ajax4jmx' Initial Release - AJAX interface as JMX console


News: ajax4jmx' Initial Release - AJAX interface as JMX console

  1. ajax4jmx has been announced on the GWT forum. It's a GWT-written library (obviously), and requires code changes for the JMX beans, but considering the nature of most interactions with JMX, it might be a good idea indeed. The project can be imported in Eclipse; other IDEs may have to copy the GWT artifacts into the correct places for a web application. It currently provides a simple JMX browser, as well as a series of JMX components for use in other applications. One of the things it highlights is the use of external components in GWT, which isn't as easy as it could be - first, what do you think of the component, and second, how would you address GWT's external component use? [Editor's note: the demo site is down at the time this news was posted.]

    Threaded Messages (2)

  2. Cool[ Go to top ]

    I don't know why this hasn't got any comments yet, too close to Christmas perhaps (oops, sorry I meant "holiday" for Americans). Anyway, this looks pretty cool, I've just tried the demo but it seem to crash on me. Anyway, I like the idea, everything has bugs so I'll try again later. -John- CTO C24 London, UK
  3. Re: Cool[ Go to top ]

    Hi John, I am not sure what you are attributing as 'cool' but if it is based on the functionality and not the fact that this is using GWT/Ajax and is open source then you might also want to check useful innovations in the industry related to JMX. Off-line Diagnostics of Container Runtime State http://blog.jinspired.com/?p=25 There are a number of advantages to our approach to inspection and analysis of system and component runtime state information such as JMX attributes. 1. We support off-line browsing of the complete repository (our name for an extension) 2. We collect and structure the repostory data based on natural user access patterns 3. We can go beyond the limitations within the API (JMX) and record complete object field state graphs (mementos) 4. We record all referenced class meta data including code sources, class loaders, interface/implementation hierarchy, annotations,.... 5. Data collection is collocated so we do not suffer common performance problems with excessive round trips between the repository sever (JMX MBeanServer) and the console. Collocated and batched! 6. We offer fast and easy access to the on-line and off-line repository from different UI views without the inherent slowness and poor interaction of Ajax based applications. 7. We can collect data from multiple repositories and across multiple processes within a cluster and merge them into a single snapshot or real-time view which is extremely useful for support organizations looking to provide effective and efficient problem analysis. 8. We have more than 25 repository extensions. 9. Repository models support annotation (notes) of the off-line model for improved collaboration across teams. 10. We also support the mapping of JMX attributes (including complex types) to JXInsight metrics. Beautiful Evidence: Metric Monitoring - http://blog.jinspired.com/?p=33 11. Updates to the console in terms of views and analysis inspections work we previously exported repository snapshots. One current limitation is that we do not support the direct manipulation (setting of attribute values) of each of the underlying repositories. This is likely to change early in the new year. Kind regards, William Louth JXInsight Product Architect JInspired "Java EE tuning, testing, tracing, and monitoring with JXInsight" http://www.jinspired.com