Anyway, just as a reminder, here are the commands to edit these settings :

Change the language with

Change the default keyboard layout

That’s it, much more comfortable…

Teambox is an open sourced project collaboration software with simple task managements, file sharing and wiki collaborations for small teams. Teambox can be used in a hosted mode (Saas version with various plans depending on how many team members and projects you need to manage) on teambox.com or the code can be downloaded from GitHub and installed on one’s own server.

Here is the step by step installation of the latest version of Teambox on a server running Debian Lenny (upgrade to squeeze coming soon). My thanks to Frédéric de Villamil for getting this going since he is more of a Ruby Expert than I am.

Installing the dependances :

Downloading and installing Ruby Enterprise and Passenger:

Configuring Apache :

Add the following line in the file :

Setting up Teambox :

Create an Apache Virtual Host for teambox inspired by the following :

When all these steps are completed, the rest of the process is what’s covered in Teambox’ own documentation.

All you have left to do is configure your teambox and create your database :

• copy the database configuration file

• edit your database configuration in teambox/config/database.yml
• edit teambox main configuration file in teambox/config/teambox.yml
  Personally, I changed the domain and the email settings only

• create your database

Finally, test your app to see if it’s working by going to teambox.mydomain.com

That was it for me. Thanks again to Fred for helping pointing me to Ruby Enterprise.

I hope this helps someone else out there…

If you don’t already have it, download and install Virtual Box. Launch it

You start by creating a New Virtual Machine by clicking New. That will launch the wizard shown below:

Click Next

Mint isn’t an option so I just picked Linux 2.6 here

I set the memory to 512Mb but that can be changed later in the settings

Select the creation of a new hard disk

That will prompt a new wizard for the hard disk creation. I named it, picked a location and set the size to 8Gb. That’s it, the virtual machine is created.

Now, back to the main menu (the first picture, way on top), this time, click on Settings.The trick to boot from the .iso file is to go to Storage, select the CDRom and mount the iso file…

Now, back to the main menu. Click Start to launch the virtual machine. It will boot from the iso file and launch Mint as a live distrib. From the desktop, you can click no the install Mint shortcut and follow the usual Linux installation steps (it’s pretty straightforward with Mint). Turn off the machine when asked to. Go back to the settings and unmount the iso file. Restart the machine. Welcome to Mint!

Here is a working example of an FTP upload within a bash script:

Extra tip: if you need to download more than one file, use mget instead of get. The same works for mput instead of put.

In order to enable that, you need to reconfigure exim (which is the MTA installed by default). Run dpkg-reconfigure exim4-config. Select the Internet mode and answer the few following questions about your domains (watch out for open relay issues). Exim will then restart.

You can test your configuration by running sendmail your-address. Type in the body of your message then type CTRL + D on the next blank line to send your message. Hopefully, it made it to your inbox.

To make sure everything is okay and your server has not become an open relay, have Online.Net test your mail server.

Easy enough.

• my website files
• my databases

Since my backup space can not be used with rsync, I used backup manager to automatically save my files.Dedibox has a great article online about how to set this up for any Unix Dedibox.

For the database backups, I downloaded and tweaked (not much) the script shared by William Durand.

Everything is called daily by cron and, depending on the day of week, a full or incremental archive is created. After having one of my blog hacked, I really had to get busy working on a backup solution for my data.

On Monday morning, I received an email from Google Webmaster Tools letting me know that my WordPress blog had been hacked and was temporarily blacklisted. Users accessing my blog using Google Chrome or Firefox were advised to stay away for safety. I appreciate Google’s notification of this problem as well as their concern (and Mozilla’s) to protect my visitors.

Why was my WordPress blog hacked ? I don’t believe it was individually targeted, I believe it was hacked because it was vulnerable. Two reasons :

1.  I hadn’t updated my blog to the latest version of WordPress. As a matter of fact, I was not even running WordPress 3.0 yet.
2. Also, I believe the file permissions I was required to use on the server by my host provider were not safe enough thought I could not change them without breaking the whole blog (Access Denied error). I largely blame my hosting provider architecture for that.
So, what did the malware look like ? It was a php code which was added to every .php file in my main wordpress folder and read something like eval(base64_decode(XXXXXXXXXXX)). Once this was decoded and evaluated, it would check whether the referrer was Bing, Yahoo or Google and if so, redirect my visitor to a malware sharing site of some kind (I didn’t go to it…). This hack has been used recently on several GoDaddy hosted wordpress blogs and other CMS. With some help from @fdevillamil and Google, I found information about the problem and started working on it.

Finally, I finally got the problem sorted thanks to several blogs posts I’d like to point to here.

1. Here is a detailed explanation of the attack
2. I downloaded and modified Peter’s fixfiles.php script in order to clean my PHP code of the malicious PHP script.
3. Sucuri Research offers a console command (requires SSH access to your server) to clean your WordPress, too. They also detailed how to fix a hacked blog intended for GoDaddy users.
4. Before resubmitting my blog for validation by Google, I made sure everything was okay using Sucuri’s malware scanner service (free for limited use) . A very useful tool.
5. I quickly searched for similar tools I could use in the future and found the following articles though I didn’t test these services yet.
6. I moved my blog to another host where I was able to set the permission more restrictively
7. I disabled several of my plugins and updated my WordPress hoping to secure my blog from too many security flaws that could lay in third party code
8. Next, I read up about a few ways to fix security holes in WordPress in these posts :

• Monitor Malicious File Changes in WordPress Blog
• Secure a WordPress Blog
• The best way to handle file permissions for Apache www-data (ServerFault discussion)
• Avoid users uploading malware

9. Finally, I updated my blog which was no luxury. It can quickly be done in a few steps which are detailed right here

Hope this helps out someone else in a similar situation and that Google reconsiders my website soon.

There are a few things I always do when setting up a new system running Debian: I install the bash-completion, I set up my terminal colors, enable coloration in vi/vim and other little improvements.

In order to make my next setup much easier and save some time, I gather the necessary steps in this post.

Installing linuxlogo

I like to be greeted by my computer. After all, I spend so much time with him everyday. So, I always install linuxlogo.

Installing bash-completion

The bash-completion package enables a much more advanced completion (still using tab) which will not only complete on apt to make it aptitude but on aptitude ins to make it aptitude install. I got used to it, I don’t want to live without it.

Wait, it doesn’t work yet, you need to enable this in your profile. Keep on reading…

Setting up my bash profile

I don’t do much here (share your tips in the comment) but activate the bash-completion, a connection logo and my prompt colors. There are two files to edit : .bash_profile and .bashrc (if they don’t exist, create them).

In .bash_profile, I add:

In .bashrc, I edit the prompt:

and in .bashrc, I add:

Once you edited one of these files, if you want to see the effects of your changes, run source<filename>. At this point, if you installed the bash-completion, it should work.

Setting up root’s bash profile

I don’y use root often but to run aptitude upgrade or such things but one thing I like, when I am root, is to be immediately aware of the danger. That’s why I make the prompt red in that case. To do that, I edit /root/.bashrc

In the same file, I also add the completion, a few aliases and I send myself an email every time root is used as a security mesure (if someone gets root access I’ll know about it)

Syntax coloration in vi/vim

Another thing I got used to and would like to see by default, the syntax coloration in vi/vim. To enjoy it, install vim

For fun, reopen that same file and this time, it will have its colors. Awesome!

Auto login with SSH using a private key (from iTerm)

This is so convenient, it must be setup (if your security requirements allow it). I always follow this tutorial by David Winter in order to set this up. Nothing complicated, just head over there and follow the few steps.

This would not be complete if I didn’t assign the SSH login a bookmark in iTerm 

Press Shift + Command + B to manage your iTerm bookmarks
Create a new bookmark, give it the name of your remote host and enter the command ssh xxx.xxx.xxx.xxx
Choose a display profile and background as well as a shortcut letter
Save your bookmark and type the key combination, you are logged in on your remote host!

Anything else?

I do not spend so much time playing with Debian. Truthfully, I setup my server, once my projects run, I’m good. The rest of the time, I use Mac OS X. Still, I know there must be a lot many more comfortable tweaks available. Please, share your favorite in the comments. If you feel like reading on, here is a forum post about customizing .bashrc and several good examples to get inspiration.

I assume that, at this point, SSH server and Apache are already installed and running. Let’s start :

1. Install the SVN packages
   aptitude install subversion
   aptitude install libapache2-svn
2. Create a repository for your project(s)
   mkdir /var/subversion
   svnadmin create –fs-type fsfs /var/subversion/project1
3. Change the repository rights so that Apache can access/write in it
   chown -R www-data:www-data /var/subversion
   chmod -R 770 /var/subversion
4. Enable the SVN module for Apache
   a2enmod dav
   a2enmod dav_svn
5. Create the accounts for your SVN user(s) and enter the password (twice) when prompted
   htpasswd -c /etc/apache2/svn.passwd user1
6. Create the VirtualHost used to access your subversion ie subversion.yourdomain.com containing the following code
   <Location /project1>
   DAV svn
   SVNPath /var/subversion/project1
   AuthType Basic
   AuthName “Subversion Repository”
   AuthUserFile /etc/apache2/svn.passwd
   Require valid-user
   </Location>
7. Restart Apache with
   apachectl configtest
   apachectl restart
8. Browse to your repository (you should be at revision 0)

   http://subversion.yourdomain.com/project1

9. Import files in project 1
   cd /var/www/project1
   svn import -m “importing project1 to svn” http:/subversion.yourdomain.com/project1
10. Refresh your browser window, your files should be there and you should be at revision 1

If you have been migrating a Subversion repository using this tutorial, you can use the following command to relocate your local copy

svn switch –relocate http://old_url/subversion-repository http://shiny_new_url/subversion-repository

That’s it. Everything works, next, if you so desire, you can enable SSL for your repository access

1. Install the SSL packages
   aptitude install openssl
   aptitude install ssl-cert
2. Generate your certificate with the following command (and answer the questions that follow)
   openssl req  -new -x509 -days 365 -nodes -out /etc/apache2/apache.pem -keyout /etc/apache2/apache.pem
3. Enable SSL in the Apache configuration
   Listen 443
   NameVirtualHost *:443
4. Edit the subversion virtual host so that
   it listens on port 443 <NameVirtualHost *:443>
   add SSLEngine on
   and SSLCertificateFile /etc/apache2/apache.pem
   in the Location node, add SSLRequireSSL
5. Activate the SSL module
   a2enmod ssl
6. Again, restart Apache
   apachectl configtest
   apachectl restart

That much is done, too!

One last – optional – step : specify different access level to different users

1. Create your subversion users using
   htpasswd /etc/apache2/svn.passwd user2
   htpasswd /etc/apache2/svn.passwd user3
2. Edit your VirtualHost for subversion.yourdomain.com and add
   AuthzSVNAccessFile /etc/apache2/svnaccess.authz
3. Edit (or create) the file /etc/apache2/svnaccess.authz using the example below
   [groups]
   group1 = user1, user2
   [/]
   @group1 = rw
   [/project2]
   user3 = rw
4. Restart Apache and test your access restrictions
   apachetcl configtest
   apachectl restart

That’all. Happy New Year!

• add the line below to /etc/apt/sources.list

  1	deb http://www.backports.org/debian lenny-backports main contrib non-free

• get the public key from lenny-backports running the following command

  1	wget -O - http://backports.org/debian/archive.key | apt-key add -

• run aptitude update
• install subversion aptitude -t lenny-backports install subversion