Re: take my index, please
| От | Tom Lane |
|---|---|
| Тема | Re: take my index, please |
| Дата | |
| Msg-id | 10085.988603130@sss.pgh.pa.us обсуждение исходный текст |
| Ответ на | take my index, please (Lyn A Headley <laheadle@cs.uchicago.edu>) |
| Список | pgsql-novice |
Lyn A Headley <laheadle@cs.uchicago.edu> writes:
> Table "allocation"
> Attribute | Type | Modifier
> ----------------+--------------------------+----------
> allocation_oid | bigint | not null
> webco=# explain select * from allocation where allocation_oid = 5;
> NOTICE: QUERY PLAN:
> Seq Scan on allocation (cost=0.00..2295.79 rows=1 width=104)
Try
select * from allocation where allocation_oid = 5::int8;
As you wrote it, it's a cross-datatype operator (int8 vs int4)
and the planner is not very smart about those at the moment.
You should probably also ask yourself whether you *really* need int8
for this column, at least for the next release or two that it's likely
to be before something is done about this problem.
regards, tom lane
В списке pgsql-novice по дате отправления: