search cancel

MOM log filling with messages - Exception when processing Central Fast Correlation incubator

book

Article ID: 8762

calendar_today

Updated On:

Products

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

Issue/Introduction

We are using EM cluster on Introscope Enterprise Manager Release 10.5.1 The MOM is getting a lot of log messages of the following format

20171107 104857 GMT [WARN] [Manager.AppMap] Exception when processing Central Fast Correlation incubator
java.lang.NullPointerException
at com.google.common.base.Preconditions.checkNotNull(Preconditions.java:210)
at com.google.common.net.HostAndPort.fromString(HostAndPort.java:168)
at com.wily.introscope.appmap.em.ontology.processing.ProxyGraphProcessor.extractRemotePairs(ProxyGraphProcessor.java:288)
at com.wily.introscope.appmap.em.ontology.processing.ProxyGraphProcessor.processGraph(ProxyGraphProcessor.java:215)
at com.wily.introscope.appmap.em.ontology.processing.GraphProcessorManagerImpl.processGraph(GraphProcessorManagerImpl.java:53)
at com.wily.introscope.appmap.events.consumers.CrossClusterCorrelationEngine$Consumer.processCorrelatedVertices(CrossClusterCorrelationEngine.java:293)
at com.ca.apm.transactiontrace.appmap.mapper.arrivalbuffer.agc.CentralCorrelatingCollectionBuffer.processCorrelatedTraces(CentralCorrelatingCollectionBuffer.java:215)
at com.ca.apm.transactiontrace.appmap.mapper.arrivalbuffer.agc.CentralFastIncubationBuffer.process(CentralFastIncubationBuffer.java:62)
at com.ca.apm.transactiontrace.appmap.mapper.arrivalbuffer.AbstractTransactionTraceIncubator$QueueConsumer.run(AbstractTransactionTraceIncubator.java:215)
at java.lang.Thread.run(Unknown Source)

and

20171107 104855 GMT [WARN] [Manager.AppMap] Exception when processing Central Slow Correlation incubator
java.lang.NullPointerException
at com.google.common.base.Preconditions.checkNotNull(Preconditions.java:210)
at com.google.common.net.HostAndPort.fromString(HostAndPort.java:168)
at com.wily.introscope.appmap.em.ontology.processing.ProxyGraphProcessor.extractRemotePairs(ProxyGraphProcessor.java:288)
at com.wily.introscope.appmap.em.ontology.processing.ProxyGraphProcessor.processGraph(ProxyGraphProcessor.java:215)
at com.wily.introscope.appmap.em.ontology.processing.GraphProcessorManagerImpl.processGraph(GraphProcessorManagerImpl.java:53)
at com.wily.introscope.appmap.events.consumers.CrossClusterCorrelationEngine$Consumer.processCorrelatedVertices(CrossClusterCorrelationEngine.java:293)
at com.ca.apm.transactiontrace.appmap.mapper.arrivalbuffer.agc.CentralCorrelatingCollectionBuffer.processCorrelatedTracesFinal(CentralCorrelatingCollectionBuffer.java:229)
at com.ca.apm.transactiontrace.appmap.mapper.arrivalbuffer.agc.CentralSlowIncubationBuffer.process(CentralSlowIncubationBuffer.java:61)
at com.ca.apm.transactiontrace.appmap.mapper.arrivalbuffer.AbstractTransactionTraceIncubator$QueueConsumer.run(AbstractTransactionTraceIncubator.java:215)
at java.lang.Thread.run(Unknown Source)

As far as we can see the system is working normally but weare concerned about the flooding of log messages as we have already received disk space alerts due to the excessive log sizes.

Cause

The message is caused by an unhandled condition when processing traces with Unknown Called URLs - this is a valid situation and should not cause an exception

Environment

APM 10.5

Resolution

The problem is being addressed in APM 10.7 and this situation will no long cause a NullPointerException.

To prevent the flooding of messages the following configuration setting can be made in the IntroscopeEnterpriseManager.properties file on the MOM

log4j.logger.Manager.AppMap=ERROR,console,logfile

 

This is a dynamic property and does not require an EM restart.

The affect of this change is that only log messages with severity of ERROR or higher will be reported by the Manager.AppMap component - other logging will not be affected.

Additional Information

00886520