Ardian Xharra skrev: > *From:* Anoo Sivadasan Pillai <mailto:aspillai@xxxxxxxxx> >> I am not using any sequences, The following batch can reproduce the >> behaviour. >> CREATE TABLE master ( m1 INT primary key , m2 int unique ) ; >> INSERT INTO master VALUES ( 1, 1 ) ; >> INSERT INTO master VALUES ( 2, 2) ; >> UPDATE master SET m2 = m2 + 1; > It's normal behaviour, because after the first update it will be 2 same > values for m2 and you don't want that since you have a unique constraint > for that column. Please note: This is a bug in Postgresql, not "normal behaviour". From a conceptual perspective, there is no "after the first update" - the statement is supposed to be atomic. Unfortunately, the problem is waiting for someone to get a great idea: http://svr5.postgresql.org/pgsql-bugs/2007-02/msg00075.php If you can't wait, you are probably better off working around the problem. Standard solution is to do: UPDATE master SET m2 = -m2; UPDATE master SET m2 = -m2+1; or something similar. Nis ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
