From owner-freebsd-questions  Fri Jan 17 04:23:56 1997
Return-Path: <owner-questions>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.4/8.8.4) id EAA26672
          for questions-outgoing; Fri, 17 Jan 1997 04:23:56 -0800 (PST)
Received: from mail.EUnet.hu (www.eunet.hu [193.225.28.100])
          by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id EAA26654
          for <freebsd-questions@freebsd.org>; Fri, 17 Jan 1997 04:23:49 -0800 (PST)
Received: by mail.EUnet.hu, id NAA01378; Fri, 17 Jan 1997 13:23:42 +0100
Received: (from zgabor@localhost) by CoDe.hu (8.7.5/8.7.3) id NAA00500 for freebsd-questions@freebsd.org; Fri, 17 Jan 1997 13:17:23 +0100 (MET)
From: Zahemszky Gabor <zgabor@CoDe.hu>
Message-Id: <199701171217.NAA00500@CoDe.hu>
Subject: Re: Crontab for killing and restarting named
To: freebsd-questions@freebsd.org (FreeBSD questions)
Date: Fri, 17 Jan 1997 13:17:23 +0100 (MET)
In-Reply-To: <199701170922.BAA09807@connectnet1.connectnet.com> from That Doug Guy at "Jan 17, 97 01:21:26 am"
X-Mailer: ELM [version 2.4ME+ PL11 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> 	Ok, this is driving me nutty.  I need to kill named once a day, and then 
> restart it.  First I tried a command line to do this in the crontab, but I couldn't get 
> that to work, so I decided to use a script.  Here is the script that I ended up with:
> 
> #!/bin/sh
> PID=`/bin/ps ax | /usr/bin/grep named | /usr/bin/grep -v grep | /usr/bin/awk 
> '{print $1}'`
> /bin/kill -9 ${PID}
> /bin/sleep 5
> /usr/sbin/named
> 
> The only problem is, it doesn't work properly.  When I run it from the command 
Yes, because it's name contains ``named'', too, and awk find it.

1)  I found it in AEllen Frisch's Essential System Administration:
instead of:
.... | grep foo | grep -v grep
it would be good:
.... | grep '[f]oo'
It's cheaper, because it uses only 1 grep process, and 1 pipe, not two of
them.  (And if you use grep in the original instead of fgrep, the time and
memory are ~ same.)

2) Your problem has a much cheaper solution:
kill -9 `cat /var/run/named.pid`
in your script, or use the systems
/usr/sbin/named.restart
or
/usr/sbin/ndc restart
scripts, designed for this problem.

Bye, Gabor