That was surprising. Very cool and creative gift I got from our team:)
I was quite surprised when after logging in to the site I saw something different than it should.
Here is what I saw:) Read more…
Amazon Web Service is a set of various services. Their scope is so wide that we couldn’t possibility describe them in a few sentences. My everyday work, however, would not be possible without two of them.
S3 – Simple Storage Service, which I will describe in another article on Content Distribution Networks in TYPO3 CMS.
EC2 – Elastic Compute Cloud – used to test various configurations of any implemented solution. In your average conditions it is hard to test the application’s behaviour on e.g. 6 different servers. That takes time – either in the server room or in-house – and always involves costs and wasted time. EC2 allows us to run any number of server units, and the low cost allows us to experiment as much as we need. Read more…
From time to time it is worth sharing the successes. Some time has passed since the occassion, however it is better later than never. I became a certified integrator of CMS TYPO3 
Various pages and various categories.
Today we are going to write about a common limitation of the tt_news plug-in integrated in TYPO3 CMS, to which no easy solution has been created. Imagine a site, which has many categories and a large number of updates. Due to the site’s composition, our clients want the tt_news records to be mounted directly to a specific category in the page tree. The site has many sections, and different news items are displayed in different sections. Defining a single view for each of these categories proves to be a problem. Remember that each instance of the plug-in needs to have its own single view. Add a global search through each and every category, and a second problem manifests itself. The search results plug-in should link each record to an appropriate single view page. We can’t create a common page, because (for example) the news:
“Oscar Awards” should be displayed in the Main > Stars > Events section,
while
“Interview with Gary Oldman” should be displayed in the Main > Stars > Interviews section.
Read more…
Not so long ago, Google performed an experiment and increased the number of results on a single search page – from 10 to 30. Although the loading time for 30 results had risen to 0.9 sec from 0.4 sec, the overall search traffic fell by 20 per cent. That shows how important a search engine (and its performance) is for a website.
I don’t want my blog to be solely about performance problems. However, when working on large sites you should constantly keep performance issues in mind, especially when thinking about minimizing costs related to any increases in traffic down the road.
The indexed_search indexing search engine built in TYPO3 leaves a lot to be desired and overall, I don’t recommend using it in extensive sites with large amounts of information. A better solution (albeit requiring more configuration work) would be to use the mnogosearch plug-in. This extension allows us to connect an external indexing engine of the same name, and to use its array of functions in conjunction with TYPO3. To find a detailed configuration description, visit the TYPO3 extension page. I will cover a number of additional tricks, which allow for indexing a sql database. Read more…
We use more and more social websites and portals, creating many user profiles in the process. But how many usernames can we possibly remember? The usernames we choose during the account registration process are often forgotten.
Is there no easier way to log on?
In fact, there are lots of possibilities. First of all, we could make use of OpenID, although I still have reservations to using the same login parameters in all websites in which I regularly participate. TYPO3 provides a simple solution by allowing the user to provide either an email address or a username. Usernames tend to be forgotten, but it’s much harder to lose track of your email address 
Username-based login is a standard feature in TYPO3, whereas the ability to use either a username or an email address in the login field needs to be programmed in. Read on to find out how to easily implement this feature through a TYPO3 service.
Read more…
NOTICE. New version available. Now you can use evo_nginx_boost without nginx. You need ony TYPO3 and memcache. New manual of evo_nginx_boost – http://techblog.evo.pl/en/evo_nginx_boost-extension/
The TYPO3 performance issue is often a subject of discussion. I found a number of various tricks, which can speed up the page generation time, including nc_staticfilecache and dmc_highperformance modules. The latest alpha version of TYPO3 contains radical changes in the caching mechanism, enabling us to choose from various support functions, such as file, memcache, database or apc.
All of the above solutions, with the exception of nc_staticfilecache, have one thing in common: in order to read cache, one has to invoke a php process, read and parse the cached data, and then send the information to the browser.
Let’s perform a simple test using the ab tool. We will check the number of queries per second for a complex Web site.
Recent Comments