Unable to logon to the P8 Domain when attempting to connect to node 2 of a IBM FileNet Content Engine (CE), deployed as a WebSphere cluster, from FileNet Enterprise Manager (FEM). Reciev an error, GCD configuration is invalid. The ServerInstance object does not exist”
Attempt to connect to node 2 of a CE WebSphere (WAS) cluster, get error popup stating Unable to logon to the P8 Domain, clicking on Details gives"
GCD configuration is invalid. The ServerInstance object does not exist
- Connecting from node 1 of the CE cluster works fine.
- The CE server on node 2
|2010-04-12T19:27:52.986Z 1A451A45 ENG FNRCE0306E - ERROR method name: getUserName principal name: ceAdmin Global Transaction: false User Transaction: false Exception Info: An unexpected exception was detected.|
com.filenet.api.exception.EngineRuntimeException: E_UNEXPECTED_EXCEPTION: An unexpected exception was detected.
java.lang.NoClassDefFoundError: com.filenet.engine.security.ProviderCache (initialization failure)
.. 43 more
The error, GCD configuration is invalid. The ServerInstance object does not exist, is normally given when the application server within the cluster is not written into the GCD (Global Configuration Data). However, when WAS server restarts, it should add any new server instances to the GCD during startup.
A full restart of WAS server needs to be done, since the CE application had been deployed to the cluster as outlined in the Installation and Upgrade Guide.
WebSphere Network Deployment (ND) CE cluster
Resolving The Problem
1. After deployment of the CE application to the WAS cluster make sure to do a full restart of WAS, including the deployment manager (dmgr), and all the nodes, as outlined in the Installation and Upgrade guide.
2. Restart the WebSphere application server where Content Engine is deployed, as follows in step 6 of the P8 Platform Installation Upgrade Guide:
|Stand-alone server||Stop and start the application server.|
|Network Deployment||Stop and start the application server where Content Engine is deployed, including the Deployment Manager and all managed nodes.|
|Cluster||Stop and start the cluster.|
17 June 2018