Re: How to obtain the real problem that avoids postgresql - Mailing list pgsql-general

From Craig O'Shannessy
Subject Re: How to obtain the real problem that avoids postgresql
Date
Msg-id Pine.LNX.4.44.0312051507590.8177-100000@mail.undercoverwear.com.au
Whole thread Raw
In response to How to obtain the real problem that avoids postgresql service starting?  ("Netto" <rcnetto@yahoo.com>)
Responses Re: How to obtain the real problem that avoids postgresql
List pgsql-general
I think this is a pretty serious shortcoming of the standard
/etc/init.d/postgresql script, it sends all output (std. out AND error to
/dev/null).

You should have a line something like this

                su -l postgres -s /bin/sh -c "/usr/bin/pg_ctl  -D $PGDATA -p /usr/bin/postmaster -o '-p ${PGPORT}'
start > /var/log/postgres.log 2>&1" < /dev/null 

You can see where I have replaced the "> /dev/null" with "> /var/log/postgres.log"

You would want to do this first, before attempting to start postgres, so


[root@somebox /#] vi /etc/init.d/postgresql
[root@somebox /#] touch /var/log/postgres.log
[root@somebox /#] chown postgres /var/log/postgres.log
[root@somebox /#] /etc/init.d/postgresql start

Now, when it fails, you will get decent logging in /var/log/postgres.log

Hope this helps.

Craig


On Tue, 2 Dec 2003, Netto wrote:

> Hi all,...
> I'd like to know where can I get information about what is happening to
> postgresql so it can't be started...
>
> all I get is:
> Starting postgresql service:                               [FAILED]
>
> Is there any log file, or a way to start the database with a kind of -v
> (verbose) option??
> Or maybe any configuration parameter to modify on postgresql files,...
>
> Any suggestion will help a lot!
>
> That's it...
> Thanks!
> Netto
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: you can get off all lists at once with the unregister command
>     (send "unregister YourEmailAddressHere" to majordomo@postgresql.org)
>


pgsql-general by date:

Previous
From: Alex Satrapa
Date:
Subject: Re: are cursors necessary?
Next
From: Martijn van Oosterhout
Date:
Subject: Re: How to obtain the real problem that avoids postgresql