By Martin Harris on July 13th, 2010
Several months ago now I posted a series of articles about offline backup. Its a topic that more and more people are dealing with. The realisation that you have many gigabytes of digital media that could easily be lost combined with better broadband is prompting a move to off-site backups. So I posted up my exploration into backing up off-site. Recently I have had a few people ask me if I am still happy with my final choice, crashplan.
Even my mum uses it!
I am very happy with the choice. In the original posts I managed to setup three home machines, one Linux one windows and a mackbook. Since then I added the iMac that I keep at my parents. This is a sign of how easy crashplan is to maintain. In a nutshell once you have set it up, you just forget about it. This has to be one of the most important things about backing up. It does not require any maintenance whatsoever. This is encouraging when your several hundred miles away from one of the machines. I can use the web to see if its backed up, and get alerts if a period goes by and it fails. I just checked it now, 100% backed up today.
What else?
The backup client allows you to set how much bandwidth and CPU its allowed to use. I use the default as I have never found it hogging either. I am probably lucky that so far several backups on different machines have not started at the same time. If it did its easily curable.
In summary
If you have anything worthwhile keeping, and want off-site backup – get a crashplan account. If you have friends with space, use crashplan for free and backup to each others machines. In fact do both!
By Martin Harris on June 22nd, 2010
I come from a Java background. So over the last decade I have moved away from managing dependencies on projects using jars in shared folders. Most professional java programmers have used Maven or Ivy. Recently I looked at Groovy and Gradle. All well and good. What do programmers use for .NET, WPF and Silverlite? I have been trying to find out and there are mixed opinions. Some write scripts to copy .dll files, others rely on the Visual Studio solutions. None of these solutions are scalable. I found this project: npanday Has anyone used it on a large project? I would love to hear your views or review. What does microsoft do? What do you do?
Please comment here I am very interested in hearing what other programmers do for .NET dependency management.
Links for the Java world of dependency management
http://maven.apache.org/
http://ant.apache.org/ivy/
http://www.gradle.org/
By Martin Harris on May 31st, 2010
Unitils provides its own DataSource. This causes problems if you want to use Spring IOC to inject your own. There is a way around this but its not elegant. In future versions of unitils I believe the team is going to provide better support for less intrusive methods. This post documents a way to use your own spring defined DataSource.
Continue reading How to link unitils to a spring wired datasource →
By Martin Harris on May 24th, 2010
I was talking to Shay Banon (Gigaspace Software Architect) he mentioned using GigaSpaceFactoryBean for writing gigaspace unit tests. I could find no examples on their site, although I concede it may be there somewhere. This shows post shows a simple example.
He also mentioned using the admin api to do full end to end integration testing. The admin api has huge potential as its able to control gigaspace containers, deploy into them and collect statistics. Its a powerful api.
This is documented on their site: http://www.gigaspaces.com/wiki/display/XAP7/Administration+and+Monitoring+API
Continue reading Unit Testing with a GigaSpaceFactoryBean →
By Martin Harris on May 21st, 2010
On the 18th of May, 2010, the very new tumbler-glass project by Studio Pragmatists uploaded Tumbler 0.2.1 to Maven. Having recently written about JBehave I found myself really liking the concept of behavior driven development. So I decided to write a similar article about Tumbler. If you want the project code its available in my example project.
4 hour time box in 20 minutes!
Once again I decided to time box the work to 4 hours. This time though the whole process only took about 20 minutes. The product owner and testers produce a story file. The Tumbler format allows for multiple stories each containing scenarios, so its possible to cover a complex set of requirements in one file. This allows for flexibility when breaking down the work into tasks. As per the usual behavior driven approach, a scenario contains the Given, When and Then sections which describe the behavior. Continue reading Tumble dried BDD from Studio Pragmatists →
