data is separated by several characters. Different from data in the CSV format, data in the CUT format is separated by strings.
--all	No	Imports the definitions of all database objects.
--table	No	The table to be exported. Separate multiple tables with commas (,). If you set this option to an asterisk (*), all tables in the database specified by the -D option are imported.
--view	No	The view to be imported. Separate multiple views with commas (,). If you set this option to an asterisk (*), all views in the database specified by the -D option are imported.
--replace-object	No	Replaces a table schema if the table schema already exists. We recommend that you do not use this option.
--rw	No	The ratio between the read threads and write threads. Default value: 0.2, which is 1 read thread to 5 write threads.
--slow	No	The memory usage threshold of the OBServer, exceeding which OBLOADER will slow down. Default value: 0.75.
--pause	No	The memory usage threshold of the OBServer, exceeding which OBLOADER will pause. Default value: 0.85.
--batch	No	The number of queries supported in a single transaction. Default value: 200.
--thread	No	The number of worker threads. Default value: CPU × 2.
--block-size	No	The size of a file block, in MB. Default value: 64.
--retry	No	Specifies whether to resume the import from the last saved point after the import is interrupted.
--external-data	No	Records data if it is exported by using a third-party tool. OBLOADER does not check the meta information about data exported by third-party tools.
--max-tps	No	The upper limit of the import speed in the unit of rows/second. Default value: Integer.MAX_VALUE.
--max-wait-timeout	No	The maximum amount of time OBLOADER waits for a major compaction to finish before it times out, in hours. Default value: 3.
--trail-delimiter	No	Deletes the last column delimiter if the data ends with column delimiters.
--with-trim	No	Truncates leading and trailing spaces of the data.
--skip-header	No	Specifies whether to skip the first row in a CSV file. By default, the first row is not skipped.
--null-string	No	The string to replace NULL. Default value: \N. If the value is not \N after the data is imported, the original value is a NULL string.
--empty-string	No	The escape character for an empty string (' '). Default value: \E.
--line-separator	No	The row delimiter. Default value: '\n'.
--column-separator	No	The column delimiter. Default value: ','.
--escape-character	No	The escape character. Default value: '\'.
--column-delimiter	No	The quotation marks. Default value: ' '.
--exclude-table	No	* If you specify a table name, the schema and data of the specified table are not exported. Example: --exclude-table 'a,b,c'. * If you do not specify a table, all table schemas and data are exported. Example: --exclude-table ''. * You can use asterisks (*) for fuzzy matching. For example, you can specify --exclude-table '*ab*' to exclude tables whose names contain ab.
--column-splitter	No	The column delimiter. Different from the CSV format where the delimiter is a single character, this option allows you to use a string, which can have several characters, to separate data.
--oss-endpoint	No	The endpoint of the region where a bucket resides. For example, for the China (Hangzhou) region, the value must be https://oss-cn-hangzhou.aliyuncs.com.
--access-key	No	The account used to access the Object Storage Service (OSS) bucket.
--secret-key	No	The secret key used to access the OSS bucket.
--bucket-uri	No	The URI of the bucket. URI format: oss://bucket/key1/key2 Example: oss:// oceanbasepublic/tools/obdumper.
--max-discards	No	Limits the number of discards generated by a single table. This option applies to a single object. For example, when you import Table A and Table B, if the discards generated by Table A reach the specified value of the max-discards option, OBLOADER stops the import of Table A and continues to import Table B. * --max-discards=0: OBLOADER sends an alert and stops the import once a discard is generated. * --max-discards=N: The N represents a positive integer. OBLOADER sends an alert and stops the import when N discards are generated. * --max-discards=-1: OBLOADER ignores the alert and continues the import.
--max-errors	No	The maximum number of errors that can be ignored. You can ignore all errors. By default, at most 10 errors can be ignored. If more than 10 errors occur for a thread, the program exits. This option applies to a single object. For example, when you import Table A and Table B, if the errors generated by Table A reach the specified value of the max-errors option, OBLOADER stops the import of Table A and continues to import Table B. * --max-errors=0: OBLOADER reports an error and stops the import once an error is generated. * --max-errors=N: The N represents a positive integer. OBLOADER reports an error and stops the import when N errors occur. * --max-errors=-1: OBLOADER ignores the errors and continues the import. Notice However, some errors cannot be ignored and must be eliminated before you can continue the import. For more information about errors that cannot be ignored, see FAQ about OBLOADER.
--replace-data	No	Supports the import of incremental data, and allows you to overwrite and insert data when a primary key or unique key conflict occurs. You can import files in the --sql, --csv, or --cut format. The --sql format is preferred.
--truncate-table	No	Specifies whether to clear tables before the import. We recommend that you do not use this option. Instead, we recommend that you manually clear tables. When an application starts, an error may be reported to indicate that the partition information cannot be found.