Re: BUG #4941: pg_stat_statements crash

Поиск
Список
Период
Сортировка
От Tom Lane
Тема Re: BUG #4941: pg_stat_statements crash
Дата
Msg-id 21301.1248661698@sss.pgh.pa.us
обсуждение исходный текст
Ответ на Re: BUG #4941: pg_stat_statements crash  (Itagaki Takahiro <itagaki.takahiro@oss.ntt.co.jp>)
Ответы Re: BUG #4941: pg_stat_statements crash
Список pgsql-bugs
Itagaki Takahiro <itagaki.takahiro@oss.ntt.co.jp> writes:
> We should call [Read dumpfile] routine only once even on Windows.

Seems to me that you should simply do the load only when found is false.

> How about executing the routine during AddinShmemInitLock is taken?

You should not hold AddinShmemInitLock any longer than really necessary.
I don't think there is a need for locking here anyway, though, since
found should be false only in the postmaster.

> The best solution might be to call shmem_startup_hook only once
> every platforms, but it is difficult without fork().

We're not changing that.  This is a bug in pgss_shmem_startup.

            regards, tom lane

В списке pgsql-bugs по дате отправления: