Greetings! On my Debian system, several volatile directories such as
/tmp are automatically cleaned on reboot. My guess would be that
you're solution would be fine, therefore.
Take care,
"Bill & Chris Shell" <[EMAIL PROTECTED]> writes:
> Hi all,
> I have been experiencing a problem with PostgreSQL, which is caused by the
> improper system shutdown. Occasionally, my CLX node will have a power
> interruption which causes the system to restart. (I know, I should get an
> UPS.) When the system restarts, Postgres fails to start. I found the
> problem to be an artifact in /tmp area with filename of /tmp/.s.PGSQL.####.
> The #### represents a three or four digit number that changes. To remedy
> this I changed the start case in /etc/rc.d/postgres to the following.
>
> start)
> echo -n "Starting service postgres"
> ## Start daemon with startproc(8). If this fails
> ## the echo return value is set appropriate.
> # added line to delete temporary files from
> # improper shutdown
> rm /tmp/.s.PGSQL.*
> #
> su - postgres -c "/sbin/startproc -l $LOGFILE
> $H -i -o -F -D$DATADIR"\
> || return=$rc_failed
> echo -e "$return"
> ;;
>
> The change enables postgres to start every time without any apparent
> problems. Does anyone know of any reason I should not delete the file? Or,
> does anyone know of a better way of handling the postgres startup problem?
>
> TU es 73, Bill
> N6WS
> [EMAIL PROTECTED]
--
Camm Maguire [EMAIL PROTECTED]
==========================================================================
"The earth is but one country, and mankind its citizens." -- Baha'u'llah
