Make a directory ( it can anywhere ) called baks/mysql

mkdir -p /baks/mysql

Create a file (it can be anywhere) called /root/mysql_backups.sh and put this script in it

#!/bin/bash

# modify the following to suit your environment
export DB_BACKUP="/baks/mysql"
export DB_USER="root"
export DB_PASSWD="your-mysql-password-goes-here"

# title and version
echo ""
echo "Backup and rotate all mysql databases"
echo "--------------------------"

rm -rf $DB_BACKUP/04
mv $DB_BACKUP/03 $DB_BACKUP/04
mv $DB_BACKUP/02 $DB_BACKUP/03
mv $DB_BACKUP/01 $DB_BACKUP/02
mkdir $DB_BACKUP/01

echo "* Creating backup..."
mysqldump --user=$DB_USER --password=$DB_PASSWD --all-databases | bzip2 > $DB_BACKUP/01/mysql-`date +%Y-%m-%d`.bz2
echo "----------------------"
echo "Done"
exit 0

Install it via cron and have it run at 3:10 am every morning.

crontab -e

10 3 * * * /root/mysql_backups.sh  > /baks/status.log

This script will save the last 4 days of data.

If using Sphinx, you need to refresh indexes when you add new content to your database. This is fairly easy to do by hand

rake thinking_sphinx:index RAILS_ENV=production

But if you want to automate this and use a cron, remember to set the PATH, SHELL and RAILS_ENV variables for your job. The environment isn’t the same when you’re doing it by hand and your index will fail silently :(

You find your specific PATH and SHELL info like so

echo $PATH
echo $SHELL

To get into the cron and set your schedule

crontab -e

PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
SHELL=/bin/bash
RAILS_ENV=production

# re-index production sphinx every 15 minutes
*/15 * * * * root cd /var/www/rails_app && /usr/local/bin/rake thinking_sphinx:index >> /dev/null 2>&1

More information available here

http://heimdull.blogspot.com/2009/05/journey-with-thinking-sphinx-and-crond.html

http://groups.google.com/group/thinking-sphinx/browse_thread/thread/5451458fae7d6124

First off you’ll need to edit your cron file. Normally, the cron files are kept under /etc/cron.daily or /etc/cron.hourly but we can just use the command line tool, crontab and pass it the -e flag, so that we can edit the file without any fuss.

sudo crontab -e

If this is the first cron you’re installing you should see a blank file or something that looks like this

# m h  dom mon dow   command

Essentially this is telling you the order of the arguments you need to specify. Each asterisk represents an interval of time. In this order they are

Minute, Hour, Day of Month, Month, Day of Week and then the path to your command. So to set up our cron to run once every five minutes we’ll write

*/5 * * * * /path/to/my/script

It’s important to note that it is */5 and not just 5. The interval for minute is 0 – 59, meaning that if you just enter a 5, you’ll run your cron job only once an hour but at the 5th minute of that hour. To specify that we want to run it every 5 minutes we need the */5.  Also, note that the cron will execute your command via the path. If you need an interpreter like ruby, rake or php, don’t forget to put that as a part of your command. For instance this will set a rake task to run every five minutes on a Ubuntu box.

*/5 * * * * cd /var/www/my_ror_application  && /usr/bin/rake RAILS_ENV=development db:migrate

You can see that my command uses cd, changing directories to my application. I then specify the full path to the rake program, /usr/bin/rake and give it arguments, including the environment and then of course, the task I want to execute. In this case I’m migrating the database. Which is obviously pointless. Maybe you’ll want to email forum news or send out activation emails.

For more information here is a great reference mkaz.com