From owner-freebsd-questions@freebsd.org Sun Jul 14 23:39:42 2019 Return-Path: Delivered-To: freebsd-questions@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id E3D71AB090 for ; Sun, 14 Jul 2019 23:39:42 +0000 (UTC) (envelope-from lee@adminart.net) Received: from mo6-p01-ob.smtp.rzone.de (mo6-p01-ob.smtp.rzone.de [IPv6:2a01:238:20a:202:5301::12]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) server-signature RSA-PSS (4096 bits) client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "*.smtp.rzone.de", Issuer "TeleSec ServerPass Class 2 CA" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 76E9C6C15B for ; Sun, 14 Jul 2019 23:39:41 +0000 (UTC) (envelope-from lee@adminart.net) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; t=1563147578; s=strato-dkim-0002; d=adminart.net; h=References:Message-ID:Date:In-Reply-To:Subject:Cc:To:From: X-RZG-CLASS-ID:X-RZG-AUTH:From:Subject:Sender; bh=L/hOjrWwwv0GO0mWPaPyxFk/ngtCgiFS88Jdy1Z+dSE=; b=Z2FYrvHZTLsrgaATWb5hA4HTRbNupbm7TaVix6VMrTTHrMoNEdr8x3BlCdLckYh7zb WPFJtkdSiBQTg4Qd3Xt0cnC8NwvDirCVS4t40WpNgv80DRLYCTo6wx9tYoX11IL4z870 bUGZPicQsansaoDG8/sUErk/WztjI7GCThOVXQsI2AU0MjAY/eZqyGgEQ3fN2il5B11h a2bTs9xaEKlMLOIC5plxPsMP1NRjdQy5tr94zKxF0MXDWXjxe8pkA8eX1PEzD6YdyjG/ tmT7KiNDIK2We9MvkeC3n83SR3F2OVa1v6fqVpNyx41NjgbydMgMI0SD9s8x7NBRixdf dfQw== X-RZG-AUTH: ":O2kGeEG7b/pS1FS4THaxjVF9w0vVgfQ9xGcjwO5WMRo5c+h5ceMqQWZ3yrBp+ARdaXvxIDf7nlw=" X-RZG-CLASS-ID: mo00 Received: from himinbjorg.adminart.net by smtp.strato.de (RZmta 44.24 DYNA|AUTH) with ESMTPSA id e0059dv6ENdcRKp (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (curve secp521r1 with 521 ECDH bits, eq. 15360 bits RSA)) (Client did not present a certificate); Mon, 15 Jul 2019 01:39:38 +0200 (CEST) Received: from toy.adminart.net ([192.168.3.55]) by himinbjorg.adminart.net with esmtps (TLSv1.2:ECDHE-RSA-AES256-GCM-SHA384:256) (Exim 4.92) (envelope-from ) id 1hmo5h-00014P-JC; Mon, 15 Jul 2019 01:39:37 +0200 Received: from lee by toy.adminart.net with local (Exim 4.92) (envelope-from ) id 1hmo5h-00010U-Dr; Mon, 15 Jul 2019 01:39:37 +0200 From: hw To: Karl Denninger Cc: freebsd-questions@freebsd.org Subject: Re: dead slow update servers In-Reply-To: (Karl Denninger's message of "Sun, 14 Jul 2019 07:23:10 -0500") Date: Mon, 15 Jul 2019 01:23:43 +0200 Organization: my virtual residence Message-ID: <87v9w4qjy8.fsf@toy.adminart.net> References: <87sgrbi3qg.fsf@toy.adminart.net> <20190712171910.GA25091@neutralgood.org> <871ryuj3ex.fsf@toy.adminart.net> <874l3qfvqw.fsf@toy.adminart.net> <20190714011303.GA25317@neutralgood.org> <87v9w58apd.fsf@toy.adminart.net> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.0.50 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Rspamd-Queue-Id: 76E9C6C15B X-Spamd-Bar: --- Authentication-Results: mx1.freebsd.org; dkim=pass header.d=adminart.net header.s=strato-dkim-0002 header.b=Z2FYrvHZ X-Spamd-Result: default: False [-3.27 / 15.00]; ARC_NA(0.00)[]; RCVD_VIA_SMTP_AUTH(0.00)[]; R_DKIM_ALLOW(-0.20)[adminart.net:s=strato-dkim-0002]; NEURAL_HAM_MEDIUM(-1.00)[-0.999,0]; FROM_HAS_DN(0.00)[]; TO_DN_SOME(0.00)[]; NEURAL_HAM_LONG(-1.00)[-1.000,0]; MIME_GOOD(-0.10)[text/plain]; DMARC_NA(0.00)[adminart.net]; HAS_ORG_HEADER(0.00)[]; RCVD_COUNT_THREE(0.00)[4]; TO_MATCH_ENVRCPT_SOME(0.00)[]; MX_GOOD(-0.01)[cached: smtpin.rzone.de]; DKIM_TRACE(0.00)[adminart.net:+]; RCPT_COUNT_TWO(0.00)[2]; NEURAL_HAM_SHORT(-0.42)[-0.424,0]; R_SPF_NA(0.00)[]; FORGED_SENDER(0.30)[hw@adminart.net,lee@adminart.net]; RCVD_IN_DNSWL_LOW(-0.10)[2.1.0.0.0.0.0.0.0.0.0.0.1.0.3.5.2.0.2.0.a.0.2.0.8.3.2.0.1.0.a.2.list.dnswl.org : 127.0.5.1]; MIME_TRACE(0.00)[0:+]; RCVD_TLS_LAST(0.00)[]; ASN(0.00)[asn:6724, ipnet:2a01:238::/32, country:DE]; FROM_NEQ_ENVFROM(0.00)[hw@adminart.net,lee@adminart.net]; IP_SCORE(-0.74)[ipnet: 2a01:238::/32(-3.27), asn: 6724(-0.41), country: DE(-0.01)] X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Jul 2019 23:39:42 -0000 Karl Denninger writes: > On 7/14/2019 00:10, hw wrote: >> "Kevin P. Neal" writes: >> >>> On Sat, Jul 13, 2019 at 05:39:51AM +0200, hw wrote: >>>> ZFS is great when you have JBODs while storage performance is >>>> irrelevant. I do not have JBODs, and in almost all cases, storage >>>> performance is relevant. >>> Huh? Is a _properly_ _designed_ ZFS setup really slower? A raidz >>> setup of N drives gets you the performance of roughly 1 drive, but a >>> mirror gets you the write performance of a titch less than one drive >>> with the read performance of N drives. How does ZFS hurt performance? >> Performance is hurt when you have N disks and only get the performance >> of a single disk from them. > > There's no free lunch.=C2=A0 If you want two copies of the data (or one p= lus > parity) you must write two copies.=C2=A0 The second one doesn't magically > appear.=C2=A0 If you think it did you were conned by something that is > cheating (e.g. said it had written something when in fact it was sitting > in a DRAM chip) and, at a bad time, you're going to discover it was > cheating. > > Murphy is a SOB. I'm not sure what your point is. Even RAID5 gives you better performance than raidz because it doesn't limit you to a single disk. >> Mirroring the N disks would require another N disks, which you don't >> have. >> >> "Performance" isn't much better defined as "properly designed" here. In >> practise, I prefer a hardware RAID5 with N disks over a raidz with N >> disks and a RAID10 over a RAID5. Unfortunately, in practise, the number >> of disks is limited because they aren't cheap and because only so many >> disks can be connected to a machine without further ado while there is a >> certain requirement for storage capacity. Reality is not proper >> designed :/ >> >> >> What do you do when you put FreeBSD on a server that has a hardware RAID >> controller which doesn't do JBOD? Use ZFS on the RAID? > > Throw said controller in the trash and get a proper one. Show me, for example, such a controller that is certified to be compatible with HP DL380 gen7 servers or Dell R710s, replacing an H700+, and doesn't cost anything. > Raid controllers were very useful a decade ago when ZFS was > trouble-ridden and the controller's firmware was less-so.=C2=A0 Now it's = the > other way around. ZFS is still trouble ridden when you're using Linux even if only because it hasn't been integrated so well due to licensing issues. Software RAID has advantages and disadvantages, same as hardware RAID. In all cases I've been using RAID, hardware RAID has always been the best option considering ease of use, reliability and performance. Of all RAIDs I've been using, ZFS has shown the worst performance which was so bad that I don't want to use it anymore. Maybe ZFS works perfect with FreeBSD and has better performance than what I've seen, but being limited to the performance of a single disk remains unless you can use a mirror. > And whether you do your Raid in hardware or software Raidz is Raidz. > > I binned the last of the hardware RAID adapters in my production > machines roughly five years ago.=C2=A0 ZFS got to be faster and more-reli= able > than they were. I'd have to try ZFS with FreeBSD before I would believe that. In any case, it leaves you with the problem of connecting the disks to the machine. It's not like you could just pull the controller out and connect the disks through thin air. Fiddling with another controller until it supports JBOD (and perhaps works in that server or doesn't) isn't an option like anything else isn't that costs extra money. I don't know much about Dell servers; do they usually support JBOD out of the box?