From owner-freebsd-questions@FreeBSD.ORG Tue Jan 31 21:09:51 2006 Return-Path: X-Original-To: questions@freebsd.org Delivered-To: freebsd-questions@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D0DD716A420 for ; Tue, 31 Jan 2006 21:09:51 +0000 (GMT) (envelope-from corwin@aeternal.net) Received: from amber.aeternal.net (amber.in.markiza.sk [62.168.76.21]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0913443D4C for ; Tue, 31 Jan 2006 21:09:50 +0000 (GMT) (envelope-from corwin@aeternal.net) Received: from localhost (localhost.aeternal.net [127.0.0.1]) by amber.aeternal.net (Postfix) with ESMTP id CE4D5B90E for ; Tue, 31 Jan 2006 22:09:46 +0100 (CET) Received: from amber.aeternal.net ([127.0.0.1]) by localhost (amber.aeternal.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 48199-04 for ; Tue, 31 Jan 2006 22:09:46 +0100 (CET) Received: from [192.168.0.44] (pleiades.aeternal.net [192.168.0.44]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by amber.aeternal.net (Postfix) with ESMTP id 34A82B909 for ; Tue, 31 Jan 2006 22:09:46 +0100 (CET) Message-ID: <43DFD21C.2000304@aeternal.net> Date: Tue, 31 Jan 2006 22:09:48 +0100 From: Martin Hudec User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: questions@freebsd.org References: <200601312202.03556.db@traceroute.dk> In-Reply-To: <200601312202.03556.db@traceroute.dk> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at aeternal.net Cc: Subject: Re: Program for restarting stopped/crashed daemons X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: corwin@aeternal.net List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 31 Jan 2006 21:09:52 -0000 Hello, db wrote: > Hi all > > I've been looking for a program to restart apache, postfix and my other server > daemons when they stop/crash, but couldn't find a good one. I've read the > sourcecode for sysutils/monit and sysutils/monitord but they seem to > need /proc and insist on poll instead of using kqueue(2) and kevent(2). So > before I start coding my own, can someone recommend a port or tell me why > kevent(2) shouldn't be used for this? And what is exactly wrong with /proc running? I have had procfs turned off, but our content management application does need to have access to /proc, because in its management of processess (fastcgi app with custom management of cron processes of its own) it have not been able to manage them without it. We used alternative like p5-Proc-ProcessTable, but it did sometimes render our system unusable, userland unresponsive (although kernel was okey). Cheers, Martin