From owner-freebsd-scsi Sun Sep 22 03:53:45 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id DAA07811 for freebsd-scsi-outgoing; Sun, 22 Sep 1996 03:53:45 -0700 (PDT) Received: from shadows.aeon.net (bsdscsi@shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id DAA07760 for ; Sun, 22 Sep 1996 03:53:35 -0700 (PDT) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.7.5/8.6.9) id NAA14141 for freebsd-scsi@freebsd.org; Sun, 22 Sep 1996 13:53:43 +0300 (EET DST) From: mika ruohotie Message-Id: <199609221053.NAA14141@shadows.aeon.net> Subject: striping/mirroring? To: freebsd-scsi@freebsd.org Date: Sun, 22 Sep 1996 13:53:43 +0300 (EET DST) X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk hi i've asked this one on usenet before too, but got no replies... i'm about to set up a server that's meant to be "serious" in some sense... the machine will be ga586hx512 boarded, about 128 megs of ram (later more), and i will get adaptec 3940uw... now, what i want to know is which scsi drives would give me the best performance/reliability... i've personally thought using barracuda 4lp, but i feel a bit sceptic it's reliability... anyway, the plot is, if freebsd is able to do it (i know how to strip disks, not how to mirror) that i'd use disk striping on 3-4 drives, and would mirror it to one... can i? how, if i can? mickey -- mika@aeon.net mika ruohotie From owner-freebsd-scsi Sun Sep 22 21:11:18 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id VAA26127 for freebsd-scsi-outgoing; Sun, 22 Sep 1996 21:11:18 -0700 (PDT) Received: from post.io.org (post.io.org [198.133.36.6]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id VAA26097 for ; Sun, 22 Sep 1996 21:11:13 -0700 (PDT) Received: from zap.io.org (taob@zap.io.org [198.133.36.81]) by post.io.org (8.7.5/8.7.3) with SMTP id AAA22861; Mon, 23 Sep 1996 00:11:06 -0400 (EDT) Date: Mon, 23 Sep 1996 00:11:06 -0400 (EDT) From: Brian Tao To: Christoph Kukulies cc: freebsd-scsi@freebsd.org Subject: Re: TLZ04 (DAT) problem In-Reply-To: <199609200925.LAA16560@gilberto.physik.rwth-aachen.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Fri, 20 Sep 1996, Christoph Kukulies wrote: > > Sep 20 11:07:13 escunix /kernel: (aha0:3:0): "DEC TLZ04 1989(C)DEC 1915" type 1 > removable SCSI 2 > > Sep 20 11:12:37 escunix /kernel: st0(aha0:3:0): ABORTED COMMAND asc:4e,0 Overlap > ped commands attempted > Sep 20 11:12:42 escunix /kernel: st0(aha0:3:0): timed out I have a very similar drive in our NetBSD NFS server (although it also worked fine when it was in my FreeBSD workstation, 2.1.0 through to 2.2-CURRENT): ncr0 targ 5 lun 0: SCSI2 1/sequential removable st0 at scsibus0: st0(ncr0:5:0): FAST SCSI-2 150ns (7 Mb/sec) offset 8. It is attached to an NCR SCSI controller though, and I haven't had any problems with it -- Brian Tao (BT300, taob@io.org, taob@ican.net) Senior Systems and Network Administrator, Internet Canada Corp. "Though this be madness, yet there is method in't" From owner-freebsd-scsi Mon Sep 23 12:18:25 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA27494 for freebsd-scsi-outgoing; Mon, 23 Sep 1996 12:18:25 -0700 (PDT) Received: from iafnl.es.iaf.nl (uucp@iafnl.es.iaf.nl [195.108.17.20]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id MAA27347 for ; Mon, 23 Sep 1996 12:18:11 -0700 (PDT) Received: by iafnl.es.iaf.nl with UUCP id AA22261 (5.67b/IDA-1.5 for freebsd-scsi@freebsd.org); Mon, 23 Sep 1996 21:17:01 +0200 Received: (from wilko@localhost) by yedi.iaf.nl (8.7.5/8.6.12) id TAA00366; Mon, 23 Sep 1996 19:04:27 +0200 (MET DST) From: Wilko Bulte Message-Id: <199609231704.TAA00366@yedi.iaf.nl> Subject: Re: TLZ04 (DAT) problem To: taob@io.org (Brian Tao) Date: Mon, 23 Sep 1996 19:04:27 +0200 (MET DST) Cc: kuku@gilberto.physik.rwth-aachen.de, freebsd-scsi@freebsd.org In-Reply-To: from "Brian Tao" at Sep 23, 96 00:11:06 am X-Mailer: ELM [version 2.4 PL24 ME8a] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Brian Tao wrote... > On Fri, 20 Sep 1996, Christoph Kukulies wrote: > > > > Sep 20 11:07:13 escunix /kernel: (aha0:3:0): "DEC TLZ04 1989(C)DEC 1915" type 1 > > removable SCSI 2 > > > > Sep 20 11:12:37 escunix /kernel: st0(aha0:3:0): ABORTED COMMAND asc:4e,0 Overlap > > ped commands attempted > > Sep 20 11:12:42 escunix /kernel: st0(aha0:3:0): timed out > > I have a very similar drive in our NetBSD NFS server (although it > also worked fine when it was in my FreeBSD workstation, 2.1.0 through > to 2.2-CURRENT): > > ncr0 targ 5 lun 0: SCSI2 1/sequential removable > st0 at scsibus0: st0(ncr0:5:0): FAST SCSI-2 150ns (7 Mb/sec) offset 8. > > It is attached to an NCR SCSI controller though, and I haven't had > any problems with it A TLZ07 is a much newer design. If I remember well it is an Archive Python in disguise. I don't know about the TLZ04 (have yet to dismantle one ;-) Wilko _ ____________________________________________________________________ | / o / / _ Bulte email: wilko@yedi.iaf.nl - Arnhem, The Netherlands |/|/ / / /( (_) Do, or do not. There is no 'try' - Yoda -------------------------------------------------------------------------- From owner-freebsd-scsi Mon Sep 23 15:10:21 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id PAA03299 for freebsd-scsi-outgoing; Mon, 23 Sep 1996 15:10:21 -0700 (PDT) Received: from manet.eecs.ukans.edu (manet.eecs.ukans.edu [129.237.80.18]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id PAA03230 for ; Mon, 23 Sep 1996 15:10:12 -0700 (PDT) Received: from localhost (conley@localhost) by manet.eecs.ukans.edu (8.7.5/8.6.12) with SMTP id QAA02961; Mon, 23 Sep 1996 16:41:41 -0500 (CDT) X-Authentication-Warning: manet.eecs.ukans.edu: conley owned process doing -bs Date: Mon, 23 Sep 1996 16:41:41 -0500 (CDT) From: "Dennis R. Conley" X-Sender: conley@manet To: Wilko Bulte cc: Brian Tao , kuku@gilberto.physik.rwth-aachen.de, freebsd-scsi@FreeBSD.org Subject: Re: TLZ04 (DAT) problem In-Reply-To: <199609231704.TAA00366@yedi.iaf.nl> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >A TLZ07 is a much newer design. If I remember well it is an Archive >Python in disguise. I don't know about the TLZ04 (have yet to dismantle >one ;-) I took the cover off a TLZ04 some time ago and found ( to my surprise ) that the drive was made by HP. From owner-freebsd-scsi Tue Sep 24 10:01:05 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id KAA29806 for freebsd-scsi-outgoing; Tue, 24 Sep 1996 10:01:05 -0700 (PDT) Received: from mail.ruhrgebiet.individual.net (in-ruhr.ruhr.de [193.100.176.38]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id KAA29581 for ; Tue, 24 Sep 1996 10:00:30 -0700 (PDT) Received: by mail.ruhrgebiet.individual.net (8.7.1/8.6.12) with UUCP id SAA19898 for freebsd.org!freebsd-scsi; Tue, 24 Sep 1996 18:21:06 +0200 (MET DST) Received: by robkaos.ruhr.de (/\oo/\ Smail3.1.29.1 #29.1) id ; Tue, 24 Sep 96 18:18 MET DST Message-Id: From: robsch@robkaos.ruhr.de (Robert Schien) Subject: How to use HP CD-R under 2.1.5R? To: freebsd-scsi@freebsd.org Date: Tue, 24 Sep 1996 18:18:10 +0200 (MET DST) X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I want to use the HP 4020i CD burner under 2.1.5R. Unfortunately, I can't get the kernel linked successfully. I get unresolved symbols like _wormcontrol, _wormclose etc. Is there anything special I have to note? I simply put device worm0 into the kernel config file. Thank you in advance. Robert From owner-freebsd-scsi Tue Sep 24 11:42:57 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id LAA01873 for freebsd-scsi-outgoing; Tue, 24 Sep 1996 11:42:57 -0700 (PDT) Received: from iafnl.es.iaf.nl (uucp@iafnl.es.iaf.nl [195.108.17.20]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id LAA01838 for ; Tue, 24 Sep 1996 11:42:46 -0700 (PDT) Received: by iafnl.es.iaf.nl with UUCP id AA03772 (5.67b/IDA-1.5 for freebsd-scsi@freebsd.org); Tue, 24 Sep 1996 20:41:56 +0200 Received: (from wilko@localhost) by yedi.iaf.nl (8.7.5/8.6.12) id TAA00798; Tue, 24 Sep 1996 19:49:38 +0200 (MET DST) From: Wilko Bulte Message-Id: <199609241749.TAA00798@yedi.iaf.nl> Subject: Re: TLZ04 (DAT) problem To: conley@eecs.ukans.edu (Dennis R. Conley) Date: Tue, 24 Sep 1996 19:49:38 +0200 (MET DST) Cc: taob@io.org, kuku@gilberto.physik.rwth-aachen.de, freebsd-scsi@freebsd.org In-Reply-To: from "Dennis R. Conley" at Sep 23, 96 04:41:41 pm X-Mailer: ELM [version 2.4 PL24 ME8a] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Dennis R. Conley wrote... > >A TLZ07 is a much newer design. If I remember well it is an Archive > >Python in disguise. I don't know about the TLZ04 (have yet to dismantle > >one ;-) > > I took the cover off a TLZ04 some time ago and found ( to my surprise ) > that the drive was made by HP. We do that all the time. The new TLZ09 eg is a Sony. Wilko _ ____________________________________________________________________ | / o / / _ Bulte email: wilko@yedi.iaf.nl - Arnhem, The Netherlands |/|/ / / /( (_) Do, or do not. There is no 'try' - Yoda -------------------------------------------------------------------------- From owner-freebsd-scsi Tue Sep 24 19:24:41 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id TAA08665 for freebsd-scsi-outgoing; Tue, 24 Sep 1996 19:24:41 -0700 (PDT) Received: from frmug.org (frmug-gw.frmug.org [193.56.58.252]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id TAA08639 for ; Tue, 24 Sep 1996 19:24:32 -0700 (PDT) Received: (from uucp@localhost) by frmug.org (8.6.8/8.6.9) with UUCP id EAA08782 for freebsd-scsi@freebsd.org; Wed, 25 Sep 1996 04:24:30 +0200 Received: (from elrond@localhost) by imladris.frmug.fr.net (8.7.5/8.7.3) id CAA29302 for freebsd-scsi@freebsd.org; Wed, 25 Sep 1996 02:48:42 +0200 (MET DST) From: Bertrand Petit Message-Id: <199609250048.CAA29302@imladris.frmug.fr.net> Subject: Two NCR adaptaters and spin up. To: freebsd-scsi@freebsd.org Date: Wed, 25 Sep 1996 02:48:40 +0200 (MET DST) X-roberto: ce sont X-remibp: Ah! It is marching. X-Mailer: ELM [version 2.4 PL20] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, This is my first message in this list and also this is the first time I dig my hands in the FreeBSD kernel. I got there because I'm building a machine equiped with two NCR adaptators (Asus SCSI-2000 for exactitude) hooked on a TP55 Asus motherboard. All of this is drived by FreeBSD 2.2-960501-SNAP. Boths of the SCSI busses are loaded with hard drives with the auto spinup feature disabled. Upon system powerup, devices on the first bus are spined up by the NCR bios: the second adaptater is leaved aside. I decided, to correct this to add sinup code in the scsi code. I've choosen to add this in sd.c sd_get_parans() function. I may be wrong at that point. The start of the function looks like this (ommiting variables declaration): /* * Is this unit ready? */ if(scsi_test_unit_ready(sc_link,flags|SCSI_SILENT)!=0) { printf("Unit is not ready trying to spinning it up\n"); if(scsi_start_unit(sc_link,flags)!=0) printf("Spinup failed!\n"); return(0); } This code almost work: the unit is detected as not ready (generating errors messages despite the SCSI_SILENT flag) the command is sent (the unit starts) and then the NCR driver barfs but I don't know why. What am I doing wrong? From the boot messages (included bellow) I guess there is something wronf with command queing. Is it necessary to wait for the queued commands to complete before sending the start unit command? Thanks for your help. Sep 24 21:50:28 phoe /kernel.test: ncr1 rev 2 int a irq 10 on pci0:10 Sep 24 21:50:29 phoe /kernel.test: ncr1 waiting for scsi devices to settle Sep 24 21:50:29 phoe /kernel.test: (ncr1:6:0): "CONNER CFP2105S 2.14GB 15" type 0 fixed SCSI 2 Sep 24 21:50:29 phoe /kernel.test: sd1(ncr1:6:0): Direct-Access Sep 24 21:50:29 phoe /kernel.test: Sep 24 21:50:29 phoe /kernel.test: sd1(ncr1:6:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8. Sep 24 21:50:29 phoe /kernel.test: Sep 24 21:50:29 phoe /kernel.test: sd1(ncr1:6:0): NOT READY asc:4,0 Sep 24 21:50:29 phoe /kernel.test: sd1(ncr1:6:0): Logical unit not ready, cause not reportable field replaceable unit: 15 Sep 24 21:50:30 phoe /kernel.test: , retries:2 Sep 24 21:50:30 phoe /kernel.test: Sep 24 21:50:30 phoe /kernel.test: sd1(ncr1:6:0): NOT READY asc:4,0 Sep 24 21:50:30 phoe /kernel.test: sd1(ncr1:6:0): Logical unit not ready, cause not reportable field replaceable unit: 15 Sep 24 21:50:30 phoe /kernel.test: , retries:1 Sep 24 21:50:30 phoe /kernel.test: Sep 24 21:50:31 phoe /kernel.test: sd1(ncr1:6:0): NOT READY asc:4,0 Sep 24 21:50:31 phoe /kernel.test: sd1(ncr1:6:0): Logical unit not ready, cause not reportable field replaceable unit: 15 Sep 24 21:50:31 phoe /kernel.test: , FAILURE Sep 24 21:50:31 phoe /kernel.test: Unit is not ready trying to spinning it up Sep 24 21:50:31 phoe /kernel.test: ncr1: aborting job ... Sep 24 21:50:31 phoe /kernel.test: ncr1:6: ERROR (90:0) (8-0-0) (8/13) @ (a54:50000000). Sep 24 21:50:31 phoe /kernel.test: script cmd = 740a8700 Sep 24 21:50:31 phoe /kernel.test: reg: da 00 00 13 47 08 06 1f 35 08 00 00 80 00 07 02. Sep 24 21:50:31 phoe /kernel.test: ncr1: restart (fatal error). Sep 24 21:50:31 phoe /kernel.test: Sep 24 21:50:31 phoe /kernel.test: sd1(ncr1:6:0): COMMAND FAILED (9 ff) @f0beea00. Sep 24 21:50:31 phoe /kernel.test: Sep 24 21:50:31 phoe /kernel.test: sd1(ncr1:6:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8. Sep 24 21:50:31 phoe /kernel.test: Sep 24 21:50:32 phoe /kernel.test: sd1(ncr1:6:0): ABORTED COMMAND asc:4e,0 Overlapped commands attempted field replaceable unit: 14 Sep 24 21:50:32 phoe /kernel.test: , retries:1 Sep 24 21:50:32 phoe /kernel.test: 0MB (0 512 byte sectors) -- %!ps | Bertrand Petit alias | elrond@imladris.frmug.fr.net | Cette zone est | % | >elrond le demi-Elfe< | elrond@freenix.fr | a louer | 550 0 translate 90 rotate/NewCenturySchlbk-Bold findfont 690 scalefont setfont -5 15 moveto(42)true charpath clip/Helvetica-Bold findfont 8 scalefont setfont 0 9 800{dup 360 mod 360 div 1 1 sethsbcolor/y exch def rand 10 mod neg 11 800 {y moveto(42)show}for}for clippath 4 setlinewidth 0 setgray stroke showpage From owner-freebsd-scsi Wed Sep 25 01:59:53 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id BAA05543 for freebsd-scsi-outgoing; Wed, 25 Sep 1996 01:59:53 -0700 (PDT) Received: from hda.com (ip26-max1-fitch.zipnet.net [199.232.245.26]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id BAA05494 for ; Wed, 25 Sep 1996 01:59:47 -0700 (PDT) Received: (from dufault@localhost) by hda.com (8.6.11/8.6.9) id EAA06551; Wed, 25 Sep 1996 04:59:51 -0400 From: Peter Dufault Message-Id: <199609250859.EAA06551@hda.com> Subject: Re: How to use HP CD-R under 2.1.5R? To: robsch@robkaos.ruhr.de (Robert Schien) Date: Wed, 25 Sep 1996 04:59:50 -0400 (EDT) Cc: freebsd-scsi@freebsd.org In-Reply-To: from "Robert Schien" at Sep 24, 96 06:18:10 pm Reply-to: hdalog@zipnet.net X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I want to use the HP 4020i CD burner under 2.1.5R. > Unfortunately, I can't get the kernel linked successfully. > I get unresolved symbols like _wormcontrol, _wormclose etc. > Is there anything special I have to note? > I simply put > device worm0 > into the kernel config file. You have to let config know it is a scsi device. This lets one add new scsi drivers without changing config. "sd", etc have special keywords added to config. Try: > device worm0 at scbus? -- Peter Dufault Real-Time Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 dufault@hda.com Fax: 508 433 5267 From owner-freebsd-scsi Wed Sep 25 02:58:05 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id CAA11733 for freebsd-scsi-outgoing; Wed, 25 Sep 1996 02:58:05 -0700 (PDT) Received: from mail.ruhrgebiet.individual.net (in-ruhr.ruhr.de [193.100.176.38]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id CAA11682 for ; Wed, 25 Sep 1996 02:58:00 -0700 (PDT) Received: by mail.ruhrgebiet.individual.net (8.7.1/8.6.12) with UUCP id LAA24704 for freebsd.org!freebsd-scsi; Wed, 25 Sep 1996 11:41:16 +0200 (MET DST) Received: by robkaos.ruhr.de (/\oo/\ Smail3.1.29.1 #29.1) id ; Wed, 25 Sep 96 11:39 MET DST Message-Id: From: robsch@robkaos.ruhr.de (Robert Schien) Subject: Re: How to use HP CD-R under 2.1.5R? In-Reply-To: <199609250859.EAA06551@hda.com> from Peter Dufault at "Sep 25, 96 04:59:50 am" To: freebsd-scsi@freebsd.org Date: Wed, 25 Sep 1996 11:39:52 +0200 (MET DST) X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > I want to use the HP 4020i CD burner under 2.1.5R. > > Unfortunately, I can't get the kernel linked successfully. > > I get unresolved symbols like _wormcontrol, _wormclose etc. > > Is there anything special I have to note? > > I simply put > > device worm0 > > into the kernel config file. > > You have to let config know it is a scsi device. This lets one > add new scsi drivers without changing config. "sd", etc have > special keywords added to config. Try: > > > device worm0 at scbus? > I tried this. Didn't help. Robert From owner-freebsd-scsi Thu Sep 26 19:38:01 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id TAA23537 for freebsd-scsi-outgoing; Thu, 26 Sep 1996 19:38:01 -0700 (PDT) Received: from silvia.HIP.Berkeley.EDU (ala-ca16-06.ix.netcom.com [204.32.168.134]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id TAA23372 for ; Thu, 26 Sep 1996 19:37:37 -0700 (PDT) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.7.6/8.6.9) id TAA18470; Thu, 26 Sep 1996 19:36:32 -0700 (PDT) Date: Thu, 26 Sep 1996 19:36:32 -0700 (PDT) Message-Id: <199609270236.TAA18470@silvia.HIP.Berkeley.EDU> To: bsdscsi@shadows.aeon.net CC: freebsd-scsi@freebsd.org In-reply-to: <199609221053.NAA14141@shadows.aeon.net> (message from mika ruohotie on Sun, 22 Sep 1996 13:53:43 +0300 (EET DST)) Subject: Re: striping/mirroring? From: asami@freebsd.org (Satoshi Asami) Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * i've asked this one on usenet before too, but got no replies... * * i'm about to set up a server that's meant to be "serious" in some * sense... * * the machine will be ga586hx512 boarded, about 128 megs of ram (later more), * and i will get adaptec 3940uw... * * now, what i want to know is which scsi drives would give me the best * performance/reliability... i've personally thought using barracuda 4lp, * but i feel a bit sceptic it's reliability... What performance? Sequential or random? I'm assuming that ga586hx512 is an Intel 430 HX with 512KB of cache, that's as good as it gets as far as motherboards go. If you want to maximize sequential access, you'll need about 6 or 7 disks to stripe across them, you'll get to about 28 MB/s with the option "I586_FAST_BCOPY". For random access, you'll need more like 30 drives through the filesystem to max out the motherboard. * anyway, the plot is, if freebsd is able to do it (i know how to strip * disks, not how to mirror) that i'd use disk striping on 3-4 drives, and * would mirror it to one... can i? how, if i can? That isn't mirroring. ;) If you have 2N drives, mirroring will make that look like a N-disk ccd. Satoshi From owner-freebsd-scsi Fri Sep 27 02:39:22 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id CAA11101 for freebsd-scsi-outgoing; Fri, 27 Sep 1996 02:39:22 -0700 (PDT) Received: from shadows.aeon.net (shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id CAA11020; Fri, 27 Sep 1996 02:39:14 -0700 (PDT) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.7.5/8.6.9) id MAA20281; Fri, 27 Sep 1996 12:38:33 +0300 (EET DST) From: mika ruohotie Message-Id: <199609270938.MAA20281@shadows.aeon.net> Subject: Re: striping/mirroring? To: asami@FreeBSD.ORG (Satoshi Asami) Date: Fri, 27 Sep 1996 12:38:32 +0300 (EET DST) Cc: freebsd-scsi@FreeBSD.ORG In-Reply-To: <199609270236.TAA18470@silvia.HIP.Berkeley.EDU> from Satoshi Asami at "Sep 26, 96 07:36:32 pm" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > * now, what i want to know is which scsi drives would give me the best > * performance/reliability... i've personally thought using barracuda 4lp, > * but i feel a bit sceptic it's reliability... > What performance? Sequential or random? I'm assuming that ga586hx512 performance as getting as high as possible read/write speed to/from hard drives. > is an Intel 430 HX with 512KB of cache, that's as good as it gets as > far as motherboards go. yes, it's gigabyte's board that's tested being the 2nd fastest board on market, only tyan is faster... (http://www.u-net.com/sysdoc) and has the said chipset, and cache. > If you want to maximize sequential access, you'll need about 6 or 7 > disks to stripe across them, you'll get to about 28 MB/s with the > option "I586_FAST_BCOPY". For random access, you'll need more like 30 > drives through the filesystem to max out the motherboard. ok only 28 mb/s? isnt it possible to get more speed? > * would mirror it to one... can i? how, if i can? > That isn't mirroring. ;) but what? > If you have 2N drives, mirroring will make that look like a N-disk > ccd. ??? i meant, that: while striping using 4 or something drives, i would love to be able to mirror whatever i am striping to one disk. just to be safe... and now that i think of it... is it even possible? anyone? i would not mind if that safe copy would not run as fast as the rest of the filesystem, but that i would have atleast some recent version of the system somewhere on _one_ disk... and while thinking more... is it sane? =) > Satoshi mickey -- mika@aeon.net mika ruohotie From owner-freebsd-scsi Fri Sep 27 04:10:43 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id EAA08149 for freebsd-scsi-outgoing; Fri, 27 Sep 1996 04:10:43 -0700 (PDT) Received: from smtp.utexas.edu (smtp.utexas.edu [128.83.126.2]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id EAA08123 for ; Fri, 27 Sep 1996 04:10:39 -0700 (PDT) Received: (qmail-queue invoked by uid 0); 27 Sep 1996 11:10:26 -0000 Received: from unknown (HELO mail.utexas.edu) (128.83.126.1) by smtp.utexas.edu with SMTP; 27 Sep 1996 11:10:26 -0000 Received: from [208.2.87.4] (cod.dataplex.net [208.2.87.4]) by mail.utexas.edu (8.6.9/8.6.6) with SMTP id GAA23706; Fri, 27 Sep 1996 06:10:25 -0500 X-Sender: rkw@eel.dataplex.net Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 27 Sep 1996 06:10:25 -0500 To: mika ruohotie From: rkw@dataplex.net (Richard Wackerbarth) Subject: Re: striping/mirroring? Cc: freebsd-scsi@FreeBSD.org Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >i meant, that: while striping using 4 or something drives, i would love >to be able to mirror whatever i am striping to one disk. just to be >safe... Look at what striping does... It takes a number of drives and makes it appear that they are one large (faster) drive. Mirroring takes two drives of the same size and stores the information redundantly. Thus if you were to mirror a 4 drive combination to a single drive, that drive would have to be 4 times as large. It would also need to be fast. From owner-freebsd-scsi Fri Sep 27 15:14:53 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id PAA29060 for freebsd-scsi-outgoing; Fri, 27 Sep 1996 15:14:53 -0700 (PDT) Received: from irz301.inf.tu-dresden.de ([141.76.1.11]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id PAA28998 for ; Fri, 27 Sep 1996 15:14:46 -0700 (PDT) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by irz301.inf.tu-dresden.de (8.6.12/8.6.12-s1) with ESMTP id AAA24714; Sat, 28 Sep 1996 00:13:46 +0200 Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id AAA16508; Sat, 28 Sep 1996 00:13:45 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.7.5/8.6.9) id XAA12175; Fri, 27 Sep 1996 23:50:45 +0200 (MET DST) From: J Wunsch Message-Id: <199609272150.XAA12175@uriah.heep.sax.de> Subject: Re: Two NCR adaptaters and spin up. To: freebsd-scsi@freebsd.org Date: Fri, 27 Sep 1996 23:50:45 +0200 (MET DST) Cc: elrond@imladris.frmug.fr.net (Bertrand Petit) Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199609250048.CAA29302@imladris.frmug.fr.net> from Bertrand Petit at "Sep 25, 96 02:48:40 am" X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E X-Mailer: ELM [version 2.4ME+ PL17 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Bertrand Petit wrote: > Boths of the SCSI busses are loaded with hard drives with the > auto spinup feature disabled. Upon system powerup, devices on the > first bus are spined up by the NCR bios: the second adaptater is > leaved aside. > > I decided, to correct this to add sinup code in the scsi code. > I've choosen to add this in sd.c sd_get_parans() function. I may be > wrong at that point. There was already code to do this, but slightly at the wrong location (a few lines too low). Have a look at the `od' driver from -current (make a side-by-side comparision), the spinup code does work there. I'm using it all the time (the `od' driver has an option to spindown the drive after unmounting the filesystem). -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Fri Sep 27 15:15:45 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id PAA29394 for freebsd-scsi-outgoing; Fri, 27 Sep 1996 15:15:45 -0700 (PDT) Received: from irz301.inf.tu-dresden.de ([141.76.1.11]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id PAA29364 for ; Fri, 27 Sep 1996 15:15:41 -0700 (PDT) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by irz301.inf.tu-dresden.de (8.6.12/8.6.12-s1) with ESMTP id AAA24786; Sat, 28 Sep 1996 00:15:14 +0200 Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id AAA16661; Sat, 28 Sep 1996 00:15:14 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.7.5/8.6.9) id UAA11188; Fri, 27 Sep 1996 20:52:17 +0200 (MET DST) From: J Wunsch Message-Id: <199609271852.UAA11188@uriah.heep.sax.de> Subject: Re: Jaztool for FreeBSD, anyone ? To: scsi@freebsd.org Date: Fri, 27 Sep 1996 20:52:17 +0200 (MET DST) Cc: jdli@freebsd.csie.nctu.edu.tw (Jian-Da Li) Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199609212039.EAA27543@FreeBSD.csie.NCTU.edu.tw> from Jian-Da Li at "Sep 22, 96 04:39:36 am" X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E X-Mailer: ELM [version 2.4ME+ PL17 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Jian-Da Li wrote: > I tried to port jaztool (http://www.cnct.com/~bwillmot/jaztool/) for > linux to freebsd without success since I don't know ioctl() and > scsi_command at all. I just hang the scsi bus nicely. :) > > This tool use ioctl() with scsi_command, I felt that it should be > very easy to port. libscsi(3) is probably your friend. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Fri Sep 27 16:46:53 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id QAA17033 for freebsd-scsi-outgoing; Fri, 27 Sep 1996 16:46:53 -0700 (PDT) Received: from irz301.inf.tu-dresden.de ([141.76.1.11]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id QAA16986 for ; Fri, 27 Sep 1996 16:46:48 -0700 (PDT) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by irz301.inf.tu-dresden.de (8.6.12/8.6.12-s1) with ESMTP id BAA27360; Sat, 28 Sep 1996 01:45:31 +0200 Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id BAA18987; Sat, 28 Sep 1996 01:45:22 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.7.5/8.6.9) id BAA13439; Sat, 28 Sep 1996 01:05:20 +0200 (MET DST) From: J Wunsch Message-Id: <199609272305.BAA13439@uriah.heep.sax.de> Subject: Re: HP T4000s tape drive To: freebsd-scsi@freebsd.org Date: Sat, 28 Sep 1996 01:05:20 +0200 (MET DST) Cc: richard@cogsci.ed.ac.uk Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <9295.199609211530@pitcairn.cogsci.ed.ac.uk> from "richard@cogsci.ed.ac.uk" at "Sep 21, 96 04:30:52 pm" X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E X-Mailer: ELM [version 2.4ME+ PL17 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As richard@cogsci.ed.ac.uk wrote: > I've just got an HP (Colorado) T4000s tape drive working under FreeBSD > 2.1.0. It's a SCSI QIC-3095 drive, 4GB uncompressed. Good news! > The main change required is that the PF bit in mode select must be set to > one. Is a "quirk" the right way to do this? I haven't yet looked at Yep, it is. > the -current code to see how the SCSI code has changed since 2.1.0. It has, massively. > The other changes are to increase SCSI_2_MAX_DENSITY_CODE from 0x17 MAX_DENSITY_CODE should be killed. I think we were already consent on this, but nobody did the deed by now. Well, mmmaybe it's already in Justin's ongoing SCSI work. > (not essential, since you can't change the density of this drive) and > add a SCSI_SILENT flag to the scsi_prevent call in st_mount_tape(). PREVENT ALLOW MEDIUM REMOVAL is optional by the SCSI-2 specs, so we should perhaps really make the call to it `silent'. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Fri Sep 27 20:19:48 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id UAA17895 for freebsd-scsi-outgoing; Fri, 27 Sep 1996 20:19:48 -0700 (PDT) Received: from freefall.freebsd.org (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id UAA17845; Fri, 27 Sep 1996 20:19:41 -0700 (PDT) Message-Id: <199609280319.UAA17845@freefall.freebsd.org> To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) cc: freebsd-scsi@freebsd.org, richard@cogsci.ed.ac.uk Subject: Re: HP T4000s tape drive In-reply-to: Your message of "Sat, 28 Sep 1996 01:05:20 +0200." <199609272305.BAA13439@uriah.heep.sax.de> Date: Fri, 27 Sep 1996 20:19:40 -0700 From: "Justin T. Gibbs" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >> The other changes are to increase SCSI_2_MAX_DENSITY_CODE from 0x17 > >MAX_DENSITY_CODE should be killed. I think we were already consent on >this, but nobody did the deed by now. Well, mmmaybe it's already in >Justin's ongoing SCSI work. st.c: revision 1.70 date: 1996/06/24 04:54:32; author: gibbs; state: Exp; lines: +22 -10 Ensure that media protection is released before attempting to eject the media in all cases. Remove SCSI_2_MAX_DENSITY_CODE definition and rely on the device to tell us if we attempt an invalid setting. Closes PR 1245. >> (not essential, since you can't change the density of this drive) and >> add a SCSI_SILENT flag to the scsi_prevent call in st_mount_tape(). > >PREVENT ALLOW MEDIUM REMOVAL is optional by the SCSI-2 specs, so we >should perhaps really make the call to it `silent'. I'll add it to the list of things to look at when I'm working on the st driver. >-- >cheers, J"org > >joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE >Never trust an operating system you don't have sources for. ;-) -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-scsi Sat Sep 28 01:11:42 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id BAA28348 for freebsd-scsi-outgoing; Sat, 28 Sep 1996 01:11:42 -0700 (PDT) Received: from shadows.aeon.net (bsdscsi@shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id BAA25315 for ; Sat, 28 Sep 1996 01:06:06 -0700 (PDT) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.7.5/8.6.9) id KAA29863; Sat, 28 Sep 1996 10:59:29 +0300 (EET DST) From: mika ruohotie Message-Id: <199609280759.KAA29863@shadows.aeon.net> Subject: Re: striping/mirroring? To: rkw@dataplex.net (Richard Wackerbarth) Date: Sat, 28 Sep 1996 10:59:29 +0300 (EET DST) Cc: freebsd-scsi@FreeBSD.org In-Reply-To: from Richard Wackerbarth at "Sep 27, 96 06:10:25 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > Look at what striping does... > It takes a number of drives and makes it appear that they are one large > (faster) drive. i know > Mirroring takes two drives of the same size and stores the information > redundantly. i know > Thus if you were to mirror a 4 drive combination to a single drive, that > drive would have to be 4 times as large. It would also need to be fast. it should be huge, yes. fast, no, if i dont want to mirror all i am striping... or if i do the way i was planning... using 2 gig drives, and only _use_ the space of one disk i'm mirroring to... original question was, can i do it with freebsd? seems to me answer is no. mickey From owner-freebsd-scsi Sat Sep 28 04:15:41 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id EAA08544 for freebsd-scsi-outgoing; Sat, 28 Sep 1996 04:15:41 -0700 (PDT) Received: from smtp.utexas.edu (smtp.utexas.edu [128.83.126.2]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id EAA08515 for ; Sat, 28 Sep 1996 04:15:37 -0700 (PDT) Received: (qmail-queue invoked by uid 0); 28 Sep 1996 11:15:35 -0000 Received: from unknown (HELO mail.utexas.edu) (128.83.126.1) by smtp.utexas.edu with SMTP; 28 Sep 1996 11:15:35 -0000 Received: from [208.2.87.4] (cod.dataplex.net [208.2.87.4]) by mail.utexas.edu (8.6.9/8.6.6) with SMTP id GAA08833; Sat, 28 Sep 1996 06:15:34 -0500 X-Sender: rkw@eel.dataplex.net Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Sat, 28 Sep 1996 06:15:34 -0500 To: mika ruohotie From: rkw@dataplex.net (Richard Wackerbarth) Subject: Re: striping/mirroring? Cc: scsi@freebsd.org Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >> Look at what striping does... >> It takes a number of drives and makes it appear that they are one large >> (faster) drive. > >i know > >> Mirroring takes two drives of the same size and stores the information >> redundantly. > >i know > >> Thus if you were to mirror a 4 drive combination to a single drive, that >> drive would have to be 4 times as large. It would also need to be fast. > >it should be huge, yes. fast, no, if i dont want to mirror all i am >striping... or if i do the way i was planning... using 2 gig drives, >and only _use_ the space of one disk i'm mirroring to... I'm lost. How can the mirror be slower than the drive that it is mirroring unless it is a "read mostly" situation and you are guaranteed no write bursts? >original question was, can i do it with freebsd? > >seems to me answer is no. I think that the answer is "yes". You can you elaborate? Describe the configuration that you have in mind. From owner-freebsd-scsi Sat Sep 28 12:26:46 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA13319 for freebsd-scsi-outgoing; Sat, 28 Sep 1996 12:26:46 -0700 (PDT) Received: from shadows.aeon.net (bsdscsi@shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id MAA12982; Sat, 28 Sep 1996 12:26:20 -0700 (PDT) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.7.5/8.6.9) id WAA05358; Sat, 28 Sep 1996 22:25:45 +0300 (EET DST) From: mika ruohotie Message-Id: <199609281925.WAA05358@shadows.aeon.net> Subject: Re: striping/mirroring? To: freebsd-isp@freebsd.org Date: Sat, 28 Sep 1996 22:25:45 +0300 (EET DST) Cc: scsi@freebsd.org In-Reply-To: from Richard Wackerbarth at "Sep 28, 96 06:15:34 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I'm lost. How can the mirror be slower than the drive that it is mirroring > unless it is a "read mostly" situation and you are guaranteed no write > bursts? hmm? > >seems to me answer is no. > I think that the answer is "yes". You can you elaborate? Describe the > configuration that you have in mind. ack... i admit first i had minor thinking error when i was planning the system at the first time... now, what i want to do is this: strip with few drives, so that i can get speed, and then mirror it all to one "safe" disk, so that i wont lost all of my data if the striping crashes... but now that i think of it, i will lose all the speed i would gain by using ccd if i'd have to "wait" until it's all being written to a one drive... so i would love to hear suggestions how to get more speed than just mounting my system thru several drives, but still having some "safety" left... actually, so that i wont waste any more of anyone's time, i think i can settle down if i get an opinion/numbers how much slower the disks will perform if i use four 2gig drives (two channel adapter used) compared if i would use them ccded? the server will be www/ftp machine which should be as fast as possible, still being safe what comes to loss of data in crashes... i was planning to run 2.1.5 on it. or should i use -current? (sent to freebsd-isp too since i think this is more isp related) mickey -- mika@aeon.net mika ruohotie net/sys admin From owner-freebsd-scsi Sat Sep 28 12:59:22 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA09418 for freebsd-scsi-outgoing; Sat, 28 Sep 1996 12:59:22 -0700 (PDT) Received: from al.imforei.apana.org.au (pjchilds@al.imforei.apana.org.au [202.12.89.41]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id MAA09344 for ; Sat, 28 Sep 1996 12:59:15 -0700 (PDT) Received: (from pjchilds@localhost) by al.imforei.apana.org.au (8.8.0/8.7.3) id FAA23341; Sun, 29 Sep 1996 05:29:13 +0930 (CST) Date: Sun, 29 Sep 1996 05:29:13 +0930 (CST) From: Peter Childs Message-Id: <199609281959.FAA23341@al.imforei.apana.org.au> To: bsdscsi@shadows.aeon.net (mika ruohotie), freebsd-scsi@freebsd.org Subject: Re: striping/mirroring? X-Newsreader: TIN [version 1.2 PL2] Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk : > Thus if you were to mirror a 4 drive combination to a single drive, that : > drive would have to be 4 times as large. It would also need to be fast. : it should be huge, yes. fast, no, if i dont want to mirror all i am : striping... or if i do the way i was planning... using 2 gig drives, : and only _use_ the space of one disk i'm mirroring to... I'm really quite interested in how you propose to have a mirrored system where drives can run at different steeds. Mirror works like this DATA ----> (going to disk) -------+--(and another copy)----+ | | disk 1 disk 2 So if i send some data to the (mirrored) disk, and disk1 is significantly faster than disk2, then disk1 will finish the write first.... so then do i just let it write more? Can you see a problem with this? What happens when disk1 is miles ahead.. who is supposed to keep track of all this data thats queued up for disk2? The OS? And if the disk 1 suddenly fails then you're screwed, and you fault tolerant system wan't very. : original question was, can i do it with freebsd? : seems to me answer is no. The reason FreeBSD can't do it is because the only people capable of writting it know that its not really the best idea. Peter -- Peter Childs --- http://www.imforei.apana.org.au/~pjchilds Finger pjchilds@al.imforei.apana.org.au for public PGP key Drag me, drop me, treat me like an object! From owner-freebsd-scsi Sat Sep 28 14:15:34 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id OAA18747 for freebsd-scsi-outgoing; Sat, 28 Sep 1996 14:15:34 -0700 (PDT) Received: from brasil.moneng.mei.com (brasil.moneng.mei.com [151.186.109.160]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id OAA18678; Sat, 28 Sep 1996 14:15:26 -0700 (PDT) Received: (from jgreco@localhost) by brasil.moneng.mei.com (8.7.Beta.1/8.7.Beta.1) id QAA00554; Sat, 28 Sep 1996 16:13:37 -0500 From: Joe Greco Message-Id: <199609282113.QAA00554@brasil.moneng.mei.com> Subject: Re: striping/mirroring? To: bsdscsi@shadows.aeon.net (mika ruohotie) Date: Sat, 28 Sep 1996 16:13:37 -0500 (CDT) Cc: freebsd-isp@freebsd.org, scsi@freebsd.org In-Reply-To: <199609281925.WAA05358@shadows.aeon.net> from "mika ruohotie" at Sep 28, 96 10:25:45 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > strip with few drives, so that i can get speed, and then mirror it all > to one "safe" disk, so that i wont lost all of my data if the striping > crashes... Having used ccd extensively for the past year, on production news server systems, I think I can safely state that you need not concern yourself too much with "striping crashes"... you are more likely to run into a dead disk than a problem with ccd. I have over two dozen heavily utilized ccd partitions in operation and have not seen any problems with ccd. If you are looking purely for reliability, look for a RAID solution. RAID's are more reliable than a SLED (Single Large Expensive Disk). If you are looking for speed, either RAID or striping may be appropriate. Since I am generally interested primarily in speed, I use ccd and its striping capabilities. ... JG From owner-freebsd-scsi Sat Sep 28 14:33:38 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id OAA27774 for freebsd-scsi-outgoing; Sat, 28 Sep 1996 14:33:38 -0700 (PDT) Received: from smtp.utexas.edu (smtp.utexas.edu [128.83.126.2]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id OAA27728 for ; Sat, 28 Sep 1996 14:33:32 -0700 (PDT) Received: (qmail-queue invoked by uid 0); 28 Sep 1996 21:33:30 -0000 Received: from unknown (HELO mail.utexas.edu) (128.83.126.1) by smtp.utexas.edu with SMTP; 28 Sep 1996 21:33:30 -0000 Received: from [208.2.87.4] (cod.dataplex.net [208.2.87.4]) by mail.utexas.edu (8.6.9/8.6.6) with SMTP id QAA32500; Sat, 28 Sep 1996 16:33:29 -0500 X-Sender: rkw@eel.dataplex.net Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Sat, 28 Sep 1996 16:33:27 -0500 To: Peter Childs From: rkw@dataplex.net (Richard Wackerbarth) Subject: Re: striping/mirroring? Cc: freebsd-scsi@freebsd.org Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I'm really quite interested in how you propose to have a mirrored > system where drives can run at different steeds. Mirror works like > this > > DATA ----> (going to disk) -------+--(and another copy)----+ > | | > disk 1 disk 2 > > So if i send some data to the (mirrored) disk, and disk1 > is significantly faster than disk2, then disk1 will finish the write > first.... so then do i just let it write more? No. However you can READ from disk1 while you wait for disk2 to finish. This may work particularly well in a case that most of the operations are reads. For example, a web server could be expected to have mostly reads with only a few writes. From owner-freebsd-scsi Sat Sep 28 15:42:23 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id PAA07072 for freebsd-scsi-outgoing; Sat, 28 Sep 1996 15:42:23 -0700 (PDT) Received: from gargoyle.carpe.net (root@gargoyle.carpe.net [194.162.243.7]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id PAA07002 for ; Sat, 28 Sep 1996 15:42:12 -0700 (PDT) Received: from helva.grefen.carpe.net (helva.grefen.carpe.net [194.162.243.129]) by gargoyle.carpe.net (8.7.4/8.7.3) with ESMTP id AAA01527; Sun, 29 Sep 1996 00:46:09 +0200 (MET DST) Received: from hex.grefen.carpe.net (root@hex [194.162.243.130]) by helva.grefen.carpe.net (8.7.5/8.7.3) with ESMTP id AAA25719; Sun, 29 Sep 1996 00:40:59 +0200 (MET DST) Received: from hex.grefen.carpe.net (grefen@localhost [127.0.0.1]) by hex.grefen.carpe.net (8.7.3/8.7.3) with ESMTP id AAA20251; Sun, 29 Sep 1996 00:40:57 +0200 (MET DST) To: Peter Childs Cc: bsdscsi@shadows.aeon.net (mika ruohotie), freebsd-scsi@freebsd.org Subject: Mirroring woth different speeds Was: Re: striping/mirroring? Reply-To: grefen@carpe.net Date: Sun, 29 Sep 1996 00:40:56 +0200 Message-ID: <20248.843950456@hex.grefen.carpe.net> From: Stefan Grefen Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <199609281959.FAA23341@al.imforei.apana.org.au> Peter Childs wrote: [...] This is a seperate issue, and becomes unrelated to the current thread: > > I'm really quite interested in how you propose to have a mirrored > system where drives can run at different steeds. Mirror works like > this > > DATA ----> (going to disk) -------+--(and another copy)----+ > | | > disk 1 disk 2 > > So if i send some data to the (mirrored) disk, and disk1 > is significantly faster than disk2, then disk1 will finish the write > first.... so then do i just let it write more? Can you see a > problem with this? What happens when disk1 is miles ahead.. who > is supposed to keep track of all this data thats queued up for > disk2? The OS? And if the disk 1 suddenly fails then you're > screwed, and you fault tolerant system wan't very. This is possible, but not with a standard filesystem. If you use a kind of transaction-based filesystem you can do it safely and in case of a crash you roleback the transactions that didn't make it to the mirror (you commit it on disk only after the mirror has been updated, but pretend otherwise it happend). Of cause the you have to make sure the commits on the mirror are done in the same order on bith disks. The difference between the 2 drives would limited only be available disk-size (for the old copies on drive 1). > > : original question was, can i do it with freebsd? > > : seems to me answer is no. > > The reason FreeBSD can't do it is because the only people > capable of writting it know that its not really the best idea. To make it short, it can (and has) been done, but not in *BSD (yet). This would be a nice and interesting spare-time project, if there was enough spare-time (or somebody would pay for it) ... For the given application it's not really needed, as it can be fixed with money (disk-drives or a RAID-box), but there are other applications where this isn't possible ('life' off-site mirrors for example, you can't buy or lease a 40MByte/s, SCSI-link with 500 Miles length ... the latency factor alone will kill you on every SCSI-transaction). And of cause it only works in read-mostly situations, but this is the case for most filesystems (most partitions on a unix system have this access pattern, obvious exceptions /tmp /var/spool/mqueue /var/spool/news ...) Stefan > > Peter > > -- > Peter Childs --- http://www.imforei.apana.org.au/~pjchilds > Finger pjchilds@al.imforei.apana.org.au for public PGP key > Drag me, drop me, treat me like an object! -- Stefan Grefen Am Grossberg 16, 55130 Mainz, Germany grefen@carpe.net +49 6131 998566 Fax:+49 6131 998568 Living on Earth may be expensive, but it includes an annual free trip around the Sun.