Archive for the ‘WebSphere Application Server’ Category

Mediation handler tooling broken

Tuesday, June 10th, 2008

Despite being sidelined by the arrival of WebSphere ESB, there are still times when it is useful to use the mediation support in WebSphere’s service integration bus (of which more in another post). It’s obviously a while since I’ve done it though as apparently the mediation handler tooling in Rational Application Developer has been broken since v7 was released. I’m told that the fix will finally make it in to 7.0.0.7.

Wednesday @ WSTC

Thursday, May 8th, 2008

The first slot of the day was one of those rare occasions where I wasn’t quite sure what to attend. In the end I opted for Alex Polozoff’s "Large Topology Tactics and Tradeoffs" presentation as I have great respect for his opinions. The emphasis here was not so much on the technology but the processes and the people. Most of it seemed common sense (at least to me) but the presentation should be a useful resource for anyone embarking on large scale WebSphere topologies.

(more…)

Web messaging lack of service

Monday, March 31st, 2008

Last week I spent a not entirely constructive hour or two with a customer trying to get them started with the web messaging service. This is a component of the Feature Pack for Web 2.0 which provides a bridge between the default messaging provider’s publish/subscribe capabilities and an Ajax (Comet) client. The InfoCenter contains a great quick start guide but it was missing one vital piece of information.
(more…)

Jython scripting

Monday, January 28th, 2008

I’ve had the pleasure of working with my UKISA colleague Andrew Simms on a number of occasions. Some may have been luck enough to see his “Monty Jython’s Scripting Circus” presentation on the conference circuit recently. If not, fear not, as he now has a developerWorks article to set you on the path to being a Jython pro!

WebSphere MQ Java in J2EE

Wednesday, January 2nd, 2008

Thanks once again to Paul Titheridge, this time for pointing me to this technote covering the use of the WebSphere MQ Java APIs in a J2EE/JEE environment. It seems that IBM does now support the use of the WebSphere MQ Java interfaces in a WebSphere Application Server environment (check the supported software for your release) but this technote provides you with lots of reasons why you shouldn’t! The Java interfaces do have functionality that isn’t currently available in the JMS API (for example message segmentation and important parts of message grouping) but that needs to be weighed carefully against the drawbacks. JMS should always be the default option unless you have a very good reason not to use it.

Cleaning up dumps

Saturday, December 1st, 2007

No – this has nothing to do with nappy changing! I’ve spent the past two weeks as part of the support team for a customer’s WebSphere Application Server environment. Although I know all the theory (I’m an IBM Certified System Administrator for Versions 5.0, 6.0 and 6.1) it’s not very often I spend long enough at any one customer to get my hands dirty. My Korn shell scripting, Jython and vi skills certainly improved rapidly during the time I was there!
(more…)

JMS problem determination

Friday, October 12th, 2007

The JMS Problem Determination Redpaper covering WebSphere Application Server Version 6.1 was published at the start of the month. As the long list of authors indicates, Rich Coppen has done a fantastic job in getting input from the whole of the development team so this is a pretty comprehensive resource.

Messaging identities

Friday, October 5th, 2007

Given Alasdair has been kind enough to include me in the acknowledgements, that guarantees him a plug for his developerWork’s comment line entitled “Connect an application to the default messaging provider using the run-as identity“. This describes how to connect to the default messaging provider in WebSphere Application Server using the application’s current runtime run-as identity rather than, as is usually the case, the credentials administratively associated with the connection factory or resource reference. As always, just because something is possible, doesn’t mean you should do it, so please head Alasdair’s comments about what to consider before implementation.