> i've not had this problem, and my DB is about 60Gb with many, many FK's. > > aren't FK's (rather, *all*) triggers disabled on the replica already?
I don't know why you didn't have this problem, maybe you don't have indexes on the tables which are parents of foreign keys. Or the parent tables are small tables in your schema... but my schema has a few huge tables which are parents of foreign keys, and have lots of indexes. If the truncate fails on those tables, and the indexes are not dropped, then both the deletion and the copy will be an order of magnitude slower than when I drop first the foreign keys (and then truncate succeeds and the indexes stay dropped during the copy). Cheers, Csaba. _______________________________________________ Slony1-general mailing list [email protected] http://gborg.postgresql.org/mailman/listinfo/slony1-general
