On Fri, Feb 10, 2006 at 08:10:36PM -0800, Jean-Christophe Roux wrote: > here are the structures of the table involved: I couldn't duplicate the problem in 8.1.2 or 8.0.6 with the code you posted. I created the given tables and rules (plus guesses for the administrators and status_list tables), then inserted several records into topics, then inserted a few records into releases, then inserted a few more records into topics. The topics_id_seq sequence incremented by one each time with no gaps; that probably means my test didn't match exactly what you're doing. Does the topics table have any rules or triggers that you didn't show? Could you post a minimal but complete test case, i.e, all SQL statements that somebody could load into an empty database to reproduce the problem? If not then it might be useful to see the EXPLAIN ANALYZE output of a series of statements that exhibit the unexpected behavior. What version of PostgreSQL are you running? -- Michael Fuhr
