From owner-freebsd-stable@FreeBSD.ORG Sat Jan 14 18:37:33 2006 Return-Path: X-Original-To: freebsd-stable@freebsd.org Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0564216A41F; Sat, 14 Jan 2006 18:37:33 +0000 (GMT) (envelope-from ken@abbott.allenmyland.com) Received: from sccrmhc12.comcast.net (sccrmhc12.comcast.net [63.240.77.82]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4798643D45; Sat, 14 Jan 2006 18:37:31 +0000 (GMT) (envelope-from ken@abbott.allenmyland.com) Received: from abbott.allenmyland.com ([68.81.206.140]) by comcast.net (sccrmhc12) with ESMTP id <2006011418372501200h5uile>; Sat, 14 Jan 2006 18:37:25 +0000 Received: by abbott.allenmyland.com (Postfix, from userid 1001) id F2BB817024; Sat, 14 Jan 2006 13:37:24 -0500 (EST) Date: Sat, 14 Jan 2006 13:37:24 -0500 From: Ken Stevenson To: Doug Barton Message-ID: <20060114183724.GA2723@abbott.allenmyland.com> Mail-Followup-To: Doug Barton , freebsd-stable@FreeBSD.org, Brooks Davis References: <20060114011006.GA83466@abbott.allenmyland.com> <43C88F18.5070902@FreeBSD.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <43C88F18.5070902@FreeBSD.org> User-Agent: Mutt/1.4.2.1i Cc: Brooks Davis , freebsd-stable@FreeBSD.org Subject: Re: postgresql doesn't start on boot-up X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 14 Jan 2006 18:37:33 -0000 On Fri, Jan 13, 2006 at 09:41:44PM -0800, Doug Barton wrote: > Ken Stevenson wrote: > >I'm new to FreeBSD and for the first time ever, I cvsupp'ed > >to freebsd-stable 6 and rebuilt everthing, following the instructions > >in the handbook. It went pretty smoothly except that several services > >failed to start and others acted badly. > > > >In particular: > > > >Initially, postfix, cups and postgresql failed to start. > > This sounds like you don't have the most recent versions of these ports > installed, as all of them have had fixes for these problems. It's also > possible that you have stale files in /usr/local/etc/rc.d that are not > related to ports that you have installed. For example, a lot of users have > at some time in the past copied cups.sh.sample to cups.sh, which prevents > updates to the script from being seen. For old style scripts it's safer to > symlink the foo.sh.sample file to foo.sh, fyi. > You were right. I upgraded postfix, cups, apache2 and postgresql and everything started normally. > >named started too late, so ntpd couldn't resolve the timeserver names > >and couldn't set the time. > > This is a more interesting problem. On all of my test systems, named is the > first to run after SERVERS, followed immediately by ntpdate and ntpd. There > is an argument to be made that it would actually be better to run ntp* > first, using IP addresses in its config files, and then run named, since if > you're running a name server for profit (as opposed to fun) accurate > timestamps on the logs are very valuable to diagnose startup problems. > OTOH, it's reasonable to assume for the general case that the current > default order is good, so I have no objection to adding named to the > REQUIRE: line in ntpdate. Ken, can you give this a try and let us know how > it works for you? I'll hold off on making this change till others have a > chance to comment. > > >I solved this per the suggestion in a prior > >post by setting: > > > >early_late_divider="NETWORKING" in rc.conf. After getting the most recent ports, I removed the early_late_divider line from rc.conf and the ntpdate problem went away. dmesg -a looks perfectly normal now. Thanks for your help. Everything's working normally now. -- Ken Stevenson Allen-Myland Inc.