When you are running SAS^® Size Profiling, you might see an error message similar to the following in the SAS^® Merchandise Intelligence middle-tier log:

INFO   | jvm 1    | 2009/04/15 23:12:13 | 2009-04-15 23:12:13,124 
SizeProfilingStoreGroupsGen-101927] ERROR com.sas.solutions.di.server - Error running 
stored process SizeProfilingStoreGroupsGen [0, 0,100565, 60, 1,1, 152642, 101927]
- server [ (prf) 1 [8] MIMain PRF - Logical Stored Process Server] -
"getConnection()" cannot be called on a factory after "shutdown()" 
or "destroy()" has been called.

It might be accompanied by the following message in the object spawner log:

20090415:23.12.13.11: 00000349:ERROR:   The Balance algorithm timed out before a server could be found.
20090415:23.12.13.11: 00000349:WARNING: The load balancing instance ldblCompRefDirectConnection call failed.
20090415:23.12.13.11: 00000349:WARNING: Unable to redirect the client request.

This issue occurs because the middle-tier might submit too many stored process requests for the object spawner to process before timing out.

To reduce the chances of encountering this problem, increase the load balancing availability timeout (on the MIMain PRF Stored Process Server) from the default 60000 ms to 120000 ms. Please see SAS Note 15661 "ERROR: The Balance algorithm timed out before a server could be found" for instructions on how to increase the availability timeout value.

Select the Hot Fix tab in this note to access the hot fix for this issue.

Operating System and Release Information

A fix for this issue for SAS Merchandise Intelligence 3.2 is available at: