Foreign key index usage [was: Re: Index lock issues]

>>>>> ""Andy" == "Andy Kent" <akent@cix.compulink.co.uk> writes: > (2) The database makes big use of foreign keys, which generate lots of > heavily-duplicate indexes all over the place. I would like to be able to > replace the generated indexes on, say, (col1,col2) with one on > (col1,col2,col3,col4) to improve their usefulness. However, OnLine > generates its own index when you add the constraint unless there's > already one with exactly the same (*only*) keys. > I can't see a way round this with 5.02. Is there one? Does 6.00 improve > things? I'm also interested in an answer to this problem. Essentially, it renders foreign keys useless. If the indexes created contain lots of duplicate entries the performance of DELETE operations is severely affected. Another problem regarding the use of declarative referential integrity is the option of deferring integrity checking until the end of a transaction. This is possible for OnLine (SET CONSTRAINTS DEFERRED). Any plans for SE yet? -- Oliver Okrongli infix Software-Systeme GmbH Phone +49 531 238090 Rebenring 33 Fax +49 531 2380935 oliver@infix.de 38106 Braunschweig F.R. Germany