Another possible reason for a production cold backup (from a presentation by an Oak Table member) - known corruption of all archived redo logs due to an Oracle bug. The worst part was having an Oracle patch that could fix the redo log corruption, but damagement would not let the dot com store be down for the few minutes needed to apply the patch. Murphy's law provided an opportunity for the database to crash and cost a company $,$$$,$$$. Have Fun :) -- //www.freelists.org/webpage/oracle-l