Ok, now that we're on the same page of music: the current nspostgres CVS on the AOLserver sourceforge site does not have Randy Kunkee's work in it. It has a complete rework of DbFail() rather than the simpler solution Randy had found.
I'm going to go over this; it creates a certain amount of redundancy to do it the way it's done now. So, I'm going to apply Randy's patch to my local copy of nspostgres and see what the differences are. Will make a report when I see what I need to see (which may be a few days, unless someone else wants to take it upon themselves to do it).
