In this post I am going to show how to extend this setup with Dropbox to make it a full fledged remote backup solution… and it’s free! (at least for the first 2 GB )

1. Obtaining an account from Dropbox

You might want to create a Dropbox account only for taking backups if you don’t want all your documents and photos synchronized to your Linux server.

Go to www.dropbox.com to create your account.

2. Dropbox installation

The installation instructions below are a slightly modified version from the Dropbox community’s to make it as easy as possible.

Start by logging in to your Linux server as the user you want to assign Dropbox to. In this example we will use root:

$ sudo su

Change to your home directory:

$ cd ~

Download Dropbox.

Stable 32-bit:

$ wget -O dropbox.tar.gz "http://www.dropbox.com/download/?plat=lnx.x86"

Or stable 64-bit:

$ wget -O dropbox.tar.gz "http://www.dropbox.com/download/?plat=lnx.x86_64"

Extract:

$ tar -xvzf dropbox.tar.gz

It will extract to .dropbox-dist.

Run Dropbox:

$ ~/.dropbox-dist/dropbox

You should see output like this:

This client is not linked to any account...
Please visit https://www.dropbox.com/cli_link?host_id=7d44a557aa58f285f2da0x67334d02c1 to link this machine.

Go to the URL given; you should see a success message at the top of your screen.
Important: Dropbox will create a ~/Dropbox folder and start synchronizing when you do this. Make sure you’ve logged in to the correct Dropbox account at www.dropbox.com before going to the URL.

Exit Dropbox by pressing CTRL+D.

3. Installing Dropbox as a service

The following is a modified single-user version of Drazenko D.’s Dropbox daemon script.

Start up your favorite editor, creating /etc/init.d/dropbox:

$ nano /etc/init.d/dropbox

Insert the following script:

start() {
    echo "Starting dropbox..."
    start-stop-daemon -b -o -c root -S -x /root/.dropbox-dist/dropbox
}

stop() {
    echo "Stopping dropbox..."
    start-stop-daemon -o -c root -K -x /root/.dropbox-dist/dropbox
}

status() {
        dbpid=$(pgrep -u root dropbox)
        if [ -z $dbpid ] ; then
            echo "dropbox not running."
        else
            echo "dropbox running."
        fi
}

case "$1" in
  start)
    start
    ;;

  stop)
    stop
    ;;

  restart|reload|force-reload)
    stop
    start
    ;;

  status)
    status
    ;;

  *)
    echo "Usage: /etc/init.d/dropbox {start|stop|reload|force-reload|restart|status}"
    exit 1

esac

exit 0

And save and exit the editor.

Set up execute permissions for the script:

$ chmod +x /etc/init.d/dropbox

Set the script to load at startup:

$ update-rc.d dropbox defaults

Run the script to start Dropbox:

$ /etc/init.d/dropbox start

Make sure Dropbox is running:

$ /etc/init.d/dropbox status

And you’re good to go . Dropbox will now run as a background service when you start your server.

4. Backing up to Dropbox

After installing Dropbox, you can use the backup script from my previous post and backup to the Dropbox instead. Like this:

$ /var/scripts/backup.sh -d ~/Dropbox/backup/lassebunk/daily -s lassebunk -m lassebunk

Or, you can manually backup files by copying them to the Dropbox folder:

$ cp myveryimportantfile.tar.gz ~/Dropbox

Conclusion

I hope you found this post helpful when creating your own remote backup solution. If you did, please let me know in the comments how you use it .

Follow me on Twitter

• Backup my websites
• Backup my MySQL databases
• Backup my PostgreSQL databases
• Delete old backups to preserve disk space

I’ll split the script into pieces to explain each part – you can download the whole script at the bottom of this post.

First we get the arguments:

• -d to specify destination
• -s to specify website to backup
• -m to specify MySQL database to backup
• -p to specify PostgreSQL database to backup

# get arguments
while [ "$1" != "" ]; do
    case $1 in
        -d | --destination )    shift
                                dest=$1
                                ;;
        -s | --site )           shift
                                site=$1
                                ;;
        -m | --mysql )          shift
                                mysqldb=$1
                                ;;
        -p | --postgres )       shift
                                postgresdb=$1
                                ;;
        * )                     echo "Unknown argument $1."
                                exit 1
    esac
    shift
done

Complain if no destination is specified:

# complain if no destination specified
if [ "$dest" == "" ]; then
  echo "Missing destination."
  exit 1
fi

Set and create backup folder. It creates folders in the destination/yyyymmddhhmmss format:

# set and create backup folder
folder="$dest/`date +%Y%m%d%H%M%S`"
mkdir -p $folder

Backup website if specified. I keep all my websites in /var/www:

# backup website
if [ "$site" != "" ]; then
  echo "Backing up site $site to $folder/www.tar.gz..."
  cd /var/www
  tar -czf $folder/www.tar.gz $site
fi

Backup MySQL database if specified. It assumes that the database has a user with the same name, and logs in with this user:

# backup mysql database
if [ "$mysqldb" != "" ]; then
  echo "Backing up database $mysqldb to $folder/db.sql.gz..."
  mysqldump -u $mysqldb $mysqldb | gzip > $folder/db.sql.gz
fi

Backup PostgreSQL database if specified. Like with MySQL, it assumes that the database has a user with the same name, and logs in with this user:

# backup postgres database
if [ "$postgresdb" != "" ]; then
  echo "Backing up database $postgresdb to $folder/db.sql.gz..."
  pg_dump -U $postgresdb $postgresdb | gzip > $folder/db.sql.gz
fi

It finally deletes old backups, keeping the last 5:

# delete old backups keeping 5
echo "Deleting old backups..."
cd $dest
ls -t | sed 1,5d | while read folder; do rm -r $folder; done

Done:

# done
echo "Done."

My crontab -e is then:

# m h  dom mon dow   command

# lassebunk
0 2 * * * /var/scripts/backup.sh -d /var/backups/sites/lassebunk/daily -s lassebunk -m lassebunk
0 2 1 * * /var/scripts/backup.sh -d /var/backups/sites/lassebunk/monthly -s lassebunk -m lassebunk

# othersite
0 2 * * * /var/scripts/backup.sh -d /var/backups/sites/othersite/daily -s othersite -p othersite
0 2 1 * * /var/scripts/backup.sh -d /var/backups/sites/othersite/monthly -s othersite -p othersite

Please also see my follow-up on this post: Using Linux and Dropbox as a remote backup solution

I hope you found it useful – if you have any comments or suggestions, please write a comment below.

Download the script

Follow me on Twitter

21:03:57 up 365 days, 21:53,  2 users,  load average: 0.00, 0.00, 0.00

I originally used this server as a staging server but later on cancelled my Slicehost virtual server and began using my own server as a production server. It now hosts lassebunk.dk, tweetmycam.com, and userdriven.dk.

Follow me on Twitter

I bought my MacBook on May 1th after viewing a screencast where a guy sets up a blog in 15 minutes using Ruby on Rails. I immediately said to my stepdad: “I’m going to buy a Mac and learn Ruby on Rails!”

And so I did – the day after, I bought a Mac.

I’ve always been a dedicated user of Microsoft‘s products. First C64, then Amiga, then Mac, then PC. I liked the way everything was tested and came from one place, unlike open source. I’ve always been saying stuff like “I like to pay for my software because then I’m be sure about the quality.” – but in reality, everyone who uses Windows and other Microsoft products know that this isn’t always the case.

So I bought the Mac, and the first thing that surprised me, was how much of day-to-day work was done in the Terminal, or command line. When I installed Ruby on Rails, it was via command line; when I installed plugins, it was via command line. Evererything command line.

Over the next few days I began to get a hang of it. I bought a couple of domain names and signed up for a slice at Slicehost – because I had read some job description that said that you should’ve at least tried to use a whole day of setting a slice.

Coming from Windows, Linux is a whole other deal to setup. I used a lot of the Slicehost articles that guides you through the whole process from setup and security to getting your slice to run as a webserver.

In the beginning I was a little nervous about all the command lines, if they would really work and so one. But the more you try it, the more calm you get. It just works! And lucky me there was a lot of helpful articles about Unix and Linux commands out there (just search for the command on Google).

Since starting out on the Mac, I’ve learned a multitude of things:

• Setting up an Ubuntu slice at Slicehost.
• Setting up a web server.
• Setting up Ubuntu Netbook Remix on my netbook.
• Setting up Ubuntu server at home, also on my netbook – which now functions as my testing server.
• Programming a little PHP – see Tweet My Cam.
• Programming Ruby on Rails.
• Tweaking some Java code, compiling and signing a java applet in one day – see a demo here.
• Download and compile software (configure, make, make install).
• Submitted a patch to a Ruby on Rails plugin on Github.
• Setting up WordPress and WordPress Mu blogs and making my own theme – see my private blog and private work blog.
• Setting up a Subversion repository at XP-Dev.com.
• Using the Subversion repository
• Loving Textmate as my favorite editor – also over Visual Studio.
• Plus a bunch of other things.

In short I’ve learned so much about the open source world that just wasn’t that easy on the Microsoft platform.

I still use Microsoft Windows and other products, but now it’s through VMware Fusion on the Mac.

I’m happy about the Mac because OS X is very unobtrusive, fast operating system and what it does, it does very good. But at the same time, I also want a netbook that’s easier to carry, so I may end up running both systems for different purposes (unless I’m just installing Ubuntu on the netbook too ).

Hope you found this post interesting – I wrote it to tell about a beautiful (yak, I know) progress from Windows to Mac and Linux. Thumbs up if this has made you want to try it too. And please tell me in the comments.

Follow me on Twitter