From owner-freebsd-scsi Sun Jun 9 03:54:16 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id DAA19266 for freebsd-scsi-outgoing; Sun, 9 Jun 1996 03:54:16 -0700 (PDT) Received: from silvia.HIP.Berkeley.EDU (silvia.HIP.Berkeley.EDU [136.152.64.181]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id DAA19208; Sun, 9 Jun 1996 03:54:09 -0700 (PDT) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.7.5/8.6.9) id DAA04925; Sun, 9 Jun 1996 03:54:01 -0700 (PDT) Date: Sun, 9 Jun 1996 03:54:01 -0700 (PDT) Message-Id: <199606091054.DAA04925@silvia.HIP.Berkeley.EDU> To: dror@hopf.dnai.com CC: freebsd-isp@freebsd.org, freebsd-scsi@freebsd.org In-reply-to: (message from Dror Matalon on Fri, 7 Jun 1996 12:28:22 -0700 (PDT)) Subject: Re: ERROR info:747d9d asc:11,0 Unrecovered read ...other SCSI issues From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * I'd love using some kind of RAID solution. So far I haven't found anything * that I really like, and that has been tested. Are you using the mylex * RAID? I looked at it a while ago and it sounded like an interesting solution, * but I first wanted to hear about other people using it. I also understand * that there are some problems with the freebsd utilities handling * large (20 Gig and more) disks, they report negative sizes etc * (32 bit problem?). This has been fixed in -current by Bruce Evans some time ago. (I'd bring in into -stable too if it's not soo deep in the kernel, but I'm not much of a kernel hacker....) * I actually think that someone could make nice bucks, by putting together * a FreeBsd box with redundant Power supplies, fans, Scsi controllers etc * and make it a, relatively, cheap NFS appliance. I'm working on a project doing just that (well, sort of). We are currently looking at FreeBSD as well as some other operating systems (duh). Right now we have the ccd driver doing striping and mirror (RAID0 and RAID1 for you paper-types). We are hoping to implement RAID4/5 parity some time this summer. Seems like you are in Berkeley too, maybe we can meet some time. We have enclosures with redundant power supply and such, it comes out to about +30% of the disk cost to set up a 64GB NFS server with P5-133 with regular striping. Satoshi From owner-freebsd-scsi Sun Jun 9 06:07:13 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id GAA09216 for freebsd-scsi-outgoing; Sun, 9 Jun 1996 06:07:13 -0700 (PDT) Received: from Sisyphos (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id GAA09182 for ; Sun, 9 Jun 1996 06:07:08 -0700 (PDT) Received: by Sisyphos id AA29281 (5.67b/IDA-1.5 for scsi@freebsd.org); Sun, 9 Jun 1996 15:06:57 +0200 Message-Id: <199606091306.AA29281@Sisyphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Sun, 9 Jun 1996 15:06:56 +0200 In-Reply-To: Michael Beckmann "Fast SCSI-2 = 6 MB/sec ?" (Jun 8, 16:39) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: Michael Beckmann Subject: Re: Fast SCSI-2 = 6 MB/sec ? Cc: scsi@freebsd.org Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Jun 8, 16:39, Michael Beckmann wrote: } Subject: Fast SCSI-2 = 6 MB/sec ? } I have just installed a Plextor SCSI CD ROM drive in my machine. I had to } find out that it supports only 6 MB/sec. Here is the output from dmesg: } (ncr0:0:0): 200ns (5 Mb/sec) offset 8. } (ncr0:0:0): "IBM OEM DFHSS4F 4141" type 0 fixed SCSI 2 } sd0(ncr0:0:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8. } (ncr0:2:0): 200ns (5 Mb/sec) offset 8. } (ncr0:2:0): "IBM OEM 0662S12 2 23" type 0 fixed SCSI 2 } sd1(ncr0:2:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8. } (ncr0:3:0): "PLEXTOR CD-ROM PX-4XCE 1.01" type 5 removable SCSI 2 } cd0(ncr0:3:0): FAST SCSI-2 175ns (6 Mb/sec) offset 8. } ^^^^^^^^^^ } cd0(ncr0:3:0): NOT READY asc:3a,0 Medium not present } (ncr0:5:0): "HP C1533A 9503" type 1 removable SCSI 2 } st0(ncr0:5:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8. } density code 0x13, variable blocks, write-enabled } I know that the CD ROM drive can never deliver a throughput of 6 MB/sec, } (actually, it should deliver around 680 kByte/sec.) but I wonder } if the CD ROM drive decreases the performance of the entire bus, thus } limiting the bus throughput to 6 MB/sec. Am I right ? Should I be } concerned about this and replace the CD drive ? } I wonder if my bus is running at 10 MByte/sec at all, because the hard } drives report both 5 MB/sec and 10 MB/sec. You don't need to worry, SCSI allows to talk to each device with its individual maximum speed. And the 6MB/s are quite impressive for a CDROM. They mean, that reading at 600KB/s will only need 10% of the available SCSI bus capacity, leaving 90% for the hard disk drives and the tape. The IBM disk drives have unique firmware features, which can cause problems at times. But the fact, that they first negotiate for 5MHz transfers when they actually are capable of 10MHz does no harm. (Well, it is an interaction with some code in the driver, which is intended towork around with other drive's bugs ... :) Anyway, all your drives are working with their maximum supported SCSI transfer rate. No reason to be worried ... Regards, STefan -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se From owner-freebsd-scsi Sun Jun 9 09:41:07 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id JAA14024 for freebsd-scsi-outgoing; Sun, 9 Jun 1996 09:41:07 -0700 (PDT) Received: from Sisyphos (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id JAA13904; Sun, 9 Jun 1996 09:40:51 -0700 (PDT) Received: by Sisyphos id AA01848 (5.67b/IDA-1.5); Sun, 9 Jun 1996 18:09:57 +0200 Message-Id: <199606091609.AA01848@Sisyphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Sun, 9 Jun 1996 18:09:57 +0200 In-Reply-To: Dave Hayes "Re: Using two NCR 53C825 controllers blows up?" (Jun 8, 16:27) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: Dave Hayes Subject: Re: Using two NCR 53C825 controllers blows up? Cc: scsi@freebsd.org, hackers@freebsd.org Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Jun 8, 16:27, Dave Hayes wrote: } Subject: Re: Using two NCR 53C825 controllers blows up? } "Michael L. VanLoon -- HeadCandy.com" writes: } >What kind of machine is this? } } Oops, sorry. It's not a name brand per se. It is an "Intel Advanced } ZP" motherboard. AMI Bios, and I don't have the revision handy. } } >However, the "second" card was actually installed with a lower address } >or IRQ or something. The only thing that made it come "after" the } >first was because it was hacked into the BIOS that way. After } >bootstrapping, NT doesn't use the BIOS anymore, and so would make the } >"second" controller into the first controller under NT, and the } >"first" controller would become the second. The drives on the } >controllers, of course, would then follow the new order. Very very } >annoying. } } That's real swell, now how do I find out whether it's doing that? } These NCR cards have jumpers to switch between INT A/B/C/D, but } I don't think PCI interrupts are as simple as ISA ones. Do I hafta } play with the jumpers? No, PLEASE always use INT A for a single function PCI device (like the NCR is) ! You may try swapping the locations of your controllers, or the direction of the PCI bus probe (it currently scans from low to high device numbers, as suggested in the PCI spec, but it seems there are PCI BIOS implementations that probe from high device numbers to low numbers). If there is one non-WIDE and one WIDE NCR card in a system, then the NCR SDMS software appears to always make the non-WIDE the primary controller (ie. will try to boot from it). Look into the LINT kernel config file for the syntax required to hardwire SCSI drives. This will permit you to specify drive names independently of the probe order. Let me know if all of this does not solve your problem ... Regards, STefan -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se From owner-freebsd-scsi Sun Jun 9 16:29:39 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id QAA10100 for freebsd-scsi-outgoing; Sun, 9 Jun 1996 16:29:39 -0700 (PDT) Received: from antares.aero.org (antares.aero.org [130.221.192.46]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id QAA10079 for ; Sun, 9 Jun 1996 16:29:37 -0700 (PDT) Received: from altair.aero.org (altair.aero.org [130.221.192.64]) by antares.aero.org (8.7.5/8.7.3) with SMTP id QAA10221 for ; Sun, 9 Jun 1996 16:29:04 -0700 (PDT) Message-Id: <199606092329.QAA10221@antares.aero.org> To: scsi@freebsd.org Subject: Toshiba XM-3501B scsi jumper config? Date: Sun, 09 Jun 1996 16:29:02 -0700 From: "Mike O'Brien" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Help help help! I'm trying to add a second drive to my FreeBSD machine and I want to change the SCSI ID of my CD-ROM drive from 1 to 2. I pulled the CD-ROM and the jumpers don't look like anything I've ever seen before. How do you set the SCSI ID on this drive? I've tried AltaVista and come up with zippo. Help! My machine's in pieces! Mike O'Brien obrien@aero.org From owner-freebsd-scsi Sun Jun 9 19:38:56 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id TAA22147 for freebsd-scsi-outgoing; Sun, 9 Jun 1996 19:38:56 -0700 (PDT) Received: from whizzo.transsys.com (whizzo.TransSys.COM [144.202.42.10]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id TAA22116 for ; Sun, 9 Jun 1996 19:38:50 -0700 (PDT) Received: from localhost.transsys.com (localhost.transsys.com [127.0.0.1]) by whizzo.transsys.com (8.7.5/8.7.3) with SMTP id WAA06736; Sun, 9 Jun 1996 22:38:22 -0400 (EDT) Message-Id: <199606100238.WAA06736@whizzo.transsys.com> X-Authentication-Warning: whizzo.transsys.com: Host localhost.transsys.com [127.0.0.1] didn't use HELO protocol To: "Mike O'Brien" cc: scsi@freebsd.org From: "Louis A. Mamakos" Subject: Re: Toshiba XM-3501B scsi jumper config? References: <199606092329.QAA10221@antares.aero.org> In-reply-to: Your message of "Sun, 09 Jun 1996 16:29:02 PDT." <199606092329.QAA10221@antares.aero.org> Date: Sun, 09 Jun 1996 22:38:21 -0400 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Help help help! I'm trying to add a second drive to my > FreeBSD machine and I want to change the SCSI ID of my CD-ROM drive > from 1 to 2. I pulled the CD-ROM and the jumpers don't look like > anything I've ever seen before. You probably don't actually have to change the SCSI target address on the CD to accomodate another SCSI read/write disk. Or are you adding a second CD? > How do you set the SCSI ID on this drive? I've tried AltaVista > and come up with zippo. Help! My machine's in pieces! I have a Toshiba XM-3601B drive, which may or may not be similar to yours. The 3601B is 4.4X speed drive. In any case, if you have an "internal" version of the drive, look at the back. On my drive, the goo on the back is (from left to right): - 3 pin AUDIO OUT connector - 7 position MODE SELECT header - 50 position SCSI bus connector - 4 position POWER connector The MODE SELECT header has 7 pairs of pins from left to right: ID1 ID2 ID3 PRTY (blank position) PRV/ALW TEST TERM ID1,ID2,ID3 encode the SCSI ID of the drive. Jumpering the pins results in a "1", leaving the jumper out results in a zero. So, SCSI ID 6 is ID1=out, ID2=IN ID4=IN. Inserting a jumper in the PRTY position causes the drive to check parity. It always generates parity. PRV/ALW - jumper removed allows CD tray to be ejected; jumper inserted prevents the tray from opening; the EJECT button is disabled. TEST - jumper out is normal. Jumper in turns on "Audio signal reproduction mode" which causes the SCSI bus to be ignored, and turns the drive into a walkman. TERM - jumper out shared term power not used. Jump in - shared term power used. Selects if drive provides term power to the SCSI bus or not. If you want to have the drive terminate (or not) the SCSI bus, then the termination resistor can be inserted (or not). Hope this helps. Louis Mamakos From owner-freebsd-scsi Sun Jun 9 21:48:14 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id VAA17895 for freebsd-scsi-outgoing; Sun, 9 Jun 1996 21:48:14 -0700 (PDT) Received: from antares.aero.org (antares.aero.org [130.221.192.46]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id VAA17870 for ; Sun, 9 Jun 1996 21:48:10 -0700 (PDT) Received: from altair.aero.org (altair.aero.org [130.221.192.64]) by antares.aero.org (8.7.5/8.7.3) with SMTP id VAA12314; Sun, 9 Jun 1996 21:47:39 -0700 (PDT) Message-Id: <199606100447.VAA12314@antares.aero.org> To: "Louis A. Mamakos" Cc: scsi@freebsd.org Subject: Re: Toshiba XM-3501B scsi jumper config? In-Reply-To: Your message of "Sun, 09 Jun 1996 19:38:21 PDT." <199606100238.WAA06736@whizzo.transsys.com> Date: Sun, 09 Jun 1996 21:47:37 -0700 From: "Mike O'Brien" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Thanks, that did the trick. I had to get out a powerful magnifying glass to see the markings for the pins. Also, the manufacturer had "parked" additional jumpers across adjacent pins on one side of the jumper set, which looked really screwball till I figured out they were non-functional. Mike From owner-freebsd-scsi Sun Jun 9 22:21:09 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id WAA29412 for freebsd-scsi-outgoing; Sun, 9 Jun 1996 22:21:09 -0700 (PDT) Received: from mail.barrnet.net (mail.barrnet.net [131.119.246.7]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id WAA29396 for ; Sun, 9 Jun 1996 22:21:07 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by mail.barrnet.net (8.7.5/MAIL-RELAY-LEN) with SMTP id WAA20548 for ; Sun, 9 Jun 1996 22:20:54 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id WAA15330; Sun, 9 Jun 1996 22:16:49 -0700 From: "Rodney W. Grimes" Message-Id: <199606100516.WAA15330@GndRsh.aac.dev.com> Subject: Re: Toshiba XM-3501B scsi jumper config? To: obrien@antares.aero.org (Mike O'Brien) Date: Sun, 9 Jun 1996 22:16:49 -0700 (PDT) Cc: louie@TransSys.COM, scsi@freebsd.org In-Reply-To: <199606100447.VAA12314@antares.aero.org> from Mike O'Brien at "Jun 9, 96 09:47:37 pm" X-Mailer: ELM [version 2.4ME+ PL11 (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 > Thanks, that did the trick. I had to get out a powerful > magnifying glass to see the markings for the pins. Also, the > manufacturer had "parked" additional jumpers across adjacent pins > on one side of the jumper set, which looked really screwball till > I figured out they were non-functional. It is quite common for vendors to ``park'' the extra jumpers on the ground side of the connectors. This makes it easy for OEM's such as myself to know what side of the connector pins are the ground side when hooking up external scsi id switches. It also saves us from digging in the parts drawers for spare jumpers ;-). The pins are not so much ``non-functional'', it just doesn't do anything when you hook 2 ground pins togeather :-) -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-scsi Mon Jun 10 08:27:42 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id IAA15935 for freebsd-scsi-outgoing; Mon, 10 Jun 1996 08:27:42 -0700 (PDT) Received: from luke.pmr.com (luke.pmr.com [206.224.65.132]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id IAA15897; Mon, 10 Jun 1996 08:27:24 -0700 (PDT) Received: (from bob@localhost) by luke.pmr.com (8.7.5/8.6.9) id KAA08933; Mon, 10 Jun 1996 10:26:32 -0500 (CDT) From: Bob Willcox Message-Id: <199606101526.KAA08933@luke.pmr.com> Subject: Re: ERROR info:747d9d asc:11,0 Unrecovered read error, other SCSI issues To: michaelv@HeadCandy.com (Michael L. VanLoon -- HeadCandy.com) Date: Mon, 10 Jun 1996 10:26:32 -0500 (CDT) Cc: wilko@yedi.iaf.nl, michael@memra.com, hdalog@zipnet.net, dror@hopf.dnai.com, freebsd-scsi@freebsd.org, freebsd-isp@freebsd.org In-Reply-To: <199606082053.NAA19410@MindBender.HeadCandy.com> from "Michael L. VanLoon -- HeadCandy.com" at "Jun 8, 96 01:52:04 pm" X-Mailer: ELM [version 2.4ME+ PL11 (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 Michael L. VanLoon -- HeadCandy.com wrote: > > > >> >Maybe Quantum's engineering does something wierd internally and doesn't > >> >test their drives on a real world activity mix that includes UNIX. > > >As Michael L. VanLoon -- HeadCandy.com wrote... > >> That would be hard to believe, considering that Quantum drives ship in > >> some HP Workstations, among others... > > >And in lots of Digital Equipment machines... > >In general: I have yet to hear of a manufacturer that never has 'junk' > >drive types every now and then. > > That has been my experience also, following these lists and Usenet. > > I think I have heard of bad runs of drives from every major > manufacturer. Especially the ones that try to push the limits, like > newer 7200rpm drives. My experience with Quantum drives has been limited to the LP240S, LP570S, and PD700S drives which I use for testing purposes. (Well, I also own a number of the DEC DSP series drives that are now made by Quantum, but I guess I don't think of them as Quantums.) I will say, that with of these particular models, they are usually the first drives to fail on a heavily loaded SCSI bus with a mix of devices. Seems to be an electrical thing, that they are more picky about bus loading than the other devices I generally use. Fact is, I often take advantage of this ``feature'' of the drives while testing, since I can count on them to intermittantly fail if I load the bus up enough with other devices. :-) -- Bob Willcox bob@luke.pmr.com Austin, TX From owner-freebsd-scsi Wed Jun 12 09:54:03 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id JAA16380 for freebsd-scsi-outgoing; Wed, 12 Jun 1996 09:54:03 -0700 (PDT) Received: from zit1.zit.th-darmstadt.de (root@zit1.zit.th-darmstadt.de [130.83.63.20]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id JAA16362; Wed, 12 Jun 1996 09:53:57 -0700 (PDT) Received: from [130.83.63.13] (apfel.zit.th-darmstadt.de [130.83.63.13]) by zit1.zit.th-darmstadt.de (8.7.5/8.6.9) with ESMTP id SAA25601; Wed, 12 Jun 1996 18:53:33 +0200 (MET DST) X-Sender: petzi@mail.zit.th-darmstadt.de Message-Id: In-Reply-To: <199606081427.QAA04029@yedi.iaf.nl> References: <199606070653.XAA08211@MindBender.HeadCandy.com> from "Michael L. VanLoon -- HeadCandy.com" at Jun 6, 96 11:52:53 pm Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 12 Jun 1996 00:22:05 +0200 To: Wilko Bulte From: Michael Beckmann Subject: Re: ERROR info:747d9d asc:11,0 Unrecovered read error, other SCSI issues Cc: freebsd-scsi@freebsd.org, freebsd-isp@freebsd.org Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >And in lots of Digital Equipment machines... Did you know that Quantum Atlas drives are based on DEC technology ? Quantum drives other than Atlas are not. I have made good experiences with Atlas drives, and would always recommend them. Michael From owner-freebsd-scsi Wed Jun 12 12:14:00 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA24436 for freebsd-scsi-outgoing; Wed, 12 Jun 1996 12:14:00 -0700 (PDT) Received: from vector.jhs.no_domain (slip139-92-18-135.pt.uk.ibm.net [139.92.18.135]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id MAA24404; Wed, 12 Jun 1996 12:13:28 -0700 (PDT) Received: (from jhs@localhost) by vector.jhs.no_domain (8.7.5/8.6.9) id RAA06093; Wed, 12 Jun 1996 17:18:52 +0200 (MET DST) Date: Wed, 12 Jun 1996 17:18:52 +0200 (MET DST) Message-Id: <199606121518.RAA06093@vector.jhs.no_domain> To: scsi@freebsd.org Cc: fabio@cesar.unicamp.br, fty@mcnc.org, gcrutchr@nightflight.com, j@uriah.heep.sax.de, jc@irbs.com, julian@freebsd.org, kuku@gilberto.physik.rwth-aachen.de, lehey.pad@sni.de, mrm@Sceard.com, nikm@ixa.net, tomppa@fidata.fi, wilko@yedi.iaf.nl, Scott Kelly , jhs@freebsd.org Subject: 8 * 0xFF bytes at intermittent multiples of 0x1000 From: "Julian H. Stacey" Reply-To: "Julian H. Stacey" Organization: Vector Systems Ltd. Address: Holz Strasse 27d, 80469 Munich, Germany Phone: +49.89.268616 Fax: +49.89.2608126 (pending modem change) Web: http://www.freebsd.org/~jhs/ Mailer: EXMH version 1.6.7 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk To scsi@freebsd.org Cc Adaptec 1542A SCSI Adapter People, Julian Elischer. [ I last posted to +1542A owners + bugs@ , but scsi@ now seems more appropriate than bugs@. I & some other 1542A people are most probably not on scsi@ list, so please be careful if trimming CC line. ] I (Julian Stacey ) did a load more hardware changes & tests, including swapping my Adaptec 1542A for a 1542B, & swapping sd0 & sd1, & eventually deduced it was not my 1542A that was mis-behaving, (returning 8 * 0xFF bytes at intermittent multiples of 0x1000), but was one of 2 HP 97548S SCSI 1 633MB disks. Either the disk is faulty, or maybe the scsi code might not be allowing for some strange sequence, or some such. __HOWEVER__ We can't dismiss it as an isolated equipment fault, as - tomppa@fidata.fi detects similar data corruptions, - scott@relay.forest.com seems to be having similar problems, but with a 1542B, - perhaps other people are suffering similar corruption without realising it. Partial Conclusion: 1542A people can `relax', to the extent that 1542B seems to be able to trigger the fault too (I don't have a1542C or 2940 etc) I've written a test program: testblock/ .c & .1 under my web page ~jhs/src/ . It merely reads & writes a large file in user mode, you don't need to be root & it does nothing nasty (except it will fill your file system with a single very large file, if you dont use `-l number_of_bytes' ) The previous owner of my disk was also a skilled FreeBSD person, he wasn't aware of a problem, (& I trust him on that ! :-) so it appears either my disk went bad when he transferred it to me, or the fault was always there, but that as it does not materialise too often, he didn't notice corruption caused. This could be a good reason for you to run my testblock.c even if you think you have no problem - think of it as a free disk check, that doesnt disrupt, no need to run dos, drop to debugger, be root, repartition, backup the file system or any other hastle :-). I don't have a scsi analyser equipment, & the bug does not cause a crash, I don't know what more I can do, except what I already am: (treating the disc as a backup with corruption guaranteed, but still usefull if another disc has a hard crash ! ) If any scsi people send me test code I'll happily compile & run it. The system exhibiting the phenomena is 2.1-Rel, but I have current src/ here too, & can easily cross compile & run a current kernel instead. ------------ > Date: Tue, 11 Jun 1996 16:56:50 -0400 > From: Scott Kelly > To: jhs@freebsd.org > Subject: Adaptec 1542A Users (from 12 Apr 1996) > > > I seem to be having similar problems, but with a 1542B... Do you know if there > has been a driver update since April? Don't know, cd /sys ; find . -type f -print | xargs grep 1542 I guess a scsi person might want to try: vi -c/1542 i386/conf/LINT i386/isa/aha1542.c i386/isa/isa.h \ scsi/README scsi/sd.c pci/ncr.c > I'm running 2.1... Me too (on the box in question). ------------ For reference, I'll append parts of my last mail: > Tomi Vainio > Has confirmed he sees the same Adaptec 1542A SCSI adapter bug that I do. > > > I connected sd1 to my 1542A and here are results: > > > > 1. No problems if testblock is only one that generates disk activity. > > 2. I launched couple find processes to sd0 and at same time I > > run testblock. Testblock failed only 1/10 of test runs. > > 3. I copied files with cp to sd1 when running testblock on > > sd1. Testblock failed on every time. > > > > Tomppa ........ > > > > ../testblock -v -l 10000000 /v/fish > > ../testblock: Neither -w or -r specified, so will both write then read. > > Using a block size of 61440, to a limit of 10000000. > > ../testblock writing then reading /v/fish. > > ../testblock: Started rewinding /v/fish. > > ../testblock: Finished rewinding /v/fish. > > ../testblock: In /v/fish, data mismatch at byte 49153 (0xc001), after 0 (0x0) previously checked ok. > > Byte read 255, byte expected 0 > > ../testblock: With /v/fish, only checked 0 bytes, 10,014,720 failed. > > ../testblock: Finished. ...... > > So it looks like a generic bug in FreeBSD code: > With a 1542A (& not a 1542B, which seems OK), > In simultaneous multiple task write mode to sd1 (or 2 or 3 or 4), > At random multiples of 0x1000 bytes, > The first 8 bytes of a block get forced to 0xFF. > (Of course it may well be that FreeBSD code is not `in error' but merely > doesnt allow for some wart in the 1542A, that's fixed in the 1542B, > but whatever, we need a fix). As above in this mail, I think I'm wrong there, it's not 1542A sepcific, I get it with 2 different 1542B's as well > Those who have not yet proven this on their system might like to try something > like this: > sync ; echo maybe even dump sd1 to tape # See below > cd <<>>/tmp > testblock -l 10000000 rubbish1 & > testblock -l 10000000 rubbish2 & > testblock -l 10000000 rubbish3 & > & do some other sd0 to sd1 copying in parallel. > Then run my 8f on all the data files youve run. > > Remember if you have a swap partition on sd1, & you swapped, > the swap may be damaged so you might crash. > If you'r really unlucky, while the system is creating new inodes for the > rubbish files, & is manipulating the file system, 8 bytes (out of several 0x1000) > bytes of file system structure data may get mangled. > > I have supplied CC readers with testblock.c & 8f.c, > for others interested, I'll toss them in http://www.freebsd.org/~jhs/src/ Since Done. Julian -- Julian H. Stacey jhs@freebsd.org http://www.freebsd.org/~jhs/ From owner-freebsd-scsi Thu Jun 13 00:52:36 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id AAA29494 for freebsd-scsi-outgoing; Thu, 13 Jun 1996 00:52:36 -0700 (PDT) Received: from proxy.siemens.at (proxy.siemens.at [192.138.228.19]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id AAA29309; Thu, 13 Jun 1996 00:51:00 -0700 (PDT) Received: from sol1.gud.siemens.co.at (sol-f.gud.siemens-austria) by proxy.siemens.at with SMTP id AA17515 (5.67a/IDA-1.5); Thu, 13 Jun 1996 09:48:44 +0200 Received: from ws2301.gud.siemens.co.at by sol1.gud.siemens.co.at with smtp (Smail3.1.28.1 #7 for ) id m0uU78y-00020DC; Thu, 13 Jun 96 09:48 MET DST Received: by ws2301.gud.siemens.co.at (1.37.109.16/1.37) id AA221852097; Thu, 13 Jun 1996 09:48:17 +0200 From: "Hr.Ladavac" Message-Id: <199606130748.AA221852097@ws2301.gud.siemens.co.at> Subject: Re: 8 * 0xFF bytes at intermittent multiples of 0x1000 To: jhs@freebsd.org Date: Thu, 13 Jun 1996 09:48:17 +0200 (MESZ) Cc: scsi@freebsd.org, fabio@cesar.unicamp.br, fty@mcnc.org, gcrutchr@nightflight.com, j@uriah.heep.sax.de, jc@irbs.com, julian@freebsd.org, kuku@gilberto.physik.rwth-aachen.de, lehey.pad@sni.de, mrm@sceard.com, nikm@ixa.net, tomppa@fidata.fi, wilko@yedi.iaf.nl, scott@relay.forest.com In-Reply-To: <199606121518.RAA06093@vector.jhs.no_domain> from "Julian H. Stacey" at Jun 12, 96 05:18:52 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 In his e-mail Julian H. Stacey wrote: > > To scsi@freebsd.org > Cc Adaptec 1542A SCSI Adapter People, Julian Elischer. > > > > I connected sd1 to my 1542A and here are results: > > > > > > 1. No problems if testblock is only one that generates disk activity. > > > 2. I launched couple find processes to sd0 and at same time I > > > run testblock. Testblock failed only 1/10 of test runs. > > > 3. I copied files with cp to sd1 when running testblock on > > > sd1. Testblock failed on every time. > > > > > > Tomppa It need not be FreeBSD specific. I had a very similar problem a couple of years ago with a 1542CF and Seagate 540 MB disk *WITH LINUX* The problems went away when I've thrown out the IDE disk on which I had the /usr partition (i.e., it was always in use.) I've used dd < /dev/zero and my own simple C program for testing. Errors appeared, but there was no regularity whatsoever. However, in my case the errors were simply inverted bits, 90% of the time 1 inverted bit, very seldom 2 bits. Since it never occured afterwards, I've dismissed it as the bus-noise generated by an ancient IDE device. Needless to say, the same machine, other machines with the very same Adaptec/Seagate combo behaved okay, even in presence of other SCSI devices. /Marino From owner-freebsd-scsi Sat Jun 15 15:44:15 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id PAA20330 for freebsd-scsi-outgoing; Sat, 15 Jun 1996 15:44:15 -0700 (PDT) Received: from vector.jhs.no_domain (slip139-92-42-189.ut.nl.ibm.net [139.92.42.189]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id PAA20258; Sat, 15 Jun 1996 15:42:32 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by vector.jhs.no_domain (8.7.5/8.6.9) with SMTP id WAA28493; Thu, 13 Jun 1996 22:29:16 +0200 (MET DST) Message-Id: <199606132029.WAA28493@vector.jhs.no_domain> X-Authentication-Warning: vector.jhs.no_domain: Host localhost [127.0.0.1] didn't use HELO protocol To: "Hr.Ladavac" cc: scsi@freebsd.org, fabio@cesar.unicamp.br, fty@mcnc.org, gcrutchr@nightflight.com, j@uriah.heep.sax.de, jc@irbs.com, julian@freebsd.org, kuku@gilberto.physik.rwth-aachen.de, lehey.pad@sni.de, mrm@sceard.com, nikm@ixa.net, tomppa@fidata.fi, wilko@yedi.iaf.nl, scott@relay.forest.com Subject: Re: 8 * 0xFF bytes at intermittent multiples of 0x1000 From: "Julian H. Stacey" Reply-To: "Julian H. Stacey" Organization: Vector Systems Ltd. Address: Holz Strasse 27d, 80469 Munich, Germany Phone: +49.89.268616 Fax: +49.89.2608126 (later) Web: http://www.freebsd.org/~jhs/ Mailer: EXMH 1.6.7, PGP available In-reply-to: Your message of "Thu, 13 Jun 1996 09:48:17 +0200." <199606130748.AA221852097@ws2301.gud.siemens.co.at> Date: Thu, 13 Jun 1996 22:29:15 +0200 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, Reference: > From: "Hr.Ladavac" > In his e-mail Julian H. Stacey wrote: > > To scsi@freebsd.org > > Cc Adaptec 1542A SCSI Adapter People, Julian Elischer. > > > > > > I connected sd1 to my 1542A and here are results: > > > > > > > > 1. No problems if testblock is only one that generates disk activity. > > > > 2. I launched couple find processes to sd0 and at same time I > > > > run testblock. Testblock failed only 1/10 of test runs. > > > > 3. I copied files with cp to sd1 when running testblock on > > > > sd1. Testblock failed on every time. > > > > > > > > Tomppa > > It need not be FreeBSD specific. I had a very similar problem a couple of > years ago with a 1542CF and Seagate 540 MB disk *WITH LINUX* Pretty scary if we're still in the same dumb scenario we were in 20 years ago, back then it was `Unix assumes perfect hardware' If we're still doing no checks on what the hard disks deliver, it's time to shiver with fear ;-) (though I realise implications WRT FFS definition & portability if we were to add checks, tricky that.) > The problems went away when I've thrown out the IDE disk on which I had > the /usr partition (i.e., it was always in use.) I've used dd < /dev/zero > and my own simple C program for testing. I have no IDE :-) My program started years on a 4.3 with no /dev/zero, & now does a rolling bytes offset within blocks, to detect skipped blocks. > Errors appeared, but there was no regularity whatsoever. Mine is regular 8 * 0xFF every few 0x1000 > However, in my > case the errors were simply inverted bits, 90% of the time 1 inverted bit, > very seldom 2 bits. Sounds like no commonality. > Since it never occured afterwards, I've dismissed it as the bus-noise > generated by an ancient IDE device. Mine is a short internal ribbon, good firm contacts, properly terminated. > Needless to say, the same machine, other machines with the very same > Adaptec/Seagate combo behaved okay, even in presence of other SCSI devices. Well at least one other person is seeing exactly what I'm seeing he posted his log. Julian -- Julian H. Stacey jhs@freebsd.org http://www.freebsd.org/~jhs/