The importance of making regular backups of your data cannot be over stated. Using the "Copy database" command from the File menu brings up a dialogue asking for a new database version. Type in a single character for the new version and press "ok" or return. If the new database already exists you will be asked whether you wish to overwrite it. Any subsequent changes you make will still be to the database that you originally opened, not to the database you have just saved to.
The database file may sometimes become fragmented. An option available when
saving is to use garbage collection. This creates the new database by only
copying over the used portions of data (and hence reduces fragmentation).
However it is quite a lot slower than the standard "Copy database" mechanism,
so if this causes problems add "
set_def COPY_DATABASE.COLLECT 0" to
your `.gaprc' file to change the default to no garbage collection. It
should be noted that garbage collection also performs a rigorous database
Do not always use the same version character for you backups. Instead keep several different backups. Otherwise you may find that both your current database and the backup has problems. It is also wise to run "check database" to verify data integrity. See section Check Database.
It is also possible to backup databases from outside
Gap by using
standard unix commands to copy both the record and index files.
Care should be taken when doing this to ensure that the database is not
being modified whilst copying. See your unix manuals for further details or
copy_db manual page (see section Copy_db) for
the external garbage collecting database copy program.