Stats Aggregation process not working due to Separate JVM not starting successfully.

Document ID : KB000005427
Last Modified Date : 14/02/2018
Show Technical Document Details
Issue:

 Stats aggregation failing to start due to a JVM error.

 [ERROR] [DailyAggregation.Thread1] [Manager.com.timestock.tess.services.processors.StatsAggregator] Separate JVM for daily stats aggregation failed to start

Environment:
All Supported APM Releases.
Cause:

 This error may appear in Enterprise Manager logs and defects aggregation may not occur.

Resolution:

 tess-default.properties or tess-customer.properties (which overrides tess-default.properties) can enable running defect aggregation in a separate JVM instance.

This error is a result of allocating inadequate resources to this JVM.

If you are seeing this error, do the following:

  1. Switch to using a 64-bit operating system, and allocate 8Gb heap for the affected Enterprise Manager,
  2. After doing the above, change this property to 'false' in tess-default.properties or tess-customer.properties

    • dailystats.aggregateInSeparateJvm=false
Additional Information:

 The APM Configuration and Administration Guide recommends that tess-default.properties and tess-customer.properties be identical across all Enterprise Managers in a cluster, including the MOM. All EMs will need to be restarted for any change to take effect.

 If you have a 32-bit operating system, keep this property set to 'true' and allocate more heap to the JVM that runs the defect aggregation service.

 To increase the amount of heap allocated to the JVM running the defect aggregation service, edit tess-default.properties or tess-customer.properties.

Locate the entry:

daily.jvmArgs= -XX:+CMSClassUnloadingEnabled -XX:+UseConcMarkSweepGC -XX:+UseBiasedLocking -XX:SurvivorRatio=8 -XX:TargetSurvivorRatio=90 -Dlog4j.configuration=log4j.properties -Xms256M -XX:ThreadStackSize=128

 Increase the heap by altering the setting: -Xms256M to a larger value.

 After copying the modified .properties file to each EM, then restart all of the EMs.