Re: Hashagg planning bug (8.0.1)
От | Rod Taylor |
---|---|
Тема | Re: Hashagg planning bug (8.0.1) |
Дата | |
Msg-id | 1115742105.6888.87.camel@home обсуждение исходный текст |
Ответ на | Re: Hashagg planning bug (8.0.1) (Tom Lane <tgl@sss.pgh.pa.us>) |
Ответы |
Re: Hashagg planning bug (8.0.1)
|
Список | pgsql-hackers |
On Tue, 2005-05-10 at 12:11 -0400, Tom Lane wrote: > Rod Taylor <pg@rbt.ca> writes: > > It would seem that the planner does not take into account whether the > > datatypes involved have the capability to use hash aggregates or not. > > > sdb=# explain select query_start, current_query from pg_locks join stat_activity on pid = procpid where granted =true and transaction in (select transaction from pg_locks where granted = false); > > ERROR: could not find hash function for hash operator 716373 > > What's stat_activity? I thought you meant pg_stat_activity, but that > works fine here. Oh, stat_activity is a view which removes idle connections from displaying and allows non-privileged users to see everything that's going on within the DB. CREATE OR REPLACE FUNCTION stat_activity() RETURNS setof pg_stat_activity SECURITY DEFINER AS 'select * from pg_stat_activity;' language sql; CREATE OR REPLACE VIEW stat_activity ASSELECT stat_activity.procpid, stat_activity.usename, stat_activity.query_start::timestamp(0) without time zone AS query_start, stat_activity.current_query FROM stat_activity() WHERE stat_activity.current_query <> '<IDLE>'::text; --
В списке pgsql-hackers по дате отправления: