The following bug has been logged online:

Bug reference: 2870
Logged by: Dmitry D. Khlebnikov
Email address: galaxy@openwall.com
PostgreSQL version: 8.2
Operating system: Openwall GNU/*/Linux
Description: incorrect man page for postgres
Details:

The man page for postgres(1) has a wrong description of the '-S' option.
Previously, the '-S' option means that postgres will start in the 'silent'
mode, but in version 8.2 the '-S' option describes the amount of memory for
sorts.

---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster

Dmitry D. Khlebnikov wrote:

> The man page for postgres(1) has a wrong description of the '-S' option.
> Previously, the '-S' option means that postgres will start in the 'silent'
> mode, but in version 8.2 the '-S' option describes the amount of memory for
> sorts.
>
I think you change the order, in 8.1 -S option describes the amount of
memory but in 8.2 it describes the silent mode. This is because there
was some mismatch options in postgres and postmaster. See "Make
command-line options of postmaster and postgres identical (Peter)" in
release notes [1].

[1] http://www.postgresql.org/docs/8.2/s...lease-8-2.html


--
Euler Taveira de Oliveira
http://www.timbira.com/


---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match

Euler Taveira de Oliveira <euler@timbira.com> writes:
> Dmitry D. Khlebnikov wrote:
>> The man page for postgres(1) has a wrong description of the '-S' option.
>> Previously, the '-S' option means that postgres will start in the 'silent'
>> mode, but in version 8.2 the '-S' option describes the amount of memory for
>> sorts.
>>
> I think you change the order, in 8.1 -S option describes the amount of
> memory but in 8.2 it describes the silent mode.

No, Dmitry has it right: the man page is mistaken. You can still get
silent mode but you have to use the long-form option (--silent-mode=1).

I've corrected the documentation error in CVS.

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 6: explain analyze is your friend

Euler,

On Wed, Jan 03, 2007 at 09:50:15PM -0200, Euler Taveira de Oliveira wrote:
> Dmitry D. Khlebnikov wrote:
>
> > The man page for postgres(1) has a wrong description of the '-S' option.
> > Previously, the '-S' option means that postgres will start in the 'silent'
> > mode, but in version 8.2 the '-S' option describes the amount of memory for
> > sorts.
> >
> I think you change the order, in 8.1 -S option describes the amount of
> memory but in 8.2 it describes the silent mode. This is because there

Perhaps. However, when I tried to follow the man page and run
'postgres' with '-S' it responded with:

db!pgsql:~$ postgres --version
postgres (PostgreSQL) 8.2.0
db!pgsql:~$ postgres -D /var/lib/pgsql -S
postgres: option requires an argument -- S
Try "postgres --help" for more information.
db!pgsql:~$ postgres --help | fgrep -- -S
-S WORK-MEM set amount of memory for sorts (in kB)
db!pgsql:~$ zcat /usr/share/man/man1/postgres.1.gz | fgrep -A 3 -- -S
\fB-S\fR
Specifies that the \fBpostgres\fR
process should start up in silent mode. That is, it will
disassociate from the user's (controlling) terminal, start its
db!pgsql:~$

All in all, I don't know how it was in the previous versions of
PostgreSQL (I was using 7.x versions), I just merely want to point out
that the man page provides the incorrect information wrt the '-S'
option.

--
(GM)


---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match

Tom,

On Wed, Jan 03, 2007 at 08:02:34PM -0500, Tom Lane wrote:

> No, Dmitry has it right: the man page is mistaken. You can still get
> silent mode but you have to use the long-form option (--silent-mode=1).
>
> I've corrected the documentation error in CVS.

I think that you have forgotten to regenerate man pages from SGML since
8.2.1 is out and postgresql-base-8.2.1.tar.bz2 contains the old man page
for postgres(1).

--
(GM)


---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match