Server configuration question

[groundclutter]

Here is our configuration...the first server (server123) 'points' to production database (Oracle 8) while the second server (server 456) 'points' to our test server.

Input server123
Output server123
server123.aps server123:6412 Crystal APS
server123.cacheserver server123Crystal Cache Server server123.eventserver server123Crystal Event Server server123.pageserver server123Crystal Page Server server123.reportserver server123Crystal Job Server server123.wcs server123Crystal Web Component Server

server456.cacheserver server456 Crystal Cache Server server456.eventserver server456 Crystal Event Server server456.pageserver server456 Crystal Page Server server456.reportserver server456 Crystal Job Server server456.wcs server456 Crystal Web Component Server

The problem is that even though the data source for reports stored in ePortfolio is pointing at the production database region, periodically (and no rhyme or reason as to the timing), the production reports seem to be trying to access the test server.

An example is as follows:

Status: Failed

External Destination: None

Start Time: 08/13/04 08:07:41

End Time: 08/13/04 08:09:42

Server Used: server456.reportserver

Error Message: File Repository Server Input is down.


We should see:


Status: Success

External Destination: None

Start Time: 08/13/04 05:45:48

End Time: 08/13/04 05:47:04

Server Used: server123.reportserver

Instance Location: frs://Output/a_220/044/004/273628/~ce42a837f405d45860a.rpt


Any thoughts as to where the configuration is screwed up that allows this to happen? The only method that I have found to fix the problem is to go into the CCM and not only disable but stop the test servers. Then, it seems to work again...but once we turn the test servers back on...at some point, not necessarily right away, we encounter these errors again (usually the following morning).

Any ideas?

 

We ran into this same scenario with our CE9 test server install. Our DBA actually copied our production APS when I requested a new instance (Oracle) for our test machine. When we went through the install, the test server accessing the copy of the production APS wrote registry keys pointing to the production server. Those services were showing up every time the services were started on the test machine. We uninstalled and reinstalled CE9, dropped all the tables in the APS, and had the same problem. The CE9 uninstall does not clean up the registry. After calling Business Objects and having them tell me that it could not be a problem with the registry settings, I went in the registry and found the problem. We manually deleted all the Crystal related keys in the registry and reinstalled CR9 and CE9. We haven't had the same issue yet and both systems are running. Has anyone ever gotten any real help from Business Objects support? Does anyone use Elite level support? And if so, is it worth it?