Guido Neitzer wrote:
Perhaps you have to do some tricks to tell the script which tables or columns should be equal and which are allowed to differ, but as far as I can see, it shouldn't be that hard. At all, it took me about a day to verify the db contents.
I'm not too fussed about a row-by-row comparison between the source and the copy. It's rather a case of a tool to check the datafiles' integrity (such as fsck, myisamchk, svnadmin verify etc). If the fact that pg_dumpall returned successfully, then i would hope that all the data is present and correct. - Rich Doughty
