How do I determine the statistical views that are dependent on a particular table?
Here is why I need to know:
I am testing an upgrade of a DB from DB2 9.1 to 9.7. Part of the upgrade requires dropping a column from a table. Every time I do this, it makes any future attempts to create procedures referring to that table fail.
After several hours of digging recently, I have found this:
https://www-304.ibm.com/support/docv...id=swg1IZ39997
The gist of this is that the act of dropping a column from a table having dependent statistical views (which cannot be regenerated) can cause irreversible issues. The preventative measure is described this way:
If altering a base table on which statistical views are
dependent:
1. disable query optimization for all statistical views
2. perform the alterations required on the base table
3. enable query optimization for the views once again.
How, though, can I determine, based on the table name, its dependent statistical views so I know which ones to alter before I drop the column? This is pretty much over my head. I do not really understand (and am not entirely sure I want to) the ins and outs of query optimization and statistical views. I am just trying to get my DB upgraded.
I posted my initial issue (but got no response) several months ago before I finally figured out it was related to statistical views:
Drop column failure
(yes...this upgrade has been on hold for that long!)