This is a python script that scans directories for Drupal sites and uses Drush to grab database backups for each site. It can also automatically git add/commit the changes.

The first time you run the script you should run it with the -h or --help option.

Requirements

• Drush 6.x
• Git
• Python (at least version 2.7)

Configuration

The "--structure-tables-key=common" option will be used for the sql-dump operation. This means you can create a drushrc.php file to exclude data from certain tables. See http://drush.ws/examples/example.drushrc.php for more information. I suggest doing this per-site (create sites/all/drush/drushrc.php).

If you don't want any tables to be skipped, either remove that line from my code, or empty the array in your drushrc.php file like this:

$options['structure-tables']['common'] = array();

If you want to exclude data from more than just the "common" tables (cache, watchdog, etc.), simply add table names to the array or create a new array and merge into the default.

Cronjobs

One of the main reasons to use a script like this is to have automatic, daily git commits of Drupal site database changes.

If you're running the script from a cronjob, make sure that Drush and Git are somewhere in the cron user's $PATH.

Here's an example cronjob (all on a single line):

• It runs every day at 02:15
• The output is emailed to agileadam@gmail.com
• The Drupal sites are located up to 1 level below /webapps/
• The backup files are saved to /db_backups/
• The --target-dir is a git repository
• The backed-up files will be automatically added / committed to git

15 2 * * * /usr/local/bin/python2.7 /usr/local/bin/drush-git-db-backup.py --commit --scan-dir=/webapps/ --target-dir=/db_backups/ -t 1 | mail -s "Drush DB Backup on server123" agileadam@gmail.com