The event will take place on Friday 24th April 18:30 – 20:00 at Nixon (map).

We will be deploying both Magnet commerce and Zuni greek social network on the cloud during the following days.

Some more useful information can be found at Rightscale’s blog.

Update: The post on Amazon’s web services blog can be found here.

Now before starting the tutorial, I would like to write a few things about the purpose of it. The idea behind it, is to be able to run a server with data stored on EBS volumes. This gives you a lot of flexibility. For example, let’s say that you have two EBS volumes: One 10GB volume for you mySQL database and another 150GB volume for serving files through apache. If your website grows and you need to have a bigger storage just for your apache server you can simply take a snapshot of that volume, create a new volume using that snapshot but with a bigger storage (up to 1000 GB) and attach it back to the server. This is a matter of just a few minutes. Moreover this can happen for CPU problems as well. Once you need more CPU, you can create a new, bigger instance and attach your EBS volumes to that. Amazon offers great options for that, which practically can cover most of internet websites. Should you need more, scaling is much easier since they are many tools out there like RightScale or scalr which can add / remove instances on the fly using very advanced triggers for that.

Now let’s get our hands dirty:)

To get started you need to do the following things:

1. Launch an instance of the following AMI: “ec2-public-images/fedora-8-i386-base-v1.07.manifest.xml“.
2. Create an EBS volume and attach it to the instance (attach it to /dev/sdh).
3. Request a new elastic IP and assign it to the instance.
4. Connect to your new server using SSH.

All the above steps can be done using ElasticFox (or any other tool, even through command line).

Once you have logged in as root to your server the first thing to do (only do this once) is to format your newly attached hard drive (EBS volume actually:)). To do this, run the following command:

mkfs.ext3 /dev/sdh

Then you need to create a directory on your server where the new volume will be mounted.

mkdir /ebs1

To mount the volume to that directory run the following command:

mount -t ext3 /dev/sdh /ebs1

Note for the steps above: If you ever terminate your instance and create a new one, you will only need to create the directory and mount the volume there. The volume would be formatted already and your data will be in there!

Now you can install apache, php and mySQL using yum:

yum -y install httpd php mysql mysql-server php-mysql

To make sure that these services are started each time your instance boots, run the following commands:

/sbin/chkconfig httpd on
/sbin/chkconfig –add mysqld
/sbin/chkconfig mysqld on

You can start apache and mySQL using:

/sbin/service httpd start
/sbin/service mysqld start

Once you have started mySQL it is advised that you do some stuff for security reasons. The first thing to do is to change the root password:

mysqladmin -u root password ‘new-password’ (where new-password is the password you want for root).

Then connect to mySQL:

mysql -u root -pnew-password (replace new-password with the password you specified before).

Remove the test database:

DROP DATABASE test;

Disable anonymous access:

DELETE FROM mysql.user WHERE user = ”;
FLUSH PRIVILEGES;

Then exit mySQL using quit.

Now it’s time to move apache to EBS:

First we need to stop the service:

/sbin/service httpd stop

Then we move httpd to EBS:

mv /etc/httpd /ebs1/httpd

And create a symbolic link to it:

ln -s /ebs1/httpd /etc

Also do this for the document root:

mv /var/www /ebs1/www
ln -s /ebs1/www /var/

And restart apache:

/sbin/service httpd start

Note: If you terminate and start a new instance simply remove the /etc/httpd and /var/www folders and just run the commands to create the symbolic links.

Then we do the same for mySQL:

/sbin/service mysqld stop
mv /var/lib/mysql /ebs1/mysql
ln -s /ebs1/mysql /var/lib
/sbin/service mysqld start

This only moves the mySQL data, not the logs. You can edit /etc/my.cnf to configure where logs are stored and also move the log directory to EBS if you want to.

So now you have a fully functioning server on EC2 that stores and retrieves data to and from and EBS volume. The EBS volume and be backed up almost instantly using a snapshot. If you don’t use innodb for your tables it is advised that you lock your tables while you take the snapshot.

Next steps / thoughts:

Some might want to create a hosting enviroment for many websites. What you can do for this is to edit your httpd.conf file (located on EBS /ebs1/httpd/conf/httpd.conf) and enable Virtual Hosts. What I did is to create a directory /ebs1/sites and then create users on fedora using the following command:

useradd -d /ebs1/sites/mysite1.com user-for-mysite1

Also don’t forget to specify a password for the user using:

passwd user-for-mysite1

Once you have created the user you can point the document root for that specific site to his home directory.

Also to enable the user to upload files to his/her website using SCP you need to allow password access through SSH (be careful on security issues though). To do this, edit /etc/ssh/sshd_config and change the line “PasswordAuthentication no” to “PasswordAuthentication yes”. Also restart sshd by running /etc/init.d/sshd restart.

The final step is to link real domains to your new server. We assume that you have already assigned one or more IPs to your server. To point a domain to a specific IP you need to edit its DNS records and add an “A record” that points to that IP. Once the domain resolves on your server you will be able to use it for your website. If you need to host multiple websites, create one VirtualHost record for each domain (using the appropriate servername and documentroot values) and off you go!

Special thanks to http://agiletesting.blogspot.com/ for the useful information that I found there for this tutorial.

Today Microsoft launched Azure, a cloud for applications designed for Microsoft Products. Azure is available for free at this time through the CTP program. Pricing will be revealed once the program is available for commercial use.

Considering that Microsoft will be launching a cloud operating system in the future I consider this move very important.

* Amazon EC2 is now in full production. The beta label is gone.
* There’s now an SLA (Service Level Agreement) for EC2.
* Microsoft Windows is now available in beta form on EC2.
* Microsoft SQL Server is now available in beta form on EC2.

I played around with ElasticFox and launched a few Windows instances. The instances work very well and there are AMIs that have built-in support for IIS, ASP.NET and MS SQL. Of course you can customize your own AMI and deploy any software you like.

In case you wonder what the computer properties look like in windows when you run an extra large instance, here is a screenshot for you:)

AWS include some of the very best services to launch and scale a start up. During the meet up we will be discussing about:

Elastic Compute Cloud (EC2):

Amazon EC2’s simple web service interface allows you to obtain and configure capacity with minimal friction. It provides you with complete control of your computing resources and lets you run on Amazon’s proven computing environment. Amazon EC2 reduces the time required to obtain and boot new server instances to minutes, allowing you to quickly scale capacity, both up and down, as your computing requirements change.

SimpleDB:

Amazon SimpleDB is a web service for running queries on structured data in real time. This service works in close conjunction with Amazon Simple Storage Service (Amazon S3) and Amazon Elastic Compute Cloud (Amazon EC2), collectively providing the ability to store, process and query data sets in the cloud.

Simple Storage Service (S3):

Amazon S3 provides a simple web services interface that can be used to store and retrieve any amount of data, at any time, from anywhere on the web.

DevPay:

Amazon DevPay is a simple-to-use online billing and account management service that makes it easy for businesses to sell applications that are built in, or run on top of, Amazon Web Services. It is designed to make running applications in the cloud and on demand easier for developers.

Third party tools to be discussed:

Rightscale:

RightScale enables companies to create Web solutions running on cloud computing providers such as Amazon EC2 that are scalable, reliable, easy to manage, and affordable.

Ylastic:

Ylastic is a unique application that is built on top of EC2, S3, SDB, SQS and FPS, and provides an intuitive and powerful user interface for managing the main Amazon Web Services – S3, EC2, SQS and SDB, along with monitoring, alerts, reporting, and other goodies. It enables you to spend your precious time focusing on bringing your ideas to life rather than dealing with the daily nitty gritty of the environment. Don’t waste your bandwidth by building your own UI, monitoring and administration tools. Leverage Ylastic and use the saved time to build value added applications.

Jollat:

Jollat is an easy-to-use GUI to manage S3 and EC2 services. It provides access to the full functionality of the services, but everything is integrated into a simple interface. Versions are available for the Mac, Windows and Linux.

Bucket Explorer:

A User Interface for Amazon S3. Helps you to transfer files to and from s3. Allows managing ACL, creating signed URLs, managing bucket logging and allows access to shared buckets and files.

Since the meet up will be discussion oriented, it is suggested that you first study the general overview of these services. We will focus on how these tools can be used efficiently, show a hands-on tutorial for starting an EC2 instance and also present some useful third party tools.

The event in take place at Eleytheroudakis Bookstore (cafe, 6th floor), Panepistimiou 17, 7pm on 29th October.

Larger view

The first part of the problem wouldn’t exist if the internet was a single social network. In real life though we have facebook, myspace, linkedin, and many other social networks. Each one serves it’s own purpose and most of the people need to use more than one social networks. This is the part of the problem that dataportability.org is trying to solve. Basically, there will be “trusted” banks of data where users will be able to grant access in order to share their data with other social networks. It’s a great idea since it will allow us to control which data we want to share on which social network. Moreover we will have a permanent identity on the web. We will be a live part of it no matter what happens to a website or a social network.

In real life though there is a quite significant problem: Will people trust those data banks that will be holding their data? From the first point of view I would say yes, because there is no difference than giving your data to facebook or myspace. However, on the other side I would say that the ultimate goal is that every person holds his own data and decides if and how he wants to share them. This could be a perfect solution but there is a huge problem! People are not connected on the internet 24/7. If I hold my data and someone searches for me, nothing would appear if I am offline at that time because my data would be with me. In a few years though I am pretty sure that everyone will be connected 24/7 either by his cell phone, his computer or even his car.
Saying so, we will have a world of always connected people who will be holding their private data. Practically, we will be able to build a p2p network where only a map/filesystem of people names (very basic data) would be hosted on a data bank and all their private data they want to share would be served by them directly.

As a conclusion I should say that technology is not there yet. It might take several years to make the internet so reliable that this whole thing could work flawlessly. But the data portability project is a good way to start with. It will make things a lot easier for everyone.

Try the advanced search form to get language specific results.

More info can be found here.