Re: Patch to implement pg_current_logfile() function

Поиск
Список
Период
Сортировка
От Karl O. Pinc
Тема Re: Patch to implement pg_current_logfile() function
Дата
Msg-id 20161127173707.0481a361@slate.meme.com
обсуждение исходный текст
Ответ на Re: Patch to implement pg_current_logfile() function  (Gilles Darold <gilles.darold@dalibo.com>)
Список pgsql-hackers
On Sun, 27 Nov 2016 21:54:46 +0100
Gilles Darold <gilles.darold@dalibo.com> wrote:

> Again, patches patch_pg_current_logfile-v14.diff.doc_linux_default-v2
> have not been included because I don't see any reason to talk
> especially about systemd. If you talk about systemd you must talk
> about other stderr handler by all systems. IMO saying that
> current_logfile is present only if logging_collector is enabled and
> log_destination include stderr or/and csvlog is enough, no need to
> talk about systemd and behavior of Linux distributions.

Fair enough.  And I'd sooner not talk about systemd or other such
specifics too.

The concern I'm attempting to address is that the patch touts
the current_logfiles file in the section on logging without
reservation.  But anyone compiling from source or using most
pre-built Linux binaries won't have the file. (And the
pg_current_logfiles() function won't work either.)

Maybe the answer is to  change
   <para>When logs are written to the file-system ...

to
   <para>When the <productname>PostgreSQL</productname> backend   database server write logs to the file-system (which
isoften   not the default configuration) ...
 

?

Or something.  This also seems verbose, yet incomplete because
although it mentions that the default is to not have the backend
write logs it does not say anything about where to look to change
the default.

The thing is, on most default setups you do get logs written to the
filesystem, but they are not written by the PG backend.

I'll let you (or anyone else who might be concerned) move this forward
because I don't seem to have a good answer.

Regards,

Karl <kop@meme.com>
Free Software:  "You don't pay back, you pay forward."                -- Robert A. Heinlein



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Robert Haas
Дата:
Сообщение: Re: UNDO and in-place update
Следующее
От: Robert Haas
Дата:
Сообщение: Re: Autovacuum breakage from a734fd5d1