Troubleshooting Steps to identify the cause of java.lang.NoClassDefFoundError exception in a Private Cloud

Document created by Ramya_Nakkapalle Employee on Jul 16, 2015Last modified by jon_brams809399 on Oct 17, 2016
Version 3Show Document
  • View in full screen mode
Process failing with NoClassDefFound errors frequently and a restart of the cloud nodes temporarily resolves the issue, but it continues to return again after some time.
Monitoring the cloud runtime environment for availability, performance and scalability should be helpful to diagnose the cause


1.Cluster level monitoring
Monitor Cluster Status panel in Atom Management or the Molecule node “views” files for node health and inter-node communication issues.

More details at this link


2.Server level monitoring
Servers would need to be monitored for obtaining and trending following metrics,especially Disk I/O wait.Trend the data over time.
•    Availability
•    CPU utilization
•    CPU load
•    Disk space free
•    Memory free
•    Disk I/O wait
•    atom.exe service running


3.Disk space monitoring for accounts
  • Set account size limit for account via cloud management/attachment quota.
  • Set properties on the selected cloud via Atom management Cloud properties.
  • Under Advanced set the following properties
            1) Disk Space Monitoring (True)
            2) Disk Space Monitoring Frequency (in milliseconds):  The value needs to be decided based on the rate at which disk size has been observed to change based on metrics gathered via server monitoring.
            3) Disk Space Warning Limit (%) - Warning written to container logs when account reaches the % of disk limit assigned to that account


More details at link:


4.JMX monitoring
Also,find the attached document that describes the attributes that you can track to monitor the performance/communication issues.  Use JMX to Monitor Your System and trend the data over time.