From owner-freebsd-threads@FreeBSD.ORG Mon Mar 22 16:50:24 2004 Return-Path: Delivered-To: freebsd-threads@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 933C716A4D0 for ; Mon, 22 Mar 2004 16:50:24 -0800 (PST) Received: from rwcrmhc13.comcast.net (rwcrmhc13.comcast.net [204.127.198.39]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7F3C543D48 for ; Mon, 22 Mar 2004 16:50:24 -0800 (PST) (envelope-from julian@elischer.org) Received: from interjet.elischer.org ([24.7.73.28]) by comcast.net (rwcrmhc13) with ESMTP id <2004032300502201500kthp3e>; Tue, 23 Mar 2004 00:50:23 +0000 Received: from localhost (localhost.elischer.org [127.0.0.1]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id QAA39859; Mon, 22 Mar 2004 16:51:40 -0800 (PST) Date: Mon, 22 Mar 2004 16:51:39 -0800 (PST) From: Julian Elischer To: Kris Gale In-Reply-To: <51632.68.14.252.60.1079987672.squirrel@mail.asn.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: freebsd-threads@freebsd.org Subject: Re: Loaded MySQL 4.0.18 w/ KSE running nicely X-BeenThere: freebsd-threads@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Threading on FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 23 Mar 2004 00:50:24 -0000 On Mon, 22 Mar 2004, Kris Gale wrote: > > WRT compiling/linking/libmap.conf. I have Mysql 4.0 running > > on -current just fine, and so does the mysql maintainer. > > This is something I've said on this list quite a few times, but it > may help shed some light on this situation: > > MySQL does run very well on -current, even handling huge loads > in terms of queries per second. The problem is with surges of > incoming connections. If multiple clients try to connect > simultaneously, things bog down terribly. > > Up until a few weeks ago, a surge of incoming connections > would end up causing my -current system to panic and reboot > about half of the time. > > A recent fix must have corrected this, because I can't reproduce > the kernel panic (which I could do very consistently before). > However, the problem with things bogging down has gotten > much, much worse. you need to help us here.. if you have the ability to simulate this then contact me and we can try various debugging techniques to see what is going on. > > I've posted a backtrace of the crash dump to the list, and also > a perl script which I was using to see the problems. > > If someone is just using a small number of clients (or script that > just makes few connections) to test MySQL, it will appear to run > beautifully. The performance in such a scenario is drastically > better than under 4.x. WHat if you artificially give it one burst of heavy load first? does it handle subsequent loads well? does it act any differently when you use process scope threads instead of system scope threads? (I think Dan sent you the patch to mysql to try this)