Date: Thu, 25 Mar 1999 18:17:48 +0100 From: sthaug@nethelp.no To: kpielorz@tdx.co.uk Cc: jason@intercom.com, freebsd-isp@FreeBSD.ORG Subject: Re: BIND 8.2 Message-ID: <23644.922382268@verdi.nethelp.no> In-Reply-To: Your message of "Wed, 24 Mar 1999 17:24:09 %2B0000" References: <36F91FB9.EA61A2D4@tdx.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
> > What is a good way to monitor a process, and
> > restart if it dies?
>
> Probably to put a script run from the crontab that checks it say, every 5
> minutes then re-run's it if it's not running...?
If it is specifically BIND 8.x we're talking about, why not use the
method that Paul Vixie uses? I figure it's good enough for me if it
works for him.
Steinar Haug, Nethelp consulting, sthaug@nethelp.no
----------------------------------------------------------------------
To: Jared Mauch <jared@voyager.net>
cc: Jared Mauch <jared@puck.nether.net>, bind-workers@isc.org,
Matt Holtz <mholtz@voyager.net>
Subject: Re: 8.2 rel core
Date: Tue, 16 Mar 1999 17:35:55 -0800
From: Paul A Vixie <paul@vix.com>
> This was on our main production server... I hastily loaded
> it too soon.
because bind does not gracefully deal with running out of virtual memory,
we run all of our production servers under this script, called keep-running:
#!/bin/sh
trap "" 1
sleep 60
while :; do
/usr/sbin/named -f >> /var/log/named 2>&1 < /dev/null
/usr/ucb/logger -p emerg "named restart"
sleep 65
done
exit
> now if only people could run their nameservers... /sigh
actually we've been running this locally on all of our production machines
for weeks. (and this bug can't affect nonrecursive servers like the roots).
> Will test with this patch.
thanks.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-isp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?23644.922382268>