Sunday, December 18, 2011


So here we are again this time to talk about "Trac" an integrated SCM and project management tool. I love this tool even though at first glance it could result a bit simplistic. The philosophy is quite clear and if you make a good usage of the tool you will be implementing plenty of recognized good practices on the software engineering even without notice it. Here you can find the original site with a lot of information, tutorials, etc.. As usual, I will focus on this tutorial on how install it and make it works, but I do not discard a future where we will integrate our source code within Trac and create a bunch of online documentation for it.

I will explain two different approaches, the classical one when we have root privileges, and a very particular one by downloading the sources and installing it without root permissions.

A. Install Trac in the classical way.

1. Install easy_install tool (More information here):

sudo apt-get install python-setuptools

2. Install Genshi, a web-template based tool that Trac is based on: (More information here.)

sudo easy_install Genshi

3. Install the python connector that will allows Trac to communicate with a MySQL Database.

sudo apt-get install python-mysqldb

4. Install Babel, the internationalization engine that Trac is based on:

sudo easy_install Babel==0.9.5

5. And finally... Install Trac!!

sudo easy_install Trac

6. If you want to use Trac with a MySQL Database, you will need to create a Database called trac and provide the needed privileges for the user 'tracuser' from MySQL monitor (More information about how to install and use MySQL on this previous blog):


GRANT ALL ON trac.* TO tracuser@localhost IDENTIFIED BY 'administrador';

If you are installing Trac in the same server that MySQL is installed, your connection query should be something like this:


If you are installing it in a different server you will have to change 'localhost' by the IP of the server were MySQL is installed

7. Create a directory where the configuration files of your Trac site will be stored e.g.  /var/trac/testsite

8. Execute initenv command indicating the directory created in the previous step (You will be asked for several paramenters, including  the connection query retrieved on the step 6) :

sudo trac-admin /var/trac/testsite initenv 

The main configuration field of your site is found here:


You can run a built-in server to start playing with your just installed tool:

sudo tracd --port 8000 /var/trac/testsite 

After doing that you should be able to browse your site:

Fig.1 Our first Trac Project

Finally I would like to mention a very interesting command that will provide the default user Administrator privileges over the whole site.

sudo trac-admin /var/trac/testsite permission add anonymous TRAC_ADMIN

B. Install Trac from the sources

1. Install Python

export PATH=$PATH:~/bin/Python-2.6.7/Lib
tar -xvf Python-2.6.7.tgz

2. Install easy_install tool

export PYTHONPATH=$PYTHONPATH:/home/javier/bin/EasyInstall
sh setuptools-0.6c11-py2.6.egg  --install-dir=~/bin/EasyInstall
export PATH=$PATH:~/bin/EasyInstall

3. Install Genshi

export PYTHONPATH=$PYTHONPATH:~/app/Genshi
easy_install --install-dir=~/app/Genshi/ Genshi

4. Install Babel

easy_install --install-dir=~/app/Babel/ Babel

5. Install Trac

easy_install --install-dir=~/app/Trac/ Trac
./trac-admin ~/tracsites/ initenv
./tracd --port 8000 ~/tracsites/ initenv

Monday, December 12, 2011

MySQL on the Cloud

On this post we are going to install and configure an Instance of MySQL on our remote server. Here you can find a post talking about this same topic.

First of all we have to install the MySQL Server packages:

sudo apt-get install mysql-server

During the installation you will be asked to introduce the root password.

In order to allow connection from any client, edit /etc/mysql/my.cnf and modify bind-address with the following value:

bind-address            =

We need to do the same at operative system level, edit /etc/hosts.allow and include the following line:

mysqld: all

We also need to set a final layer of security at application level. Introduce the following command and introduce the password that you set in the installation phase:

mysql -u root -p

Once you are inside the MySQL console type (The semicolon at the end is quite important, do not miss it):

Welcome to the MySQL monitor.  Commands end with ; or \g.
Your MySQL connection id is 51
Server version: 5.1.41-3ubuntu12.10 (Ubuntu)

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

mysql> GRANT ALL PRIVILEGES ON *.* TO 'root'@'%' IDENTIFIED BY 'root';

After entering the above command type quit to leave the console:

mysql> quit

Now you can connect from your local pc to your remote database on the cloud. There are several interesting tools to manage and use your MySQL database server, my favorites are MySQL Administrator and MySQL Server. You can install it:

sudo apt-get install mysql-client mysql-query-browser mysql-admin

Fig.1 MySQL Administrator login

Log in with your credentials and your cloud instance identifier and enjoy!!


Sunday, December 11, 2011

Mercurial with Apache... on the Cloud!!

And here we go with our third tutorial explaining how to deploy and enjoy applications 'In the cloud'. This time we will take advantage of our previous post where we configured Apache to serve securely and authenticated information across the Internet and we will deploy a distributed source code management system 'Mercurial'. You can find a lot of information about what is and what is  Mercurial used for in the official web page

A. Installing and configuring Apache with Mercurial

And here we go...

The first thing we have to do is install python (Mercurial is developed on python and mercurial packages):

sudo apt-get install python python-dev mercurial

The next step is create our repository folder, and give it the appropriated permissions  from where Apache will interact:

sudo mkdir -p /var/hg/repos
sudo  touch /var/hg/hgweb.config
sudo chown -R www-data:www-data /var/hg

Edit the file /var/hg/hgweb.config that you just created and include the following Mercurial configuration directives:

repos/ = repos/

style = gitweb

Now we are going to copy a little script that will allow to interact apache with our Mercurial repository and again we will set the correct permissions:

cd /var/hg/
sudo cp /usr/share/doc/mercurial/examples/hgwebdir.cgi ./
sudo chown www-data:www-data ./hgwebdir.cgi
sudo chmod +x ./hgwebdir.cgi

In the following step we are going to initialize a repository called 'myRepos', this is the general way to do it, so if in the future you need to initialize another repository you just need to execute this command again by changing the name of 'myRepos' by the new name.

sudo su - www-data -c "hg init /var/hg/repos/myRepos"

Edit your new repository config file /var/hg/repos/myRepos/.hg/hgrc to include information about itself and allow push from everywhere (push is mercurial terminology)

contact = Javier
description = My first Mercurial repository
allow_push = *

# Allow the download for the last snapshot in compacted format
allow_archive = gz zip bz2 

And last but not least we have to configure Apache to start serving from our repository, this is done by adding the following section on the /etc/apache2/sites-available/default-ssl:

AddHandler cgi-script .cgi
ScriptAliasMatch      ^/hg(.*) /var/hg/hgwebdir.cgi$1
<Directory "/var/hg">
  Options FollowSymLinks +ExecCGI
  AllowOverride All
  Order allow,deny
  Allow from all
  AuthType Basic
  AuthName "Password Required"
  AuthUserFile /etc/apache2/passwd/passwords
  Require valid-user

 You can place it right after the <Directory /var/www/> section so your config file will looks like:

<Directory /var/www/>
 Options Indexes FollowSymLinks MultiViews
 AllowOverride None
 Order allow,deny
 allow from all
 AuthType Basic
 AuthName "Password Required"
 AuthUserFile /etc/apache2/passwd/passwords
 Require valid-user
AddHandler cgi-script .cgi
ScriptAliasMatch      ^/hg(.*) /var/hg/hgwebdir.cgi$1
<Directory "/var/hg">
  Options FollowSymLinks +ExecCGI
  AllowOverride All
  Order allow,deny
  Allow from all
  AuthType Basic
  AuthName "Password Required"
  AuthUserFile /etc/apache2/passwd/passwords
  Require valid-user

Restart apache:

sudo apache2ctl restart

And that's all!! You can now access to your repository via web:

B. Using mercurial from our local machine

First of all you will need to install Mercurial in your local machine:

sudo apt-get install mercurial mercurial-common

Now that we have created a remote repository, we can clone it locally an perform our changes:

(From your local pc)

mkdir ~/MercurialRepositories

cd ~/MercurialRepositories

hg clone

You will receive a set of messages as following:

warning: certificate with fingerprint 6a:11:f3:d4:be:ed:54:f3:d3:78:ca:6b:03:58:55:b0:3b:25:44:58 not verified (check hostfingerprints or web.cacerts config setting)
http authorization required
realm: Password Required
user: javier
warning: certificate with fingerprint 6a:11:f3:d4:be:ed:54:f3:d3:78:ca:6b:03:58:55:b0:3b:25:44:58 not verified (check hostfingerprints or web.cacerts config setting)
destination directory: myRepos
warning: certificate with fingerprint 6a:11:f3:d4:be:ed:54:f3:d3:78:ca:6b:03:58:55:b0:3b:25:44:58 not verified (check hostfingerprints or web.cacerts config setting)
no changes found
updating to branch default
resolving manifests
0 files updated, 0 files merged, 0 files removed, 0 files unresolved

If you want to avoid all these annoying warnings just edit  the file ~/MercurialRepositories/myRepos/.hg/hgrc

And add the following lines:

[hostfingerprints] = 6a:11:f3:d4:be:ed:54:f3:d3:78:ca:6b:03:58:55:b0:3b:25:44:58

Where the value of the fingerprint is the one displayed on the warning



A. Some useful commands on mercurial:

hg sum
Shows state of the working directory (include info of the repository).

--remote (shows if there are incomings)

hg st
Shows differences of the working directory with the current working repository.

hg resolve -m <file>
Mark file as resolved (we have to make a commit afterwards)

hg incoming
Show if there is a new changeset coming from the central repository
-p Show the differences between files

hg outcoming
Show if there is a new changeset going to the central repository

Apache on the Cloud

I will explain step by step how to configure Apache within a Ubuntu server running in the Cloud. In my previous post I explained how to configure the Amazon Web Service in order to have an Ubuntu Server running 7x24. At the end of this tutorial we will have a public html page visible to everybody.

A. Connect into your remote instance and install Apache

ssh -i Ubuntu_10_04_32.pem

sudo apt-get update

sudo apt-get install apache2 apache2-utils apache2-threaded-dev

Belive it or not you just created and publish a web page on the Internet. Open your favorite web browser and paste on the url the identifier of your cloud instance (You can find it in your connector command, in my case

Fig.1 Our public page

B. Configure Apache for ssh connection and authentication

Ok, that was pretty impressive, but what happens if you want to transmit and receive all the information encrypted, and actually what if I do not want everybody accessing to my web page. All this can be achieved by configuring apache. We will explain here the steps needed on Ubuntu, but if you are seriously thinking on developing web applications you should get some background on Apache. Here you can find the official apache documentation.

On the Ubuntu Server 10.04 the version of apache installed by default is the 2.2

ubuntu@ip-10-48-98-200:~$ apache2ctl -v
Server version: Apache/2.2.14 (Ubuntu)
Server built:   Nov  3 2011 03:31:27

So the file you are looking for is httpd-docs-2.2.14.en.pdf

That being said here you can find a set of commands that you will use in your daily life with Apache

sudo update-rc.d -f apache2 remove --> Prevent Apache autostart
sudo update-rc.d apache2 defaults   -->  Restore Apache autostart

sudo apache2ctl -k start --> Starts apache
sudo apache2ctl -k stops --> Stop apache
sudo apache2ctl restart --> Restart apache (Needed when you change a configuration file to be took into account)

a2dismod/a2enmod --> Disable/Enable module

a2dissite/a2ensite --> Disable/Enable site

B.1 Configure Apache for ssh connection

Enable the Apache ssl module, the default ssl site and restart Apache:

sudo a2enmod ssl
sudo a2ensite default-ssl
apache2ctl restart

By doing this you should be able to access to your web site via https. By default you are using a Non verified certificate so Firefox will warn you about this. Just add the exception and continue.

If you want to use your own certificates just follow these steps:

Create the directory where you want to store the certificate:

sudo mkdir /etc/apache2/ssl

Generate the certificate (You will have to enter several information that will be attached within the certificate):

sudo openssl req -new -x509 -days 365 -nodes -out /etc/apache2/ssl/apache.pem -keyout /etc/apache2/ssl/apache.pem

Generating a 1024 bit RSA private key
writing new private key to '/etc/apache2/ssl/apache.pem'
You are about to be asked to enter information that will be incorporated
into your certificate request.
What you are about to enter is what is called a Distinguished Name or a DN.
There are quite a few fields but you can leave some blank
For some fields there will be a default value,
If you enter '.', the field will be left blank.
Country Name (2 letter code) [AU]:FR
State or Province Name (full name) [Some-State]:PACA
Locality Name (eg, city) []:Antibes
Organization Name (eg, company) [Internet Widgits Pty Ltd]:N/A
Organizational Unit Name (eg, section) []:N/A
Common Name (eg, YOUR name) []:Javier
Email Address []

Now you can find your own certificate under /etc/apache2/ssl:

ll  /etc/apache2/ssl
drwxr-xr-x 2 root root 4096 2011-12-11 12:08 ./
drwxr-xr-x 8 root root 4096 2011-12-11 12:07 ../
-rw-r--r-- 1 root root 2136 2011-12-11 12:09 apache.pem 

Finally we have to modify the Apache config file to indicate that it has to use the certificate that we just generated. This is done by editing the file:


Remove these two lines:

SSLCertificateFile    /etc/ssl/certs/ssl-cert-snakeoil.pem
SSLCertificateKeyFile /etc/ssl/private/ssl-cert-snakeoil.key

And include this line indicating the path to your certificate (Our certificate include the key itself, so is not needed to include the equivalent second line):

SSLCertificateFile /etc/apache2/ssl/apache.pem 

Once done restart apache:

sudo apache2ctl restart

You will have to add a new exception to Firefox:

If the button to accept is not enables (as it is my case) just clear all the recent history and try again:

The last thing we are going to do is to disable the 'Non-https' access, so everybody trying to access our site will have to do it using https. There is a more elegant solution that is redirecting the request from the 'Non-https' to 'https', again this can be done by manipulating the Apache config files, but this is out of the scope of this post.

To disable the 'Non-http' access just type:

sudo a2dissite 000-default
sudo apache2ctl restart

Now if you try to access to your site without https, ( you will receive a nice 'Not Found' mozilla error.

If you want to avoid apache warning when restarting the server, just comment or remove the following lines in the config file /etc/apache2/ports.conf:

NameVirtualHost *:80
Listen 80

B. Enabling Authentication

Now that we have set up a secure conversation by using ssl, we can ask our users for a password and control the access to our site.

The first thing we have to do is create a folder and a file where we are going to store all our users ans its associated password:

sudo mkdir /etc/apache2/passwd
sudo touch /etc/apache2/passwd/passwords

Now we create an user using the apache tool htpasswd

sudo htpasswd -c /etc/apache2/passwd/passwords javier 

You will be asked to introduce your password:

New password:
Re-type new password:
Adding password for user javier

If you have a look to the password file you will see that a new line with the user 'javier' and an encrypted password has been created:

cat /etc/apache2/passwd/passwords

Now we just need to configure apache to ask for a password when an user is trying to access to our site.

Open the configuration file:

/etc/apache2/sites-available/default-ssl default

And modify the following section:

        <Directory /var/www/>
                Options Indexes FollowSymLinks MultiViews
                AllowOverride None
                Order allow,deny
                allow from all


        <Directory /var/www/>
                Options Indexes FollowSymLinks MultiViews
                AllowOverride None
                Order allow,deny
                allow from all
                AuthType Basic
                AuthName "Password Required"
                AuthUserFile /etc/apache2/passwd/passwords
                Require valid-user

and restart Apache again:

sudo apache2ctl restart

Now when you try to access to your site you will be asked to introduce an user and a password (You might need to clear the Firefox history to avoid the browser cache):

And that's all folks!! You have configured (well you have mostly used the default configuration) a secure conversation plus authentication for your web site. This is the basic security access protection for an enormous amount of web sites out there on the Internet.

Saturday, December 10, 2011

Ubuntu on the Cloud

It has been a long time since the last post but I hope that this interesting post will compensate the wait.

Today we are going to talk about something that has become very trendy in the last years 'The cloud'. The first question is quite obvious... What is the Cloud? Based on Wikipedia's definition: 'Cloud computing is the delivery of computing as a service rather than a product, whereby shared resources, software, and information are provided to computers and other devices as a utility (like the electricity grid) over a network (typically the Internet).'

Do not feel silly if after reading this definition you are still wondering what is the Cloud... Actually the cloud can potentially means a lot of things. If you ask an Ipod user what is the Cloud he will probably relate it with remote storage, the ability of playing his music from several devices and some other Mac related services that are not very interesting for us at this moment.

The cloud I am going to explain on this post is mostly related with the 'Infrastructure as a Service' cloud approach. Yes ok, but... What that's means?? That means in its most basic meaning an Ubuntu running somewhere.

Somewhere is usually a Data-center, like the ones constructed and used by Facebook to provide services to an enormous amount of people, and it is basically a set of thousands servers running (and consuming) continuously.

Fig.1 Amazon Date-Center in Oregon

Fig.2 A typical Data-Center from the inside

And here is where the business comes. Companies like Amazon with hundred of Data-centers all around the world hosting their own web applications (Does tell you something?) decided to step forward and offer a piece of this Data-center cake to anybody willing at pay the price they established. And guess what, the technology used to provide remotes operative system is based on virtualization techniques.

I bet you are wondering right now how much does it cost having a full accessible Ubuntu running on the cloud 7x24... The answer is nothing (During the first year)!! Amazon offers you a free-tier that consists on a limited usage (750 hours of Amazon EC2 Linux Micro Instance usage [613 MB of memory and 32-bit and 64-bit platform support] ).

If you make your calculations a 31 days month has 744 hours, so basically you can run this micro instance for free during a whole year. Actually there are more limitations, like in the amount of data that you transmit to your remote Ubuntu and the number of http transactions (100.000). In any case this is more than enough for a little web server hosting a database and serving our dynamic web applications. Actually you have to take into account that each time that you start an stop your instance (Your remote Ubuntu) it count as an extra-hour. Here '' you can find all the conditions of the free-tier usage.

Here  you can find a calculator that allows you make an idea about how much you will pay based on your usage and the type of Operative System that you are hosting (You can host a Windows Server instead than an Ubuntu).

A. Creating a remote Instance

So let's go!! All you need is an Amazon account (you probably have one already) and sign on the Amazon Web Services (from now on AWS) account

Once you have created your AWS account you can log on the AWS Management console and create your instances (An instance == An operative system running on the cloud).

Fig.3 Dashboard page

From the dashboard select your region on the left and if it does not exists create a 'Default' security group (Under Networking & Security) add the rules to allow accessing to ssh, web, mysql, web https, and all that you potentially could need (You can take the following screen-shot as reference).

Fig.4 Security group

Back to the Dashboard, click on launch instance, select Launch Classic Wizard and under the tab “Community AMIs” search for the Ubuntu instance that fits with your needs. Here you can find a list of Ubuntu releases with its associated amids. Remember that only the micro instances are for free!!
Fig.5 AMID Selection

In our case, for Europe Region we have two choices:

eu-west-1 64-bit ebs ami-cc0e3cb8 ec2-run-instances ami-cc0e3cb8 --instance-type t1.micro --region eu-west-1 --key ${EC2_KEYPAIR_EU_WEST_1}

eu-west-1 32-bit ebs ami-c00e3cb4 ec2-run-instances ami-c00e3cb4 --instance-type t1.micro --region eu-west-1 --key ${EC2_KEYPAIR_EU_WEST_1}

In my case I will choose the 32 bits instance.

Select no preferences on the Availability zone screen, and leave all the data as appear by default on the Advance Instance Option screen.

Include a tag to identify your instance (this is normally used when you have to manage a big amount of instances) and

Create a new 'Key Pair' in the next screen. This will allow you to connect from your local computer by doing an ssh. If you loose this file you will need to create a new pair and associate your instance with the new one, so save it carfully.

Fig.6 Keys Creation and download

Choose the security group that you created in the previous step and click on Continue.
Fig.7 Security group selection

A final recap like the shown below will be displayed. Just click on launch and you will have your remote Instance running!

Fig.8 Instance recap

You can see your instance under the tab Instances.

Fig.9 Instances list

B. Connecting to the Remote Instance

If you want to connect to your remote instance (I bet you want) right click on your instance and select connect. This will generate a ssh command like: 

ssh -i UbuntuServer32.pem

You just have to open a local Terminal, go to the folder where you downloaded your private key (on the step where your instance) and type the ssh command. If everything is ok and you configured the security rules as indicated, you should get a welcome screen like the following one:

Fig.10 Remote connection


Tuesday, January 25, 2011

Virtualization with VirtualBox

VirtualBox is an open source software virtualization system. For those not really familiarized with virtualization technologies I will say that basically it is a program that allows you to run an operative system inside another operative system.

For instance you could run a Windows XP inside our brand new Ubuntu 10.04 (See fig.1). Of course you may wonder why the hell would you even want to have a Windows XP running in your computer, but this is a complete different history.

Fig.1 Windows XP running inside Ubuntu 10.04

In today's domestic world there are plenty of professional applications like Oracle that could be interesting to have and install but that it takes a lot of resources from the machine, here we can find a good usage of a virtual machine. You can create your virtual machine, install Oracle and have it ready to be used whenever you need it.

In professional world virtualized environment are really in trend. You can find a lot of different architectures (Several virtual machines running in the same computer at the same time working as a replication of a real distributed system, the same architecture serving as Testing environment in the different testing phases, virtualized platforms as development environment, integration environment etc).

Maybe the most famous commercial tool for virtualization is VMware, VirtualBox as far as I tested offers almost the same but for free. Here you can find an interesting comparative done by a blogger.

A. Installing VirtualBox

VirtualBox is included in the Ubuntu repository so you just need to open Synaptic (System --> Administration --> Synaptic Package Manager) and look for virtualbox-ose package. Once installed you can find it under Applications --> Accessories

B. Creating New Virtual Machine

VirtualBox is provided with a quite intuitive graphical interface, to create a new Virtual Machine just click on the wheel at the top left of the interface.
A wizard will (See fig.2) guide you all over the installation, asking you for the name and type of the virtual machine, the amount of ram that you want to dedicate for it and if you want to create a new Virtual hard disk or use an existing one.

Fig.2 New virtual machine wizard

If you want to create a new Virtual hard disk I highly recommend you to select the "Dinamically expanding storage" so the file in your computer associated to this Virtual hard disk will grow as far as you store information inside the virtual machine.

C. Configuring a Virtual Machine
If you want to access to your virtual machine in a easy way from your main Operative System (Host operative system) you need to configure the virtual machine network as a bridged adapter. To do so, select you virtual machine (Ensure that is powered off because otherwise you will not be able to modify any parameter) click on network and set it up as in Fig.3

Fig.3 Configuring the Network

D. Troubleshooting

If you retrieve an error like "  * modprobe vboxnetflt failed. Please use 'dmesg' to find out why [fail]" when you try to run the virtual machine with a bridged connection, try to remove the package virtualbox-ose-guest-dkms. You can find a better explanation for this issue here.

Monday, January 24, 2011

C Development Tools

C Development Tools

Life changed for C and C++ Developers with the arrival of CDT for Eclipse. Eclipse is one of the most famous IDEs (Integrated development environments) based in plugin approach architecture allowing to program in different technologies and languages under the same graphical framework.

CDT is an open source plugin that enhance Eclipse with the needed features to develop in C and C++ both large and small projects. It brings all the typical characteristics of an IDE (Syntax highlighting, Code completion etc) and it can be of course combined with other external tools like control version systems (CVS, SVN etc).

A. Installing Eclipse.

You have basically two ways to install Eclipse:

The classical one (Synaptic → Eclipse) and

Directly downloading from Eclipse download page (Eclipse for C/C++ Developers).

If you choose the second option, you can download the version that includes CDT from the Eclipse download page (See Fig.1) and therefore you can avoid the “Installing CDT” step, but if you are planning on download a whole Eclipse for each feature that you need you will probably waste a lot of space in your hard disk. Therefore I recommend you to choose the first option.

Fig.1 Eclipse download page

B. Installing CDT.

If you choose the first option, you need to follow a couple of extra steps to install an run CDT:

1. Go to the download main page of the CDT Project and log the repository version that you need. In my case I am going to install it on a Eclipse Galileo (Is the one that is installed by default in Ubuntu 10.04) so my repository is

2. Open Eclipse and click on Help –> Install New Software. A new windows is opened, just paste the repository address in the first text box (See fig.2) and wait until the CDT Installation packages appears.

Fig.2 Adding CDT to Eclipse

3. Select both installation packages (CDT Main Features and CDT Optional Features) and click Next. Leave the next screens as it appears and if you do not receive any error message it means that CDT is now installed and integrated in your Eclipse.