Pivotal Cloud Foundry APM agents are not displaying all instances.

Document ID : KB000007213
Last Modified Date : 14/02/2018
Show Technical Document Details
Issue:

 APM agents running Pivotal Cloud Foundry not showing all instances. 

 The following error were showing in Collector IntroscopeEnterpriseManager.logs:

 com.wily.isengard.postofficehub.link.v1.CacheCorruptionException: Internal cache is  corrupt. Cannot determine class type for Object

 [ERROR] [Dispatcher 1] [Manager.IncomingMessageDeliveryTask] Cannot deserialize  message while reading from node at: Socket Transport connected with Socket[addr=zsdsdsdfv0.distribution.edf.fr/43333435,port=52098,localport=5001]
com.wily.isengard.postofficehub.link.v1.CacheCorruptionException: Internal cache is corrupt. Cannot determine class type for Object 1. A prior class deserialization 

Environment:
All supported APM releases.
Cause:

 Cache corruption can be caused by problems during queries, particularly under high query load. When this happens, it usually means the internal cache is corrupt on a collector(s), and needs clearing.

Resolution:

 Clear the temporary cache files on Standalone Enterprise Manager (EM) or MOM and Collectors if this is a cluster environment:

- Stop the EM or EM cluster
- Delete the cache files by removing the following files on all of the EM's:

<EM home>/product/enterprisemanager/configuration/org*    (note there are 4 directories, DO NOT delete config.ini file)
<EM home>/work/*.*

-Then restart the EM or EM cluster, wait for a while, and verify whether this message is still occurring in the logs.

Additional Information:

Corrupted cached files can cause different issues.