sqadm pump(1)

sqadm pump - Squore data import

sqadm pump [-S dir] [-W dir] [--clob-size n] [--rows n] [--jdbc-url url] [--jdbc-user user] dump

This command restores a Squore PostgreSQL database backup into an Oracle schema.

It parses a PostgreSQL dump file (binary or plain text format, optionally gzipped), generates input files suitable for the sqlldr Oracle tool, and iterates on all tables.

The sqlldr shall be properly accessible from the PATH environment variable.

Temporary data files are generated in the current working directory (unless overridden by the --working-dir option), so make sure there is enough space on the disk partition. Table data is processed by chunks, whose sizes are controlled by the --rows option (unlimited by default), with impacts on both disk and memory usage.

The sqadm pump command shall be executed on a stopped Squore server, but a running database.

-S, --sql-dir dir

Directory with SQL scripts to create schema.

-W, --working-dir dir

Directory to store temporary (and large) files. The current working directory is used if not supplied.

--clob-size n

Buffer size for CLOB columns, in bytes.

The default value is 1048576 bytes (1 MB). The sqlldr utility will fail if a value larger than this limit is found in the data file.

--rows n

Number of rows to process at once.

Large tables should be split with the --rows option. When not supplied, a table is processed in one block and requires a lot of memory (sqlldr issue).

--jdbc-url url

Database url of the form jdbc:subprotocol:subname.

--jdbc-user user

Database username.

dump

The PostgreSQL dump file, in binary or plain text format.