Have you ever thought about a method of saving sessions without getting a lot of session files or using a selfmade session handler?
The answer for this question is memcache. A module giving you the ability to store variables in memory for a defined time. You can install the module on a Linux and a Windows Server. I only want to explain you how to set it up on Linux Debian Etch.

At first you have to do is to install the server module.

apt-get install memcached

If the application is already started, we have to kill the process to setup what we want.

memcached -d -u www-data -m 2048 -l 127.0.0.1 -p 11211

Explain this command:
memcached (command)
-d (the parameter to run it in background)
-u www-data (to run it under the standard apache2 user www-data)
-m (Sets the space of the memcached server (in Mega Byte) (in this example 2048 MB (Mega Byte))
-l (Sets the IP the memcached server listens to)
-p (Sets the Port the memcached server listens to)

Now the memcache server runs as www-data with a space of 2024 MB RAM which is listen to localhost:11211. The next step is to involve PHP. For that we have to enable the module memcache.

You could get the latest version here. Please use the wget command to download the file. (Note: replace X.X.X. with the current version)

cd /usr/local/src
wget http://pecl.php.net/get/memcache-X.X.X.tgz
tar zxvf memcache-X.X.X.tgz
cd memcache-X.X.X
apt-get install php5-dev
phpize
./configure
make && make install

Okay we have compiled the files. Now you have to move the memcache.so to your extension folder of PHP5. In my case I found the memcache.so in /usr/local/lib/php/extensions/no-debug-non-zts-20060613/memcache.so, but maybe your location is different.

Now copy the file to /usr/lib/php5/20060613+lfs

The next step is to get the latest version of the memcache php5 extension in our extension directory. All we have to do now is to configure the php.ini by adding the following line.

extension=memcache.so

Before you could restart the apache2 server you need to set up the memcache that it will be used as session handler. The standard will look like this:

session.save_handler = files
session.save_path = "N;MODE;/path"
session.gc_probability = 0

You have to change it to:

session.save_handler = memcache
session.save_path="tcp://127.0.0.1:11211?persistent=1&weight=1&timeout=1&retry_interval=15"
session.gc_probability = 1

Save the ini file and restart your apache2 server.

/etc/init.d/apache2 restart

When you take now a look into your phpinfo it should look like this:

(the white spaces are private settings)
Congratulations! If you create now a session it will be stored into the memory instead of creating a file.

As I checked my emails this morning, I was suprised as well as happy. I got an email from google that they have added TechnoBase.FM to their news search.

I’m really glad about that because hopefully more people get to know our radio. Now more readers are attracted to our page and so the participation of our news posters will be honored.

You can have a look by yourself: http://news.google.de/news?pz=1&ned=de&hl=de&q=site%3Atechnobase.fm

A friend told me something about a tool called fiddler2. It is a web debugging proxy. You only have to start this program and surf in the internet. You can see which pages / files are opened by each application, you can check the status (HTTP Status) and the used protocol, too.

It helps a lot by searching for errors!

As I descripted in that post, we have updated the news system. Now it is possible to post news at one special page or at all pages. To avoid any problems and allow you to choose your favourite radio news, you can now subscribe radio special rss feeds.

http://feeds2.feedburner.com/technobase/news/only
http://feeds2.feedburner.com/housetime/news/only
http://feeds2.feedburner.com/hardbase/news/only
http://feeds2.feedburner.com/trancebase/news/only
http://feeds2.feedburner.com/coretime/news/only

Enjoy it! ;-)

Toby, a friend of mine, and Professor Trommler presentated the offline filesystem at “Chemnitzer Linux Tage (Germany)“.

The OFS (offline filesystem) is a offline-layer which is (or should be) able to extend every filesystem with offline capabilities. This means, that the user is able to work on his network documents, event when he is disconnected from the server. When connecten is restored, the filesystem automatically synchronizes the files.

It is just a small update for us, but a big for SEO.

Now on the We aRe oNe Pages the news have a new link and the page title changed, too. For exampe the new release of Phunkless is linked on http://www.technobase.fm/news/9029-phunkless-i-wont-let-you-down.

Enjoy it!

As descripted in this news we released the new releases system on the We aRe oNe pages. Now you could search for releases, have a nice overview about all the releases and an interesting detail page. This system was created by Remolus, a member of the rarecore project!

But, that was only the first step. New features are coming soon:
- Artist overview with all releases grouped by an artist
- Many links to different online shops for buying the tracks (don’t search a long time)
- Link the tracklist to the releases
- Create your own charts by clicking on a track of the tracklist
- Comments and vote releases

Our 5th We aRe oNe radio, called CoreTime.FM, has started! You can read some information about on that news.

It is a radio only for hard-, french, speed- and darkcore, … and is leaded by Björn “derade” Kuiper , the founder of Hard Music Entertainment. He is member of the We aRe oNe Community since 28th March 2005.

The page is designed by spYed – one of the best designer I know and a member of the rarecore project!

I have found a nice free icon set and I want to use these icons for my next update.
These free icons are exactly that, free! You can use them anywhere, all they want you to do is to link these icons by linking to their page.

This icons have a nice look!

It is often helpful to use an autocompleter on your website. Perhaps in search dialogs? An example is google on an extra search page.

I tested some free tools, but I had always the problem that there was no possibility to use two or more autocompleter on one page and to organize the dependency. E.g. insert a name of a country in the first textbox and in the second textbox insert a name of a city. On the city textbox you only want to see cities belonging to the inserted country – no other cities.

This problem was the reason why I developed my own autocompleter using xajax. This framework allows me to respond on the key events and select the values from the database. I could write my own functions and use any parameters.

I generated a textbox with a hidden div below which is calling a function on the onkeyup event by using xajax. In this test I mark the searched string with a span and on CSS I colored it blue. Additional I added a mouseover using CSS.

Demo *** NEW VERSION ***

download: Autocompleter v2 (119.55KB)
added: 08/11/2009
clicks: 1419

That’s all!