On 2014-05-08 22:21:43 +0900, Fujii Masao wrote:
> On Wed, May 7, 2014 at 4:57 PM, Amit Kapila <amit.kapila16@gmail.com> wrote:
> > On Tue, May 6, 2014 at 7:04 PM, Christoph Berg <cb@df7cb.de> wrote:
> >> Hi,
> >>
> >> if you split configuration and data by setting data_directory,
> >> postgresql.auto.conf is writen to the data directory
> >> (/var/lib/postgresql/9.4/main in Debian), but tried to be read from
> >> the etc directory (/etc/postgresql/9.4/main).
> >>
> >> One place to fix it would be in ProcessConfigFile in
> >> src/backend/utils/misc/guc-file.l:162 by always setting
> >> CallingFileName = NULL in src/backend/utils/misc/guc-file.l:162, but
> >> that breaks later in AbsoluteConfigLocation() when data_directory is
> >> NULL. (As the comment in ProcessConfigFile says.)
> >
> > This problem occurs because we don't have the value of data_directory
> > set in postgresql.conf by the time we want to parse .auto.conf file
> > during server start. The value of data_directory is only available after
> > processing of config files. To fix it, we need to store the value of
> > data_directory during parse of postgresql.conf file so that we can use it
> > till data_directory is actually set. Attached patch fixes the problem.
> > Could you please once confirm if it fixes the problem in your
> > env./scenario.
>
> Maybe this is nitpicking, but what happens when postgresql.auto.conf also
> includes the setting of data_directory? This is possible because we can
> set data_directory via ALTER SYSTEM now. Should we just ignore such
> problematic setting in postgresql.auto.conf with warning message?
I think that's a case of "Doctor, it hurts when I do this. Doctor: don't
do that then".
Greetings,
Andres Freund
-- Andres Freund http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services