From owner-freebsd-fs@FreeBSD.ORG Mon Apr 15 07:11:53 2013 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by hub.freebsd.org (Postfix) with ESMTP id 6FBD7EC0 for ; Mon, 15 Apr 2013 07:11:53 +0000 (UTC) (envelope-from zbeeble@gmail.com) Received: from mail-vc0-f171.google.com (mail-vc0-f171.google.com [209.85.220.171]) by mx1.freebsd.org (Postfix) with ESMTP id 323C18B0 for ; Mon, 15 Apr 2013 07:11:52 +0000 (UTC) Received: by mail-vc0-f171.google.com with SMTP id ha12so3571713vcb.30 for ; Mon, 15 Apr 2013 00:11:52 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=afVyt8kwLlcc4OxZUOw3rc18xJVOOYIy0xAvUE8/+Ac=; b=ueLxWq5yaENF4yuT6LpaMFd3WTrPNlbiNFjEf79t/draMPz9qqP2B8Exao1MlIR07e QYP6RRCza7L56KamZzez7j5Hbrabiv8AChmV3B5hmU3M8hba6pillwx+kH/qWHYT8QhD fxHy9VL43bQg2W3TV8kFnRqDUf1W4HLWutvtv39HVy/kNpUhZfrt74JHbbnDF9h78I4d xbAyrLrlfwDPXPO70fskkqK+ilLZ37E8yw7Hk32bcV6DtFXrd7ppm6BmCobkRHbHKPLI OVBkTKiGyxWz/yF+hJl935NHdUZ/VumfcitU15wccSAsqLdrTOgL+kHsj9SLkQNDbtKN dZ4A== MIME-Version: 1.0 X-Received: by 10.52.75.8 with SMTP id y8mr12937656vdv.2.1366009912224; Mon, 15 Apr 2013 00:11:52 -0700 (PDT) Received: by 10.220.91.83 with HTTP; Mon, 15 Apr 2013 00:11:52 -0700 (PDT) In-Reply-To: <20130414194440.GB38338@icarus.home.lan> References: <516A8092.2080002@o2.pl> <9C59759CB64B4BE282C1D1345DD0C78E@multiplay.co.uk> <516AF61B.7060204@o2.pl> <20130414185117.GA38259@icarus.home.lan> <20130414194440.GB38338@icarus.home.lan> Date: Mon, 15 Apr 2013 03:11:52 -0400 Message-ID: Subject: Re: A failed drive causes system to hang From: Zaphod Beeblebrox To: Jeremy Chadwick Content-Type: text/plain; charset=ISO-8859-1 X-Content-Filtered-By: Mailman/MimeDel 2.1.14 Cc: freebsd-fs , =?ISO-8859-1?B?UmFkaW8gbcS5P29keWNoIGJhbmR5dMQ/xT93?= , support@lists.pcbsd.org X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 15 Apr 2013 07:11:53 -0000 If I was using my plain old UN*X mailer, I'd try to honor your request for a new thread (by editing the headers)... but I don't see any method by which google allows this. Anyways... rather than discuss my (admittedly vague) "me too" on the drive issue, I'd like to comment on the meta issue you raise. On Sun, Apr 14, 2013 at 3:44 PM, Jeremy Chadwick wrote: > > There is already too much crap going on in this thread with 4 different > people with what are 4 different issues, and nobody at this point is > able to keep track of it all (including the participants). > > This situation happens way, WAY too often with storage-related matters > on the list. ANYTHING ZFS-related and ANYTHING storage-related results > in bandwagon-jumping and threads that spiral out of control/become > almost useless and certainly impossible to follow. It needs to stop. > I think what's happening here is that the whole storage subsystem is (at this point) good enough that people who have problems are encountering fairly obscure but serious corner cases... but that since there isn't much hardware advice from core anymore, it's assumed by the sufferers that these issues must conflate since general experience leaves us thinking there are very few issues. When I say hardware advice... many common list readers might pick up on hardware opinions dropped here but it's easy to miss them and they remain uncollected. Worse, when software workarounds and/or fixed hardware revisions occur, there is again no reflection. Some driver man pages make some statements about hardware capabilities... but other hardware has none. ... and since I'm saying this, I'll volunteer... We need for each class of hardware a simple table of information. As an example, the columns for block storage might be: - chipset (list) - driver (name) - hot swap (y/n) - known to hang on drive failures (y/n) - pmp (y/n, 1:n) - queuing (type) - block sizes (512, 4k, ...) - relative performance (cpu heavy, scatter-gather, etc) - memory support (32 bit, 64 bit, bounce buffers) - "recommended" Similar lists can easily be generated for NICs, motherboards, video (a particular mess) and whatnot. There isn't an incentive for a computer retailer to put together working hardware as lists of components could then easily be bought ... undercutting the margin --- it seems to me that knowledge inside the community needs to be fostered. So... what am I volunteering for? I would be happy to maintain a portion of the FreeBSD wiki with hardware information from components right up to systems in this form, but I would need input from the driver writers ... who are in the best position to know ... what works and what doesn't.