From owner-freebsd-stable@FreeBSD.ORG Wed May 13 19:14:54 2009 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 34544106566C; Wed, 13 May 2009 19:14:54 +0000 (UTC) (envelope-from scrappy@hub.org) Received: from hub.org (hub.org [200.46.204.220]) by mx1.freebsd.org (Postfix) with ESMTP id 023348FC13; Wed, 13 May 2009 19:14:53 +0000 (UTC) (envelope-from scrappy@hub.org) Received: from maia.hub.org (maia-4.hub.org [200.46.204.183]) by hub.org (Postfix) with ESMTP id F125453BC6F; Wed, 13 May 2009 16:14:51 -0300 (ADT) Received: from hub.org ([200.46.204.220]) by maia.hub.org (mx1.hub.org [200.46.204.183]) (amavisd-maia, port 10024) with ESMTP id 28845-06; Wed, 13 May 2009 16:14:51 -0300 (ADT) Received: by hub.org (Postfix, from userid 1002) id B4C1A53BC68; Wed, 13 May 2009 16:14:51 -0300 (ADT) Received: from localhost (localhost [127.0.0.1]) by hub.org (Postfix) with ESMTP id A3C6A53BC63; Wed, 13 May 2009 16:14:51 -0300 (ADT) Date: Wed, 13 May 2009 16:14:51 -0300 (ADT) From: "Marc G. Fournier" To: Steven Hartland In-Reply-To: <6670CCCC546E45E0A28628C45B37A074@multiplay.co.uk> Message-ID: <20090513161014.A17646@hub.org> References: <20090513040719.D17646@hub.org><200905131009.00403.jhb@freebsd.org><20090513133143.M17646@hub.org> <200905131252.15171.jhb@freebsd.org> <20090513142806.V17646@hub.org> <6670CCCC546E45E0A28628C45B37A074@multiplay.co.uk> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: freebsd-stable@freebsd.org, John Baldwin Subject: Re: More data on 7.2-RELEASE "hangs" 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: Wed, 13 May 2009 19:14:54 -0000 On Wed, 13 May 2009, Steven Hartland wrote: > We've seen things similar to this when an process uncommon process does > a query which locks the a table for a large amount of time on mysql. Sooooo many reasons why I hate MySQL :( One thing that we are trying right now is actually along these lines ... we've been working with MySQL 5.1 + NDBD for clustering ... after the last hang, we disabled both the NDBD startup, and mysql, to see if that is the cause, so nice to have some validation on this one ... > In our example this turned out to be an admin query in vbulletin. When > it happened it turned a machine which was purring along quite nicely > into a totally unresponsive machine in a matter of a few seconds as > apache spawned more process that also then instantly stalled... Let me check that the next time around ... compare the specific # of http processes between monitor runs and see if there is a 'sudden jump' ... We'll see hwo the next 'test period' works out, with that MySQL stuff offline ... the other thing I've been working on is moving jails off of that server, one at a time, to see if I can narrow down which one is causing the spike ... I will focus on the mysql backend ones going forward, to eliminate those ... Thx ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email . scrappy@hub.org MSN . scrappy@hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664