syslog-ng 3.0.1 restart script

[munthajeeb]

Hi,

I upgraded to syslog-ng 3.0.1 on Linux. But the restart script stops it but is unable to start
at times. Is there some change I need to do the startup scripts ? is there a new version for syslog-ng 3.0 ? The script is given below

-------------------------------------------
INIT_PROG="/usr/local/my/sbin/syslog-ng"    # Full path to daemon
INIT_OPTS="-f /usr/local/my/etc/syslog-ng.conf"                      # options passed to daemon

PATH=/bin:/sbin:/usr/bin:/usr/sbin:/usr/local/my/bin:/usr/local/my/sbin

INIT_NAME=`basename "$INIT_PROG"`

# Source Redhat function library.
#
. /etc/rc.d/init.d/functions


RETVAL=0

umask 077
ulimit -c 0

# See how we were called.
case "$1" in
  start)
        echo -n "Starting $INIT_NAME: "
        daemon --check $INIT_PROG "$INIT_PROG $INIT_OPTS"
        RETVAL=$?
        echo -n "Starting Kernel Logger: "
        [ -x "/sbin/klogd" ] && daemon klogd
        echo
        [ $RETVAL -eq 0 ] && touch "/var/lock/subsys/${INIT_NAME}"
        ;;
  stop)
        echo -n "Stopping $INIT_NAME: "
        killproc $INIT_PROG
        RETVAL=$?
        echo -n "Stopping Kernel Logger: "
        [ -x "/sbin/klogd" ] && killproc klogd
        echo
        [ $RETVAL -eq 0 ] && rm -f "/var/lock/subsys/${INIT_NAME}"
        ;;
  status)
        status $INIT_PROG
        RETVAL=$?
        ;;
  restart|reload)
        $0 stop
        $0 start
        RETVAL=$?
        ;;
  *)
        echo "Usage: $0 {start|stop|status|restart|reload}"
        exit 1
esac

exit $RETVAL
--------------------------

[mutex]

is there any indication as to why it doesn't start in the logs?

[munthajeeb]

I dont get any error message.
What I see is :-

-----------------
Aug  6 01:00:01 syslog01 crond[28465]: (root) CMD (/usr/local/my/scripts/restartsyslog.pl > /dev/null 2>&1)
Aug  6 01:00:01 syslog01 syslog-ng[29980]: Termination requested via signal, terminating;
Aug  6 01:00:01 syslog01 syslog-ng[29980]: syslog-ng shutting down; version='3.0.1'
-----------------

restartsyslog.pl script is called by cronjob and it only calls the above init.d syslog-ng startup script with restart option.
/usr/local/my/etc/init.d/syslog-ng restart

[Admin]

I have a hard time thinking this is it, but it appears that the stop and start commands are more or less initiated simuilateneously.  You could try changing:

Code:

  restart|reload)
        $0 stop
        $0 start

to

Code:

  restart|reload)
        $0 stop && $0 start

[munthajeeb]

Thanks you very much.
Will try it and let you know the result.

[munthajeeb]

This did not help.
but was able to find one thing, it fails to restart only if this script is triggered by
cronJob. If I trigger it manually it works fine.

any clue why this behaviour ?

[munthajeeb]

I was able to catch the error this time, by redirecting the output from the script to
a log file :

Stopping syslog-ng: [  OK  ] Stopping Kernel Logger: [  OK  ]
Starting syslog-ng: /usr/local/my/sbin/syslog-ng: error while loading shared libraries: libevtlog.so.0: cannot open shared object file: No such file or directory
[FAILED] Starting Kernel Logger: [  OK  ]

What can be the isuue ?