From owner-freebsd-geom@FreeBSD.ORG Sat Nov 3 03:46:03 2007 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 58D0816A417 for ; Sat, 3 Nov 2007 03:46:03 +0000 (UTC) (envelope-from lulf@stud.ntnu.no) Received: from fri.itea.ntnu.no (fri.itea.ntnu.no [129.241.7.60]) by mx1.freebsd.org (Postfix) with ESMTP id 1878B13C4B5 for ; Sat, 3 Nov 2007 03:46:02 +0000 (UTC) (envelope-from lulf@stud.ntnu.no) Received: from localhost (localhost [127.0.0.1]) by fri.itea.ntnu.no (Postfix) with ESMTP id 2AB828C05; Sat, 3 Nov 2007 02:42:50 +0100 (CET) Received: from caracal.stud.ntnu.no (caracal.stud.ntnu.no [129.241.56.185]) by fri.itea.ntnu.no (Postfix) with ESMTP; Sat, 3 Nov 2007 02:42:49 +0100 (CET) Received: by caracal.stud.ntnu.no (Postfix, from userid 2312) id 79D0B62421A; Sat, 3 Nov 2007 02:43:06 +0100 (CET) Date: Sat, 3 Nov 2007 02:43:06 +0100 From: Ulf Lilleengen To: Peter Giessel Message-ID: <20071103014306.GA22755@stud.ntnu.no> References: <8d4842b50710310814w3880f7d3ldf8abe3a236cbcc8@mail.gmail.com> <20071031215756.GB1670@stud.ntnu.no> <472AA59F.3020103@rootnode.com> <0001DFFC-0115-1000-9A80-3F81219C1B16-Webmail-10013@mac.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <0001DFFC-0115-1000-9A80-3F81219C1B16-Webmail-10013@mac.com> User-Agent: Mutt/1.5.9i X-Content-Scanned: with sophos and spamassassin at mailgw.ntnu.no. Cc: freebsd-geom@freebsd.org Subject: Re: gvinum and raid5 X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 03 Nov 2007 03:46:03 -0000 On fre, nov 02, 2007 at 12:38:36 -0700, Peter Giessel wrote: > On Friday, November 02, 2007, at 01:04AM, "Joe Koberg" wrote: > >Ulf Lilleengen wrote: > >> On ons, okt 31, 2007 at 12:14:18 -0300, Marco Haddad wrote: > >> > >>> I found in recent researchs that a lot of people say gvinum should not be > >>> trusted, when it comes to raid5. I began to get worried. Am I alone using > >>> > >>> > >> I'm working on it, and there are definately people still using it. (I've > >> recieved a number of private mails as well as those seen on this list). IMO, > >> gvinum can be trusted when it comes to raid5. I've not experienced any > >> corruption-bugs or anything like that with it. > >> > > > >The source of the mistrust may be the fact that few software-only RAID-5 > >systems can guarantee write consistency across a multi-drive > >read-update-write cycle in the case of, e.g., power failure. > > That may be the true source, but my source of mistrust comes from a few > drive failures and gvinum's inability to rebuild the replaced drive. > > Worked fine under vinum in tests, tried the same thing in gvinum (granted, > this was under FreeBSD 5), and the array failed to rebuild. > > I can't be 100% sure it wasn't a flakey ATA controller and not gvinum's > fault, and I no longer have access to the box to play with, but when I was > playing with gvinum, replacing a failed drive usually resulted in panics. Well, all I can say is that I've tested this many times with gvinum in CURRENT/7.x/6.x as well as my SoC work, and I made updates to the manpage to give examples on how to do this as well. Also, for the software RAID-5 problems... they are hard to "fix" since gvinum doesn't really know anything about the consumers. However, it could be interesting to try out different optimizations like not reading parity when having a sufficiently large request, or some sort of write cache until one can issue a large enough request. -- Ulf Lilleengen