Hi. I'm seeing the following errors in the splunk4jmx.log on the server. What's the best way to track down the attributes that are causing this error?
2011-10-31 00:01:00,345 ERROR [Thread-1] Error : No such attribute: committed
2011-10-31 00:01:00,348 ERROR [Thread-2] Error : No such attribute: init
2011-10-31 00:01:00,349 ERROR [Thread-1] Error : No such attribute: init
2011-10-31 00:01:00,429 ERROR [Thread-3] Error : No such attribute: value
2011-10-31 00:01:00,432 ERROR [Thread-3] Error : No such attribute: used
2011-10-31 00:01:00,433 ERROR [Thread-4] Error : No such attribute: value
2011-10-31 00:01:00,435 ERROR [Thread-4] Error : No such attribute: used
2011-10-31 00:01:00,439 ERROR [Thread-3] Error : No such attribute: value
2011-10-31 00:01:00,441 ERROR [Thread-3] Error : No such attribute: max
The config file I'm using is the original config updated to include 4 servers. All the demo dashboards are working but these errors are using up a lot of space.
Those messages might be coming from a slightly older version of SPLUNK4JMX, and in particular, attributes declared in these definitions :
<mbean domain="java.lang" properties="type=GarbageCollector,name=ParNew">
<mbean domain="java.lang" properties="type=MemoryPool,name=CMS Old Gen">
You could actually delete these elements from the default config.xml file, and it won't affect the demo dashboards.
These aren't really errors per say , but rather certain Mbean attributes aren't present or set in your target JVM.
The default config file is only really supposed to be a guide/example of various scenarios (using simple/composite/tabular attributes) for you to then go and configure your own specific environment.ie: Different JVMs will have different Garbage Collectors, Memory Pool Names , ergo , different java.lang MBeans and attributes available.
Best thing to do is to use a tool like JConsole to connect to your JVMs, browse the available MBeans and then make your config.xml file more specific to your particular JVM environment.