Saturday, September 11, 2010

Guvnor JCR Implementation Support: JackRabbit and ModeShape

I have started working on the Guvnor project recently. Guvnor manages a SOA repository and sits on top of a JCR implementation. Up until now there really where not too many JCR implementation available. Really JackRabbit was the only viable Open Source gig in town. This has changed now the ModeShape project really has been picking up steam. They recently completed implementing the JCR 2.0 specification. ModeShape has been designed from the ground up with scalability in mind, and it can run in a clustered environment (using JGroups).

--Kurt

Tuesday, July 20, 2010

Matt Ridley explains how open source works

Although he doesn't know it he is explaining the inner workings of open
source, and why it will be more successful as the number of people with
computers increases.

http://www.ted.com/talks/matt_ridley_when_ideas_have_sex.html

--Kurt

Thursday, July 8, 2010

OSX and slow DNS issue

I just 'fixed' my macbook from having terribly slow DNS lookup performance by turning off IPv6 in both of my network adapters (switch from automatic to off). The difference is not funny.

Cheers,

--Kurt

Thursday, May 13, 2010

Guerrilla SOA in the Cloud.

This evening I'm listening to a presentation by John Graham "From the ESB to REST and Clouds an Open Source", who is talking about ESB, REST and Clouds at the SUN campus in Burlington, MA. It is a nice overview of the evolution of system architectures, with the conclusion that as system became more distributed we developed different integration solutions; from Corba, Asynch Messaging to ESB. The talk is spiced up by some good word plays ("Being thrown under the ESB"), quizzes and code examples. It's entertaining.



After a little while we arrive at Guerrilla SOA coined by Jim Webber, which is a lighter style of SOA. Finally we arrive at how REST can be used in a cloud environment in a complementary fashion to manage ESBs.

Cheers,

--Kurt

Friday, May 7, 2010

Match made in Heaven: Maven Multi-Project and Eclipse Working Sets

Recently Maven dropped support for 'nested modules' which was nice from an organizational point of view, but since all the maven modules shared one classpath in this case it wasn't really working. So I started using the 'import maven modules' after checking out the main project, which creates different project for each maven module. This works well, but it explodes the amount of project in my project explorer. I have been looking for a way to organize this and today I tripped over this article from Byron, about Categorization of projects in Eclipse using 'Working Sets'. This is what I had been looking for, and it was there all this time!

Cheers,

--Kurt

Tuesday, May 4, 2010

JBoss Quick Reference Card

This is the best Getting Started Guide on JBoss I have seen! Short and to the point.

--Kurt

jBPM Developer Guide Book (P)Review

When I returned from our spring vacation trip one of the things I found waiting for me on the porch was a copy of the book "jBPM Developer Guide". After my last jBPM book review I mentioned that the target audience for that book was not the jBPM developer. Well that seems to be addressed now. Here is a link the chapter 5, if you're interested while I'm reading the book myself.

Cheers,

--Kurt

Wednesday, March 17, 2010

JBossESB MBean Service Deployment Order

In JBossESB, MBeans can be deployed inside of .esb archives. The deployment is pretty much standard and the mbean configuration lives in the jbossesb-service.xml right in the root of the .esb archive. If you need the mbean to be up before the rest of the .esb deployment, then you can reference the mbean in the META-INF/deployment.xml. If the mbean depends on another .esb service archive to be deployed first the you need to add that dependency to the jbossesb-service.xml.

If you're looking for an example check out the jbossesb.esb.

--Kurt

Sunday, February 7, 2010

Cool MySQL Database Queries

1. Select a count of all rows in your database:

SELECT SUM(table_rows) FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'database_name';


2. Select count of all rows of each table in your database:

SELECT table_name, table_rows FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'database_name';


3. Gets stats on a per hour basis, written out to CSV:

SELECT DATE_FORMAT(date, '%m/%d/%Y %H:00:00') AS date, DATE_FORMAT(date, '%Y%m%d%H') AS hour, count(id) AS requests, max(time), avg(time), count(distinct userName) INTO OUTFILE '/tmp/usage.csv' FIELDS TERMINATED BY ',' FROM statslog GROUP BY hour ORDER BY hour ASC;