From owner-freebsd-stable@FreeBSD.ORG Mon Sep 22 02:07:33 2003 Return-Path: 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 0032F16A4B3 for ; Mon, 22 Sep 2003 02:07:32 -0700 (PDT) Received: from salmon.maths.tcd.ie (salmon.maths.tcd.ie [134.226.81.11]) by mx1.FreeBSD.org (Postfix) with SMTP id 8B3F843F3F for ; Mon, 22 Sep 2003 02:07:31 -0700 (PDT) (envelope-from dwmalone@maths.tcd.ie) Received: from walton.maths.tcd.ie by salmon.maths.tcd.ie with SMTP id ; 22 Sep 2003 10:07:20 +0100 (BST) Date: Mon, 22 Sep 2003 10:07:19 +0100 From: David Malone To: Aleksey Ovcharenko Message-ID: <20030922090719.GA23435@walton.maths.tcd.ie> References: <3F6A98F3.7080801@freebsd.org> <200309221018.18036.alexovch@ic.kharkov.ua> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200309221018.18036.alexovch@ic.kharkov.ua> User-Agent: Mutt/1.5.3i Sender: dwmalone@maths.tcd.ie cc: stable@FreeBSD.ORG Subject: Re: 4.9 stability update X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 22 Sep 2003 09:07:33 -0000 On Mon, Sep 22, 2003 at 10:18:18AM +0300, Aleksey Ovcharenko wrote: > Ok, how about dirhash problem http://www.freebsd.org/cgi/query-pr.cgi?pr=51016 This problem predates PAE in -stable, just in case there is any confusion. > It's still same even on 4.9-PRERELEASE. On high load with ~4k virtual users on > mail server with ufs_dirhash turned on it cause kernel panic. I remember looking at this for you before, and it looked tricky to resolve. It had many of the traits of a hardware problem or a general memory corruption problem. It wasn't clear to me if dirhash was the problem or just detecting it. Did I suggested running with "sysctl vfs.ufs.dirhash_docheck=1"? It might change the panic to be closer to when the corruption occurs. If you can give me access to the vmcore and debugging kernel you have, I can have another look, but I'm not optimistic. David.