Better way to monitor JVM containers: PingJMX

Discussions

News: Better way to monitor JVM containers: PingJMX

  1. Better way to monitor JVM containers: PingJMX (2 messages)

    Typically JVMs are monitored by using remote JMX monitoring tools. There are signficant problems wth this approach. Examples: how do you monitor a farm of JVMs without having and administrative headache of setting up remote JMX configurations, ports, SSL, etc etc. With all the talk about cyber security, JMX remote connectivity opens up a way for hackers to exploit these administrative ports.

    PingJMX, open source JMX streaming framework, allows a developer accomplish just that. PingJMX allows developers to stream JMX metrics form JVM out to the central location, or any other destionation.

    For more information see:https://nastel.github.io/PingJMX/

    Here are some of the features of PingJMX:

    • Periodic JVM heartbeat* Monitor memory utilization, GC activity, memory leaks
    • High/Low, normal vs. abnormal CPU usage
    • Monitor threading, runtime and other JVM performance metrics
    • Monitor standard and custom MBean attributes
    • Conditional actions based on MBean attribute values
    • Conditional streaming based on custom filters* Application state dumps on VM shutdown for diagnostics

    PingJMX project is opne source and released under Apache 2 license.

    Threaded Messages (2)

  2. Or Jolokia[ Go to top ]

    For JVMs running servlet containers, Jolokia is a good HTTP-based alternative to JMX' native APIs.

  3. Interesting[ Go to top ]

    Looks like a cool project, but it is still based on request/reply type approach where clients connect and issue requets over JSON/HTTP bridge. PingJMX is somewhat different in that it is pushing metrics out to some destination. Both approaches are complementary, in my view, depending on what you are trying to do.

    In some cases remote access is need, yet others it is preferable to push/stream metrics out.