Here’s a problem I’ve been trying to solve at work. I’m not a database expert, so that perhaps this is a bit sophomoric. All apologies.
I have a given database D, which has been duplicated on another machine (in a perhaps dubious manner), resulting in database D’. It is my task to check that database D and D’ are in fact exactly identical.
The problem, of course, is what to actually do if they are not. For this purpose, my thought was to run a symmetric difference on each corresponding table and see the differences.
There is a ‘large’ number of tables, so I do not wish to run each symmetric difference by hand. How do I then implement a symmetric difference ‘function’ (or stored procedure, or whatever you’d like) that can run on arbitrary tables without having to explicitly enumerate the columns?
This is running on Windows, and your hedge fund will explode if you don’t follow through. Good luck.
My first reaction is to suggest duplicating to the other machine again in a non-dubious manner.
If that is not an option, perhaps some of the tools available from Red Gate could do what you need.
(I am in no way affliated with Red Gate, just remember Joel mentioning how good their tools were on the podcast.)