From owner-freebsd-security Mon Dec 21 07:30:22 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA05506 for freebsd-security-outgoing; Mon, 21 Dec 1998 07:30:22 -0800 (PST) (envelope-from owner-freebsd-security@FreeBSD.ORG) Received: from verdi.nethelp.no (verdi.nethelp.no [158.36.41.162]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id HAA05487 for ; Mon, 21 Dec 1998 07:30:18 -0800 (PST) (envelope-from sthaug@nethelp.no) From: sthaug@nethelp.no Received: (qmail 18929 invoked by uid 1001); 21 Dec 1998 15:30:14 +0000 (GMT) To: eivind@yes.no Cc: des@flood.ping.uio.no, dillon@FreeBSD.ORG, security@FreeBSD.ORG Subject: Re: cvs commit: src/etc rc.conf In-Reply-To: Your message of "Mon, 21 Dec 1998 16:11:10 +0100" References: <19981221161110.E14124@follo.net> X-Mailer: Mew version 1.05+ on Emacs 19.34.2 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Mon, 21 Dec 1998 16:30:14 +0100 Message-ID: <18927.914254214@verdi.nethelp.no> Sender: owner-freebsd-security@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > If named is run in the sandbox, it will have to be restarted every > > time an interface comes up after being down an hour or more - less if > > you lower interface-interval in /etc/namedb/named.conf, which you > > probably will if you run a caching nameserver on a box that has a > > dynamic IP address (e.g. a dialout gateway). It will also complain > > loudly every time it receives any of SIGHUP, SIGINT, SIGILL, SIGSYS or > > SIGTERM unless you perform the appropriate named.conf magic to move > > the pid and dump files to a directory writeable by bind:bind. > > > > OBTW, the /etc/named/s/ hack is just that - a hack, and an ugly one at > > that. > > > > You'll just have to come to terms with the fact that named needs > > privs. > > ... unless you do a series of small modifications. It is not as if > rescanning the interfaces is a _large_ task, or one that couldn't be > done by a forked out half of named, decreasing the chance of a problem > spreading. named, possibly with some small modifications, could easily run in the sandbox for a fairly large class of important configurations, namely the ISP which runs primary and/or secondary service for thousands of domains on one box - and this box is a dedicated name server. (On such a box, interfaces change rarely if at all - so I would be quite comfortable with removing the code for rescanning of interfaces. An initial scan would still be necessary.) Steinar Haug, Nethelp consulting, sthaug@nethelp.no To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-security" in the body of the message