Re: Possible bug in update?
От | Markus Kräutner |
---|---|
Тема | Re: Possible bug in update? |
Дата | |
Msg-id | 200308281228.30356.kraeutne@in.tum.de обсуждение исходный текст |
Ответ на | Re: Possible bug in update? (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-hackers |
> > At the end of several bulk inserts the following statements get executed: > > > > begin > > update pom_c2996785 set p3=3268837857008 where p3=34804837857008 > > update pom_c2996795 set p3=3268837857008 where p3=34804837857008 > > update pom_c2996804 set p3=3268837857008 where p3=34804837857008 > > commit > > > > When I check the database the last update has not been executed. Manually > > executing it in psql does work. > > ... ergo, the bug is not in the backend, but in whatever non-psql client > software you are using. Since you haven't mentioned what that is, > further speculation is difficult ... I managed to construct a sample script which also fails in psql. It seems to be the hash index. If I leave out the "create index" line or use btree instead everything works fine. I will use btree now :) Thank you for your help. Markus
В списке pgsql-hackers по дате отправления: