Data Pump - Use a Parameter File

Instead of typing commands on the command line, in many situations it’s preferable to store the commands in a file and then reference the file when executing Data Pump export or import. Using parameter files makes tasks more repeatable and less prone to errors. You can place the commands in a file once and then reference that file multiple times.

Additionally, some of the Data Pump commands (like FLASHBACK_TIME) require the use of quotation marks; in these situations, it’s sometimes hard to predict how the operating system will interpret the quotation marks. Whenever a command requires quotation marks, it’s highly preferable to use a parameter file.

To use a parameter file, first create an operating text file that contains the commands you want to use to control the behavior of your job. This example uses the Linux/Unix vi command to create a text file named imp.par:

$ vi imp.par

Now, place the following commands in the imp.par file:

userid=darl/foo
directory=dp
dumpfile=invp.dmp
logfile=invp.log
tables=f_sales

Next, reference the parameter file via the PARFILE command-line option:

$ impdp parfile=imp.par

Data Pump import processes the parameters in the file as if they were typed on the command line. If you find yourself repeatedly typing the same commands, then consider using a parameter file to increase your efficiency.