We upgraded to Hyperion Planning 11.1.2.4.000 and moved to new servers running on Windows 2012R2 from 11.1.2.3.500 on Windows 2008R2. The Planning applications on both environments are essentially the same (we migrated just recently) except that the Essbase application in the new environment was allocated more index and data caches as the new servers have more memory.
Everything seems to be working fine – calculation times and data loads are faster. However, we have been experiencing planning database refresh failures which results to data corruption. According to Oracle Support, the xcp files shows that users are accessing the application while the refresh is going on. I am almost certain that my users’ work habits have not changed and this has been on going even in 11.1.2.3.500. Except with a couple of refresh failures which was attributed to a small swap file and the antivirus, we have not had any other refresh failures in 11.1.2.3.500.
To prevent the failures, Oracle Support recommended locking out users during the refresh by “alter application appname disable connects” and “alter system logout session on database appname.DBname” but these only disables Essbase access. To disable Planning access, I also had to set the appropriate “Application Maintenance Mode”.
Is there one way of preventing application access (both through planning and Essbase) and still be able to perform a planning refresh or Essbase database restructure?
Any thoughts on why users accessing the application in 11.1.2.4.00 are causing failures and not in 11.1.2.3.500?
Thanks,