SYMPTOM
During the start-up of one node of a multiple node Strategy Intelligence Server cluster, an error similar to the following message is logged to the DSSErrors.log:
2014-12-06 15:34:24.422+01:00 [HOST:Node1][PID:12910812][THR:141869][Document Server][Error] RWDCacheManager: /data/mstr941/Cluster/ClusterCaches/ServersNode1_P7525257011E32F6100100080632F05A0/RWDCache/61236E3C11E47D31010000803501407C_info.rwdc does not exist
2014-12-06 15:34:24.423+01:00 [HOST:Node1][PID:12910812][THR:141869][Document Server][Error] RWDCacheManager: /data/mstr941/Cluster/ClusterCaches/ServersNode1_P7525257011E32F6100100080632F05A0/RWDCache/61236E3C11E47D31010000803501407C_info.rwdc does not exist
STEPS TO REPRODUCE1. Setup a 2-node Strategy Intelligence Server cluster and configure cache sharing.
2. Run a document from Strategy Web to create a document cache on any node.
3. Shutdown the Strategy Intelligence Server which did NOT create the document cache.
4. Restart this Strategy Intelligence Server again and errors similar to the above are logged.
CAUSEThe cache pool tables, which contain entries about the local caches and also about the foreign node's cache, do not remove the entries corresponding to the foreign node's caches at the time of shutdown. At the time of shutdown, the Strategy Intelligence Server which is being shutdown is supposed to clean up the pool tables and remove the entries corresponding to the caches on the foreign node. However, as these are not removed, at the time of start up, the Strategy Intelligence Server looks for these caches on it's local folder and since it cannot find them (as the caches actually reside on the other node's folder), the errors are seen.
ACTIONEven though error messages are displayed in the logs, no loss of functionality occurs. The error messages may be safely ignored.
