http://metatheory.wordpress.com/2007/12/11/atg-management-using-jmx/

-Andrew

I will keep ‘em coming:) No worries:)

I’ve checked out that group, however at the moment many of the questions seem to be from people who are just trying to avoid reading the manuals, so I haven’t gotten too active there. But I will definitely keep an eye on it.

Thanks!

BTW I should also add that if you’re interested in ATG technology you should subscribe to the google group ATG_Tech. There’s a growing community of egar ATG developers posting and sharing information. Many current and former ATG employees have joined.

http://groups.google.com/group/atg_tech

Thanks for leaving a comment. I had no idea anyone at ATG read this:)

Because you mentioned it I had to search the docs (2006.3) to find JMX mentioned under EJB clustering, in one paragraph on one page. (I’m ignoring the JBoss admin guides references to the JBoss JMX console for the obvious reasons.) So I’d hardly say that ATG has used JMX extensively. In fact this sentence: "We do not currently support the use of third-party tools to invoke these JMX interfaces" seems to imply that JMX isn’t exactly embraced. So while ATG may support JMX/MBeans as part of the J2EE spec, it’s a far cry from something like JBoss where you have the entire platform exposed for management and monitoring via JMX, and tons of documentation and write-ups for third party tools, using JMX management consoles, etc…

I’m not trying to imply that Customer Intelligence isn’t a worthwhile product, I’m just pointing out that much of the core product seems to be languishing. For instance, in the 2006.3 Repository Documentation you find this gem:

"Demonstrates that D5 repositories (unlike D4.5) do not require a "seq_num" column."

Clearly the Repository API and documentation hasn’t changed much since Dynamo 5. Or at least not enough to warrant new examples. I could go on for ages about the strengths and weaknesses (as I see them anyhow) of Repositories, but maybe that’s a separate post.

Don’t get me wrong, I still really like using ATG, but the more time passes, and the more I spend time using JBoss, Seam, Spring, etc… the more apparent the stagnation appears.

Does 2007.1 radically embrace JMX? Or is that one paragraph under EJB Clustering in the J2EE Dev Guide the extent of ATG’s JMX usage? I’m not trying to be snarky, but I’m a little dismayed that paragraph represents the best rebuttal there is to my post above.

ATG used to absolutely lead in this field. Before there was jsp, before there was hibernate, ATG was THE pioneer and had the best technical framework around. While JBoss has yet to field the personalization and commerce layers (and may never), the underlying app server and projects like Seam, really exemplify the power of some of the newer (and I’m using this loosely as JSF for instance is over 3 years old) technologies. As an ex-ATG employee I hate feeling let down when I switch from my personal projects on Seam on JBoss to my day job on ATG.

Anyhow, I’m happy to chat about this any time if you want more information.

Thanks!

–
Devon