RE: how do I set DateStyle = European in postmaster ? - Mailing list pgsql-admin

From Gary Stainburn
Subject RE: how do I set DateStyle = European in postmaster ?
Date
Msg-id 01BFE7F9.30E5C220@garys.ringways.co.uk
Whole thread Raw
In response to how do I set DateStyle = European in postmaster ?  (Michael Holopainen <michael@laserle.fi>)
List pgsql-admin
-----Original Message-----
From:    Michael Holopainen [SMTP:michael@laserle.fi]
Sent:    Friday, July 07, 2000 7:27 AM
To:    Gary Stainburn
Subject:    Re: [ADMIN] how do I set DateStyle = European in postmaster ?

>Thank you for your help, but
>what is correct postmaster options syntax in (linux) startup script?
>
>Here are the variations I have seen
>
>-oe
>-o -e
>'-o -e'
>-o "-e"
>"-oe"
>
>atleast -oe and -o -e do not work. So I suspect it is -o "-e"
>Am I correct ?

Choices 1 and 5 are the same and wrong.
Choices 2 and 4 are the same and right.
Choice 2 is the one that I use.

Most of this formatting is down to the shell and not postmaster, but
what you should be telling postmaster with the '-o' argument is to pass
the next argument '-e' to the backend process.

What happens when you make a connection to the postmaster, either by
starting psql, DBI, pgaccess, odbc connection etc, is that postmaster
forks another process to actually do the work.  That is why you need to
do it like this.

With choice 1, you have just one argument.  With choice 3 you still only
have one argument because you have enclosed it inside quotes.  When the
shell interprets the line it uses spaces to split the arguments up.
Quotes override this behaviour by keeping everything inside them -
including the spaces - as one argument.
This is why choice 1 = 5 and 2 = 3.

Can you post the script that you use to start the postmaster and I will
have a look at it.  Also, it is always a good idea to Cc: the mailing
list with any follow-ups because you then get other people seeing the
message.

>
>(postgres is in use so, I can't just restart it when I want. So it would
>be good to KNOW excatly what to do, coz when I stop it, I do not have
>time to play with it [I'm sure you know the feeling...
>"can_i_use_the_db_yet, (30 sec later)can_i_use_the_db_yet, etc. "] )

I know exactly what you mean, and to make it worse, I have a 'real-time'
feed from a logger that *expects* it to be there 24x7 and allows very
little latitude - it has enough buffer for a few minutes only, just
about enough time to reboot the PC.

>
>-michael (A)
[snip]

-----------------------------------------
Gary Stainburn.
Work: http://www.ringways.co.uk mailto:gary.stainburn@ringways.co.uk
REVCOM: http://www.revcom.org.uk mailto:gary.stainburn@revcom.org.uk
-----------------------------------------
Murphy's Laws: (326) If it jams - force it. If it breaks, it needed replacing anyway.
-----------------------------------------



pgsql-admin by date:

Previous
From: kumar1@home.com (Prasanth A. Kumar)
Date:
Subject: Re: Memory management
Next
From: Steve Sargent
Date:
Subject: Cusor handling