Not seeing JMX Active MQ metrics
search cancel

Not seeing JMX Active MQ metrics


Article ID: 194122


Updated On:


CA Application Performance Management Agent (APM / Wily / Introscope) CA Application Performance Management (APM / Wily / Introscope) INTROSCOPE DX Application Performance Management


Seeing an issue in Production that was fixed in the Test environment.

Below find the summary of the previous request.

introscope.agent.metricClamp from 5K to 200K. Also added the below JMX filter and Attributes but after adding the filter we are unable to find org.Apache.ActiveMQ folder under JMX folder in APM console. So we reverted this back to the original filter. Now we are not able to see all previous mbeans that were visible before this change. Also need to know whether there is an option available in APM to remove the metric clamp manually from the backend.|*|*|*|*,org.apache.activemq|brokerName=*|destinationName=*|destinationType=Queue,java.lang|type=Threading:*ThreadCount,java.lang|type=ClassLoading*:*ClassCount,java.nio|*|type=BufferPool:MemoryUsed,java.lang*type=GarbageCollector:CollectionTime


Original Filter


Application Performance Management 10.7


1) There are multiple errors as below noticed :
 [ERROR] [IntroscopeAgent.Agent] Introscope AutoProbe will not run because the provided directives files are not valid: Can't proceed because there are conflicting tracer class mappings for JMSFrontendMarker, JMSBackendMarker.

Please ask the customer to disable the default JMS bundle (jms.pbd) when including ActiveMQ bundle. This should take care of this probe.

Removing autoprobe errors
The ApacheMQ bundle that is included in the Agent package is conflicting with the JMS package, which supports AMQ
Remove the ApacheMQ bundle from the Agent package
Is there a reason to use the Agent for this application instead of

2) APM reaches a clamp and takes 2 weeks to clear. Why is this?
The only action item was how long should you expect the clamp go away.

You see from the logs the clamp goes away quickly. But the clamp still appears in the Investigator. This should not be. If needed, please open a new case with EM/Agent Logs and a screenshot showing the clamp.

When an Agent is clamped there will be messages in both the Agent and EM log files

3) Needs to know whether there is an option available in APM to remove the metric clamp manually from the backend. No there is not

4) See case description . Set JMX filters. They are not working

Case Description should have been not seeing JMX Active MQ metrics. Thats a lot more descriptive than the filter is not working. Please keep in mind for future cases

It looks like a simple filter misconfiguration done in the Agent profile.

The clamp and the filter are

This maybe a syntax issue, there are fewer pipes in the new filter for activemq than the original, 3 vs. 4
The Agent probably stopped processing when it found the bad syntax so did not show anymore MBean metrics
Try this filter, Move the activemq setting to the end and add another pipe (|*) to match the number in the original|*|*|*|*,java.lang|type=Threading:*ThreadCount,java.lang|type=ClassLoading*:*ClassCount,java.nio|*|type=BufferPool:MemoryUsed,java.lang*type=GarbageCollector:CollectionTime,org.apache.activemq|brokerName=*|destinationName=*|destinationType=Queue|*

Please ask the customer to change the below property

com.wily.use.platform.mbeanserver=true - > To com.wily.use.platform.mbeanserver=true

There is a space added to the above property by mistake.