From owner-freebsd-stable@FreeBSD.ORG Tue Jan 8 23:48:17 2008 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 B72E316A419 for ; Tue, 8 Jan 2008 23:48:17 +0000 (UTC) (envelope-from stephen.rumble@utoronto.ca) Received: from bureau8.utcc.utoronto.ca (bureau8.utcc.utoronto.ca [128.100.132.18]) by mx1.freebsd.org (Postfix) with ESMTP id 7340513C447 for ; Tue, 8 Jan 2008 23:48:17 +0000 (UTC) (envelope-from stephen.rumble@utoronto.ca) Received: from webmail7.ns.utoronto.ca ([128.100.132.79] EHLO localhost ident: IDENT-NOT-QUERIED [port 52549]) by bureau8.utcc.utoronto.ca with ESMTP id <24994-19488> convert rfc822-to-8bit; Tue, 8 Jan 2008 18:47:51 -0500 Received: from 76-10-169-143.dsl.teksavvy.com (76-10-169-143.dsl.teksavvy.com [76.10.169.143]) by webmail.utoronto.ca (Horde MIME library) with HTTP; Tue, 08 Jan 2008 18:47:51 -0500 Message-ID: <20080108184751.36w5c4ds6y3o0g4g@webmail.utoronto.ca> Date: Tue, 08 Jan 2008 18:47:51 -0500 From: "Stephen M. Rumble" To: Daniel Eriksson References: <20080108172846.2lglrcvo0qsk88o0@webmail.utoronto.ca> <4F9C9299A10AE74E89EA580D14AA10A61A188A@royal64.emp.zapto.org> In-Reply-To: <4F9C9299A10AE74E89EA580D14AA10A61A188A@royal64.emp.zapto.org> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; format="flowed" Content-Disposition: inline Content-Transfer-Encoding: 8BIT User-Agent: Internet Messaging Program (IMP) H3 (4.1.4) Cc: freebsd-stable@freebsd.org Subject: RE: RELENG_7: zfs mirror causes ata timeout 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: Tue, 08 Jan 2008 23:48:17 -0000 Quoting Daniel Eriksson : > Stephen M. Rumble wrote: > >> The only interesting bit of evidence I could find is that when these >> errors do occur, smartctl reports an increase in the >> Start_Stop_Count >> field on ad6. ad4, which appears to work fine, doesn't demonstrate >> this and has a much lower value. > > This looks a lot like the drive momentarily shutting down due to a power > outage/dip, only to immediately start again. Well, there's usually a sort of click, perhaps as though the drive is parking itself, near when the errors occur. I guess this is it resetting. > Are you sure the power supplies you've tested are good and powerful > enough to power your box? I've tried three supplies. One old, two new. The current one is 300 watts, the largest was 400 watts and the system uses about 40 idle, 60 loaded (it's a mobile cpu/chipset). I doubt supplied power is the issue. > Have you tried replacing the SATA power cables (as well as the actual > data cables)? Are you using the SATA power connectors that shipped with > the PSU or a Y-cable with a molex plug? Molex<->molex connections are > notoriously unreliable (either the plug breaks allowing one of the > connecting cables to halfway slip out, or the connection is simply not > electrically sound due to bad tolerances). I've used both molex<->SATA adapters for the old power supply, as well as SATA connectors for the new one. The issues are always the same, it seems. I'm starting to lose track of everything I've tried. Just to be sure, I'll swap power connections between the drives and see what happens. Thanks for the input, Steve