From owner-freebsd-current@FreeBSD.ORG Wed Oct 3 21:06:59 2007 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 58D8416A420; Wed, 3 Oct 2007 21:06:59 +0000 (UTC) (envelope-from scode@hyperion.scode.org) Received: from hyperion.scode.org (hyperion.scode.org [85.17.42.115]) by mx1.freebsd.org (Postfix) with ESMTP id 0711513C457; Wed, 3 Oct 2007 21:06:58 +0000 (UTC) (envelope-from scode@hyperion.scode.org) Received: by hyperion.scode.org (Postfix, from userid 1001) id 9589323C41F; Wed, 3 Oct 2007 22:50:48 +0200 (CEST) Date: Wed, 3 Oct 2007 22:50:48 +0200 From: Peter Schuller To: Sverre Svenningsen Message-ID: <20071003205048.GA57400@hyperion.scode.org> References: <7344605.82541191344652015.JavaMail.root@zmail.illuminati.org> <20071002181404.GA4717@garage.freebsd.pl> <3939DF4C-D688-40B2-9A9A-1400E528AB07@online.no> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="Q68bSM7Ycu6FN28Q" Content-Disposition: inline In-Reply-To: <3939DF4C-D688-40B2-9A9A-1400E528AB07@online.no> User-Agent: Mutt/1.5.16 (2007-06-09) Cc: freebsd-current , Pawel Jakub Dawidek , Brooks Talley Subject: Re: ZFS corrupting data, even just sitting idle X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 03 Oct 2007 21:06:59 -0000 --Q68bSM7Ycu6FN28Q Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable > Supposedly this card uses a Marvell 88SX6081 chipset, which as far as i= =20 > could tell is handled by Sorens ATA driver. Looks like work done elsewher= e=20 > in the kernel is making that driver misbehave in all sorts of weird ways= =20 > now. How confirmed is this? I am using that exact chipset on CURRENT from about 2007-09-05, with raidz2 on a 6 drive array + a single drive for the system (both on ZFS). I haven't seen checksum errors on scrubbing yet. How recent would this regression be? And what would the symptomes be? Is there a discussion on an ML about this? If it's regressed that would be sad, not only because I happen to have such a controller, but because it's seems to be about the *only* many-port SATA controller you can find, unless you shell out lots of money for a hardware RAID controller you (1) don't even want (2) might even make reliability worse if you want to run ZFS. So for my money it seems to be the best controller available for a home (or otherwise low-budget) server. Is there anything I can to do help resolve any issues, other than fixing them myself (sorry)? Providing a card to somebody for example? --=20 / Peter Schuller PGP userID: 0xE9758B7D or 'Peter Schuller ' Key retrieval: Send an E-Mail to getpgpkey@scode.org E-Mail: peter.schuller@infidyne.com Web: http://www.scode.org --Q68bSM7Ycu6FN28Q Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFHBACnDNor2+l1i30RAsSoAKCcvtv4Uj/p2rak5HfJp5Zz1IbVxgCgycr9 M4k0rUJ+40TsyS6lFiud9WI= =VVrY -----END PGP SIGNATURE----- --Q68bSM7Ycu6FN28Q--