So far, data is being written to the archive all of the time and no attention has been paid to cleaning out the archive again to free up space in the filesystem. PostgreSQL cannot do this job for us because it has no idea whether we want to use the archive again. Therefore, we are in charge of cleaning up the transaction log. Of course, we can also use a backup tool—however, it is important to know that PostgreSQL has no chance of doing the cleanup for us.

Suppose we want to clean up an old transaction log that is not needed anymore. Maybe we want to keep several base backups around and clean out all transaction logs that won't be needed anymore to restore one of those backups.

In this case, the pg_archivecleanup command-line tool is exactly what we need. We can simply pass the archive directory and the name of the backup file to the pg_archivecleanup command, and it will make sure that files are removed from disk. Using this tool makes life easier for us because we don't have to figure out which transaction log files to keep on our own. Here's how it works:

pg_archivecleanup removes older WAL files from PostgreSQL archives.
Usage:
  pg_archivecleanup [OPTION]... ARCHIVELOCATION OLDESTKEPTWALFILE
Options:
  -d             generate debug output (verbose mode)
  -n             dry run, show the names of the files that would be removed
  -V, --version  output version information, then exit
  -x EXT         clean up files if they have this extension
  -?, --help     show this help, then exit
For use as archive_cleanup_command in postgresql.conf:
  archive_cleanup_command = 'pg_archivecleanup [OPTION]... ARCHIVELOCATION %r'
e.g.
  archive_cleanup_command = 'pg_archivecleanup /mnt/server/archiverdir %r'
Or for use as a standalone archive cleaner:
e.g.
  pg_archivecleanup /mnt/server/archiverdir 000000010000000000000010.00000020.backup

Report bugs to <[email protected]>.

This tool can be used with ease and is available on all platforms.

Now that we've taken a look at transaction log archiving and PITR, we can focus our attention on one of the most widely used features in the PostgreSQL world today: streaming replication.