The Busy Java Developer's Guide to Monitoring
Crashes? Outages? Slow response? We all know that it's never your code that causes these things, but for some reason those pesky system administrators still insist on paging you at 4AM to come in and fix those problems, anyway. For some reason, they just keep expecting you to support this thing, even after QA said it was OK!
In this presentation, we'll talk about some of the reasons why monitoring in a production application is so important, and why it's not just a system administrator concern. Next, we'll talk about some simple monitoring facilities that don't require any new APIs or new libraries, yet still deliver some useful monitoring capabilities to developers, sysadmins and management. Next, we'll look at using JMX tools to keep an eye on what's going on inside of the JVM itself and/or your applications server, then how to use JMX to create your own points of monitoring interest. Finally, we'll look at how system administrators can build their own monitoring tools by using languages like Groovy and/or JRuby as the ultimate JMX client. And, for the really hard-core, we'll even take a look at how to use tools like JVMTI and/or java.lang.instrument to write custom monitoring agents.
