From owner-freebsd-scsi Mon Jan 20 06:33:57 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.4/8.8.4) id GAA03332 for freebsd-scsi-outgoing; Mon, 20 Jan 1997 06:33:57 -0800 (PST) Received: from mixer.visi.com (root@mixer.visi.com [204.73.178.1]) by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id GAA03324 for ; Mon, 20 Jan 1997 06:33:53 -0800 (PST) Received: from undquirt (undquirt.visi.com [206.11.194.26]) by mixer.visi.com (8.8.4/8.7.5) with ESMTP id IAA21644 for ; Mon, 20 Jan 1997 08:33:49 -0600 (CST) From: "Brent J. Nordquist" Posted-Date: Mon, 20 Jan 1997 08:33:49 -0600 (CST) Received: (from bjn@localhost) by undquirt (8.8.4/8.7.3) for freebsd-scsi@freebsd.org id IAA01384; Mon, 20 Jan 1997 08:32:57 -0600 (CST) Message-Id: <199701201432.IAA01384@undquirt> Subject: Boot with amd0 driver jumps into debugger To: freebsd-scsi@freebsd.org Date: Mon, 20 Jan 1997 08:32:57 -0600 (CST) Reply-to: bjn@visi.com (Brent J. Nordquist) 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 have a Zeos machine whose motherboard has a built-in AMD SCSI controller. However, I have a much better Adaptec 2940UW adapter card installed, so the AMD controller has nothing attached to it and I have it disabled in the BIOS. Until this weekend, I was using FreeBSD 2.2-current circa Aug 96. It used to print "pci0:9 ... AMD ... [no driver assigned]", then it would find the ahc0 driver and all would be well. After a cvsup to tag=RELENG_2_2 this weekend, I noticed that there is a new driver amd0 which is listed first in the default kernel config. file. Booting with this driver installed in that order results in dropping into the debugger. (I can produce the exact output if someone wants it.) To fix this for me, I just disabled amd0 and rebuilt my kernel, so I'm OK now. But I wanted to flag this for someone's attention, so that others would be able to use FreeBSD-2.2 out of the box without a kernel recompile. (To wit, how do you do a kernel recompile when you can't boot/install off the floppy?) Thanks! -- Brent J. Nordquist bjn@visi.com +1 612 827-2747 From owner-freebsd-scsi Mon Jan 20 18:42:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.4/8.8.4) id SAA21431 for freebsd-scsi-outgoing; Mon, 20 Jan 1997 18:42:48 -0800 (PST) Received: from mixer.visi.com (root@mixer.visi.com [204.73.178.1]) by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id SAA21425 for ; Mon, 20 Jan 1997 18:42:46 -0800 (PST) Received: from undquirt (undquirt.visi.com [206.11.194.26]) by mixer.visi.com (8.8.4/8.7.5) with ESMTP id UAA03720 for ; Mon, 20 Jan 1997 20:42:36 -0600 (CST) From: "Brent J. Nordquist" Posted-Date: Mon, 20 Jan 1997 20:42:36 -0600 (CST) Received: (from bjn@localhost) by undquirt (8.8.4/8.7.3) for freebsd-scsi@freebsd.org id UAA00309; Mon, 20 Jan 1997 20:41:39 -0600 (CST) Message-Id: <199701210241.UAA00309@undquirt> Subject: Re: Boot with amd0 driver jumps into debugger To: freebsd-scsi@freebsd.org Date: Mon, 20 Jan 1997 20:41:39 -0600 (CST) Reply-to: bjn@visi.com (Brent J. Nordquist) 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 Doh! I wasn't subscribed to freebsd-scsi. Sorry to ask, but if anyone replied to the mailing list and didn't copy me, could you please send your message to me directly? Thanks! (I'm subscribed now.) -- Brent J. Nordquist bjn@visi.com +1 612 827-2747 From owner-freebsd-scsi Tue Jan 21 02:56:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.4/8.8.4) id CAA13588 for freebsd-scsi-outgoing; Tue, 21 Jan 1997 02:56:38 -0800 (PST) Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58]) by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id CAA13490; Tue, 21 Jan 1997 02:55:18 -0800 (PST) Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.8.3/8.8.3) id MAA19603; Tue, 21 Jan 1997 12:55:03 +0200 (SAT) From: John Hay Message-Id: <199701211055.MAA19603@zibbi.mikom.csir.co.za> Subject: NCR SCSI problem with HP C1553A tape jukebox To: se@FreeBSD.ORG, FreeBSD-SCSI@FreeBSD.ORG Date: Tue, 21 Jan 1997 12:55:03 +0200 (SAT) X-Mailer: ELM [version 2.4ME+ PL24 (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 have been using the HP C1553A tape jukebox with the NCR SCSI controller for a while on FreeBSD 2.1.x on a Pentium machine and it was working fine. Now I have upgraded the machine to a 200MHz Pentium Pro and I am getting "st0(ncr0:5:0): COMMAND FAILED (4 88) @f112ce00." messages when using the tape. It will work for a while and then give this error at random places. I am still using the exact same controller and cable and the jukebox is in an external box. It is the only device on that SCSI bus and it is terminated on the controller and at the end of the cable. I had a look at the status codes and the 4 seems to indicate that the host was finished, but the 88 does not seem to exist in the S_xxxxx status codes or can you OR the bits? Is it possible that there is a timing problem in the driver somewhere? Have someone else came across this problem with a solution maybe? Thanks John -- John Hay -- John.Hay@mikom.csir.co.za From owner-freebsd-scsi Tue Jan 21 13:33:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.4/8.8.4) id NAA15287 for freebsd-scsi-outgoing; Tue, 21 Jan 1997 13:33:55 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.4/8.8.4) with SMTP id NAA15232; Tue, 21 Jan 1997 13:32:58 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id WAA18036; Tue, 21 Jan 1997 22:31:48 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.4/8.6.9) id WAA23372; Tue, 21 Jan 1997 22:20:34 +0100 (MET) Message-ID: Date: Tue, 21 Jan 1997 22:20:34 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: jhay@zibbi.mikom.csir.co.za (John Hay) Cc: se@FreeBSD.ORG, FreeBSD-SCSI@FreeBSD.ORG Subject: Re: NCR SCSI problem with HP C1553A tape jukebox References: <199701211055.MAA19603@zibbi.mikom.csir.co.za> X-Mailer: Mutt 0.55-PL10 Mime-Version: 1.0 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 Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199701211055.MAA19603@zibbi.mikom.csir.co.za>; from John Hay on Jan 21, 1997 12:55:03 +0200 Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As John Hay wrote: > I have been using the HP C1553A tape jukebox with the NCR SCSI controller > for a while This drive is known to not grok tagged commands. Use options FAILSAFE. -- 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 Tue Jan 21 15:52:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.4/8.8.4) id PAA23259 for freebsd-scsi-outgoing; Tue, 21 Jan 1997 15:52:00 -0800 (PST) Received: from Sisyphos.MI.Uni-Koeln.DE (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.8.4/8.8.4) with SMTP id PAA23229; Tue, 21 Jan 1997 15:51:01 -0800 (PST) Received: from x14.mi.uni-koeln.de (annexr3-11.slip.Uni-Koeln.DE) by Sisyphos.MI.Uni-Koeln.DE with SMTP id AA16649 (5.67b/IDA-1.5); Wed, 22 Jan 1997 00:50:49 +0100 Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.4/8.6.9) id AAA23926; Wed, 22 Jan 1997 00:50:46 +0100 (CET) Message-Id: Date: Wed, 22 Jan 1997 00:50:46 +0100 From: se@freebsd.org (Stefan Esser) To: jhay@zibbi.mikom.csir.co.za (John Hay) Cc: se@freebsd.org, FreeBSD-SCSI@freebsd.org Subject: Re: NCR SCSI problem with HP C1553A tape jukebox References: <199701211055.MAA19603@zibbi.mikom.csir.co.za> X-Mailer: Mutt 0.55-PL15 Mime-Version: 1.0 In-Reply-To: <199701211055.MAA19603@zibbi.mikom.csir.co.za>; from John Hay on Jan 21, 1997 12:55:03 +0200 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Jan 21, jhay@zibbi.mikom.csir.co.za (John Hay) wrote: > Hi, > > I have been using the HP C1553A tape jukebox with the NCR SCSI controller > for a while on FreeBSD 2.1.x on a Pentium machine and it was working > fine. Now I have upgraded the machine to a 200MHz Pentium Pro and I am > getting "st0(ncr0:5:0): COMMAND FAILED (4 88) @f112ce00." messages > when using the tape. It will work for a while and then give this error > at random places. I am still using the exact same controller and Could you please send complete error messages from the message log ? > cable and the jukebox is in an external box. It is the only device > on that SCSI bus and it is terminated on the controller and at the > end of the cable. > > I had a look at the status codes and the 4 seems to indicate that > the host was finished, but the 88 does not seem to exist in the > S_xxxxx status codes or can you OR the bits? Yes, you can :) (Well, kind of: The 0x80 is a special case, which is used internally by the driver.) > Is it possible that there is a timing problem in the driver somewhere? Not very likely. There are no timing loops, except when the chip has to be reset (which should not happen under normal use.) > Have someone else came across this problem with a solution maybe? There have been reports of problems with other PCI bus-master cards on Natoma motherboards. But I need more information in order to understand what's wrong ... Regards, STefan From owner-freebsd-scsi Tue Jan 21 22:36:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA17937 for freebsd-scsi-outgoing; Tue, 21 Jan 1997 22:36:00 -0800 (PST) Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA17882; Tue, 21 Jan 1997 22:33:35 -0800 (PST) Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.8.3/8.8.3) id IAA01199; Wed, 22 Jan 1997 08:33:02 +0200 (SAT) From: John Hay Message-Id: <199701220633.IAA01199@zibbi.mikom.csir.co.za> Subject: Re: NCR SCSI problem with HP C1553A tape jukebox In-Reply-To: from Stefan Esser at "Jan 22, 97 00:50:46 am" To: se@freebsd.org (Stefan Esser) Date: Wed, 22 Jan 1997 08:33:01 +0200 (SAT) Cc: FreeBSD-SCSI@freebsd.org X-Mailer: ELM [version 2.4ME+ PL24 (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 > On Jan 21, jhay@zibbi.mikom.csir.co.za (John Hay) wrote: > > Hi, > > > > I have been using the HP C1553A tape jukebox with the NCR SCSI controller > > for a while on FreeBSD 2.1.x on a Pentium machine and it was working > > fine. Now I have upgraded the machine to a 200MHz Pentium Pro and I am > > getting "st0(ncr0:5:0): COMMAND FAILED (4 88) @f112ce00." messages > > when using the tape. It will work for a while and then give this error > > at random places. I am still using the exact same controller and > > Could you please send complete error messages from the message log ? OK I'll attach the output of dmesg at the end. I had a look at /var/log/ messages and it is the same except for other messages by named and other daemons that looks perfectly normal. I did boot with -v just incase it gives some usefull extra info. There are no other messages to indicate an error except the "COMMAND FAILED" message. Tar just reports: "tar: can't write to /dev/nrst0 : Input/output error" > Not very likely. There are no timing loops, except when the chip has > to be reset (which should not happen under normal use.) Hmmm. I think you are right. I did a tar again and started ncrcontrol in another window. It seems like the data stops flowing long (like tens of seconds) before the error messages pops out. It is more like it missed an interrupt or something else. > There have been reports of problems with other PCI bus-master cards > on Natoma motherboards. But I need more information in order to > understand what's wrong ... Well mine is a Natoma. :-( Thanks. John -- John Hay -- John.Hay@mikom.csir.co.za n 21 15:02:30 SAT 1997 jhay@zibbi.mikom.csir.co.za:/a/FreeBSD/sys/compile/ZIBBI CPU: 199-MHz unknown (Pentium-class CPU) Origin = "GenuineIntel" Id = 0x617 Stepping=7 Features=0xf9ff,MTRR,PGE,MCA,CMOV> real memory = 67108864 (65536K bytes) avail memory = 64155648 (62652K bytes) pcibus_setup(1): mode1res=0x80000000 (0x80000000), mode2res=0xff (0x0e) pcibus_setup(2): mode1res=0x80000000 (0x80000000) pcibus_check: device 0 is there (id=12378086) Probing for devices on PCI bus 0: configuration mode 1 allows 32 devices. chip0 rev 2 on pci0:0 chip1 rev 1 on pci0:7:0 chip2 rev 0 on pci0:7:1 mapreg[20] type=1 addr=0000ffa0 size=0010. de0 rev 17 int a irq 9 on pci0:11 mapreg[10] type=1 addr=0000f480 size=0080. mapreg[14] type=0 addr=ffbde800 size=0080. reg16: ioaddr=0xf480 size=0x80 de0: SMC 9332 DC21140 [10-100Mb/s] pass 1.1 de0: address 00:00:c0:58:0e:9c de0: enabling 10baseT port ncr0 rev 2 int a irq 11 on pci0:15 mapreg[10] type=1 addr=0000f800 size=0100. mapreg[14] type=0 addr=ffbdec00 size=0100. reg20: virtual=0xf570bc00 physical=0xffbdec00 size=0x100 ncr0: restart (scsi reset). ncr0 scanning for targets 0..6 (V2 pl23 95/09/07) (ncr0:5:0): "HP C1553A 9503" type 1 removable SCSI 2 st0(ncr0:5:0): Sequential-Access st0(ncr0:5:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8. density code 0x24, drive empty ahc0 rev 0 int a irq 10 on pci0:17 mapreg[10] type=1 addr=0000fc00 size=0100. mapreg[14] type=0 addr=ffbdf000 size=1000. ahc0: Reading SEEPROM...done. ahc0: aic7880 Wide Channel, SCSI Id=7, 16 SCBs ahc0: Reseting Channel A ahc0: Downloading Sequencer Program...Done ahc0: Probing channel A ahc0: target 0 synchronous at 10.0MHz, offset = 0xf (ahc0:0:0): "CONNER CFP2105S 2.14GB 2B4B" type 0 fixed SCSI 2 sd0(ahc0:0:0): Direct-Access 2048MB (4194304 512 byte sectors) sd0(ahc0:0:0): with 3940 cyls, 10 heads, and an average 106 sectors/track ahc0: target 2 synchronous at 10.0MHz, offset = 0xf (ahc0:2:0): "CONNER CFP1060S 1.05GB 203C" type 0 fixed SCSI 2 sd1(ahc0:2:0): Direct-Access 1013MB (2074880 512 byte sectors) sd1(ahc0:2:0): with 2756 cyls, 8 heads, and an average 94 sectors/track ahc0: target 4 synchronous at 20.0MHz, offset = 0xf (ahc0:4:0): "SEAGATE ST32155N 0532" type 0 fixed SCSI 2 sd2(ahc0:4:0): Direct-Access 2049MB (4197405 512 byte sectors) sd2(ahc0:4:0): with 4177 cyls, 8 heads, and an average 125 sectors/track ahc0: target 5 synchronous at 20.0MHz, offset = 0xf (ahc0:5:0): "SEAGATE ST32155N 0532" type 0 fixed SCSI 2 sd3(ahc0:5:0): Direct-Access 2049MB (4197405 512 byte sectors) sd3(ahc0:5:0): with 4177 cyls, 8 heads, and an average 125 sectors/track pci0: uses 4480 bytes of memory from ffbde800 upto ffbdffff. pci0: uses 656 bytes of I/O space from f480 upto ffaf. Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> ed0 not found at 0x280 sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 72065B fd0: 1.44MB 3.5in npx0 on motherboard npx0: INT 16 interface Device configuration finished. Considering FFS root f/s. Configuring root and swap devs. configure() finished. BIOS Geometries: 0:0104fe3f 0..260=261 cylinders, 0..254=255 heads, 1..63=63 sectors 1:03f43f20 0..1012=1013 cylinders, 0..63=64 heads, 1..32=32 sectors 2:0104fe3f 0..260=261 cylinders, 0..254=255 heads, 1..63=63 sectors 3:0104fe3f 0..260=261 cylinders, 0..254=255 heads, 1..63=63 sectors 0 accounted for sd0s1: type 0xa5, start 63, end = 4191263, size 4191201 : OK sd2s1: type 0xa5, start 63, end = 4192964, size 4192902 : OK sd3s1: type 0xa5, start 125, end = 4196999, size 4196875 sd3s1: C/H/S start 0/1/1 (61) != start 125: invalid st0(ncr0:5:0): COMMAND FAILED (4 88) @f112ce00. st0(ncr0:5:0): COMMAND FAILED (4 88) @f112ce00. From owner-freebsd-scsi Wed Jan 22 09:46:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA16266 for freebsd-scsi-outgoing; Wed, 22 Jan 1997 09:46:22 -0800 (PST) Received: from prometeo.prometeo.it (prometeo.prometeo.it [194.21.36.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA16259; Wed, 22 Jan 1997 09:46:07 -0800 (PST) Received: from tgioc.prometeo.it (tgioc.prometeo.it [194.21.36.34]) by prometeo.prometeo.it (8.7.5/8.7.3) with SMTP id SAA24778; Wed, 22 Jan 1997 18:53:07 +0100 (MET) Message-Id: <199701221753.SAA24778@prometeo.prometeo.it> Comments: Authenticated sender is From: "Coi Giovanni" Organization: Prometeo srl To: freebsd-question@FreeBSD.org, freebsd-scsi@FreeBSD.org Date: Wed, 22 Jan 1997 18:40:27 +0000 Subject: AHA-1522A and DAT Data phase error Reply-to: Coi_Giovanni@prometeo.it X-Confirm-Reading-To: Coi_Giovanni@prometeo.it X-pmrqc: 1 Priority: normal X-mailer: Pegasus Mail for Win32 (v2.50) Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Hello, I have a SONY SDT-5200 3.26 tape backup (DAT) which I am trying to run. The SCSI controller is a AHA-1522A (ISA bus) when I try to write on tape (with dump for example) I get the following error messages on console: st0(aic0:5:0): ABORTED COMMAND info:4 asc:4b,0 Data phase error st0(aic0:5:0): HARDWARE FAILURE info:14 asc:44,0 Internal target failure st0(aic0:5:0): HARDWARE FAILURE info:14 asc:44,0 Internal target failure st0(aic0:5:0): HARDWARE FAILURE info:1 asc:44,0 Internal target failure the dump command is dump 0uBbf 2000000 10 /dev/rst0 /dev/wd0a but I try also with dump 0uBbf 2000000 10 /dev/rst0 / whit the same results and this is the output of dump on stderr: DUMP: Date of this level 0 dump: Wed Jan 22 13:52:54 1997 DUMP: Date of last level 0 dump: the epoch DUMP: Dumping /dev/rwd0a (/) to /dev/rst0 DUMP: mapping (Pass I) [regular files] DUMP: mapping (Pass II) [directories] DUMP: estimated 15086 tape blocks on 0.01 tape(s). DUMP: dumping (Pass III) [directories] DUMP: dumping (Pass IV) [regular files] ... error messages on console.... DUMP: End of tape detected DUMP: Closing /dev/rst0 DUMP: Change Volumes: Mount volume #2 DUMP: Is the new volume mounted and ready to go?: ("yes" or "no") I reply no DUMP: Do you want to abort?: ("yes" or "no") I reply yes DUMP: The ENTIRE dump is aborted. Note: a) after I run the dump command after signaling DUMP: dumping (Pass IV) [reguler files] the system freeze. It do not respond at console, no on other telnet session, no on other vitrual console. After it finish (with error) all return to normal state. b) I have tryed the following command: tar cvf /dev/rst0 /etc and it goes. Then I can run also tar tvf /dev/rst0 and see the correct listing of etc's file. I think it was because there are few data in /etc. c) the _same_ DAT works fine when I connect it to a windows NT 4.0 box (but with a SCSI controller more quick, an PCI Adaptec AHA-2940 d) as you can see from the startup log (below) I have also a SCSI CDROM that work fine (I have copied with success a file of about 10Mbyte from it). Anyone has some suggestions? Please reply directly to me or on freebsd-question list. TIA. Follow my system configuration parameter: --------------------------------------------------------------- controller aic0 at isa? port 0x140 bio irq 10 vector aicintr controller scbus0 at aic0 # Single bus device disk sd0 at scbus0 target 4 unit 0 tape st0 at scbus0 target 5 unit 0 device cd0 at scbus0 target 3 unit 0 options SCSI_DELAY=30 options SCSIDEBUG --------------------------------------------------------------- Follow my startup log ----------------------------------------------------------- FreeBSD 2.1.5-RELEASE #0: Wed Jan 22 15:56:04 MET 1997 CPU: i486DX (486-class CPU) real memory = 16777216 (16384K bytes) avail memory = 14835712 (14488K bytes) Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA mono <4 virtual consoles, flags=0x0> ed0 at 0x340-0x35f irq 5 on isa ed0: address 00:00:1b:3a:c4:88, type NE2000 (16 bit) lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16450 sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16450 aic0 at 0x140-0x15f irq 10 on isa aic0 waiting for scsi devices to settle (aic0:3:0): "NEC CD-ROM DRIVE:501 2.2" type 5 removable SCSI 2 cd0(aic0:3:0): CD-ROM cd present.[328201 x 2048 byte records] (aic0:4:0): "IBM MTA-3230TC2210!B 0" type 7 removable SCSI 2 uk0(aic0:4:0): Unknown (aic0:5:0): "SONY SDT-5200 3.26" type 1 removable SCSI 2 st0(aic0:5:0): Sequential-Access density code 0x13, 512-byte blocks, write-enabled wdc0 at 0x1f0-0x1f7 irq 14 flags 0x80008000 on isa wdc0: unit 0 (wd0): wd0: 814MB (1667232 sectors), 1654 cyls, 16 heads, 63 S/T, 512 B/S wdc0: unit 1 (wd1): wd1: 515MB (1056384 sectors), 1048 cyls, 16 heads, 63 S/T, 512 B/S fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 765 fd0: 1.44MB 3.5in mcd0: type Mitsumi FX001D, version info: D 2 mcd0 at 0x320- 0x323 irq 11 on isa npx0 on motherboard npx0: INT 16 interface ----------------------------------------------------------- ---------------------------------------------------------- Prometeo srl - Progetti e Metodologie Informatiche Internet Services Provider Coi Giovanni Voice : +39 (041)5701366 Via Giudecca 15 Fax : +39 (041)5701005 30035 MIRANO (VE) - ITALY e-mail: coi@prometeo.it http://www.prometeo.it From owner-freebsd-scsi Thu Jan 23 11:51:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA08008 for freebsd-scsi-outgoing; Thu, 23 Jan 1997 11:51:11 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA08003 for ; Thu, 23 Jan 1997 11:51:07 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id UAA02682; Thu, 23 Jan 1997 20:50:42 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.4/8.6.9) id UAA01576; Thu, 23 Jan 1997 20:35:14 +0100 (MET) Message-ID: Date: Thu, 23 Jan 1997 20:35:14 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: chris@bb.cc.wa.us (Chris Coleman) Cc: freebsd-scsi@FreeBSD.org (FreeBSD SCSI list) Subject: Re: Tape Backup Drive Not working. References: X-Mailer: Mutt 0.55-PL10 Mime-Version: 1.0 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 Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Chris Coleman on Jan 21, 1997 17:38:02 -0800 Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk (Btw., your mail was mis-referenced. I've also moved this to the SCSI list.) As Chris Coleman wrote: > I have a 4gig SCSI tape backup drive. Which one? (Probe string) > /kernel st0 (ahc:0:3:0): ILLEGAL REQUEST asc: 24,0 Invalid field in CDB > /kernel st0: bad request, must be between 0 and 0 > /kernel st0 (ahc:0:3:0): ILLEGAL REQUEST asc: 24,0 Invalid field in CDB Which blocksize setting? (output of `mt status') -- 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 Thu Jan 23 18:41:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA02037 for freebsd-scsi-outgoing; Thu, 23 Jan 1997 18:41:32 -0800 (PST) Received: from aries.bb.cc.wa.us (root@[208.8.136.11]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA02032 for ; Thu, 23 Jan 1997 18:41:29 -0800 (PST) Received: from localhost (chris@localhost) by aries.bb.cc.wa.us (8.8.3/8.6.9) with SMTP id SAA16763; Thu, 23 Jan 1997 18:36:02 -0800 (PST) Date: Thu, 23 Jan 1997 18:36:01 -0800 (PST) From: Chris Coleman Reply-To: Chris Coleman To: Joerg Wunsch cc: FreeBSD SCSI list Subject: Re: Tape Backup Drive Not working. In-Reply-To: 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 Thu, 23 Jan 1997, J Wunsch wrote: > (Btw., your mail was mis-referenced. I've also moved this to the > SCSI list.) Thanks, i didnt even think of that, I only subscribe to -hackers. > > > I have a 4gig SCSI tape backup drive. > Which one? (Probe string) "Conner CTT8000-S 1.17" Type 1 removable SCSI2 st0(ahc:0:3:0):Sequential-Access Density Code 0x45, Drive Empty > > > /kernel st0 (ahc:0:3:0): ILLEGAL REQUEST asc: 24,0 Invalid field in CDB > > /kernel st0: bad request, must be between 0 and 0 > > /kernel st0 (ahc:0:3:0): ILLEGAL REQUEST asc: 24,0 Invalid field in CDB > > Which blocksize setting? (output of `mt status') > -- mt status: Present Mode: Density = 0x00 Blocksize Variable ----------Available Modes---------------- Mode0: Density 0x00 Blocksize Variable Mode1: Density 0x00 Blocksize Variable Mode2: Density 0x00 Blocksize Variable Mode3: Density 0x00 Blocksize Variable Thanks, I was afraid i wasn't going to get an answer. Chris Coleman (chris@aries.bb.cc.wa.us) Computer Support Technician I (509)-766-8873 Big Bend Community College Internet Instructor Death is life's way of telling you you're fired. From owner-freebsd-scsi Fri Jan 24 01:50:52 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA18115 for freebsd-scsi-outgoing; Fri, 24 Jan 1997 01:50:52 -0800 (PST) Received: from relay.nuxi.com (nuxi.ucdavis.edu [128.120.37.176]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA18109 for ; Fri, 24 Jan 1997 01:50:50 -0800 (PST) Received: from dragon.nuxi.com (reqd-091.ucdavis.edu [128.120.251.211]) by relay.nuxi.com (8.7.6/8.6.12) with ESMTP id BAA06439 for ; Fri, 24 Jan 1997 01:51:41 -0800 (PST) Received: (from obrien@localhost) by dragon.nuxi.com (8.8.4/8.7.3) id BAA00430; Fri, 24 Jan 1997 01:50:39 -0800 (PST) Message-ID: Date: Fri, 24 Jan 1997 01:50:38 -0800 From: obrien@dragon.cs.ucdavis.edu (David O'Brien) To: freebsd-scsi@freebsd.org Subject: unstable 2.2-BETA X-Mailer: Mutt 0.57-PL4 Mime-Version: 1.0 Reply-To: deobrien@ucdavis.edu Organization: The NUXI *BSD group X-PGP-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Keyid: 34F9F9D5 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi all, I'm sorry to bother you, but I've got a very unstable 2.2-BETA box right now. I'm believe my problems to be SCSI subsystem related, but I could be wrong. Is the below something to to worry about? (ie, is my SCSI disk going bad, or is this a kernel problem? ------------------------------------------------------------------------------ sd0(ahc0:0:0): partity error during Data-In phase ahc0: ahc)intr reference scb not valid during scsiint 0x17 scb(1) sd0(ahc0:0:0): timed out in datain phase, SCSISIGI==0xb6 SEQADDR == 0x13c ahc0: Issued Channel A Bus Reset. 1 SCBs aborted sd0(ahc0:0:0): UNIT ATTENTION asc:29,0 sd0(ahc0:0:0): Power on, reset, or bus device reset occured, retries 3 ------------------------------------------------------------------------------ I upgraded my 486/66 AHA-2840 to a Tyan/AMD-K5 AHA-2940U about two weeks ago. At first I wasn't having stability problems with the AHA-2940U. But in the last week my machine has become quite unstatable. Since then I filled my root partition to 110%, and recompiled a new kernel. I took some SCSI devices out of the chain, used some different cables, and moved the AHA-2940U to a different PCI slot. I haven't gotten the kernel messages in many days. Instead I now get spontaniously reboots. And I haven't seen any of the bus reset errors that I did at first. It seems to happen when I'm running kernel PPP using /usr/sbin/pppd /dev/modem 38400 debug -vj crtscts asyncmap 0 \ noipdefault defaultroute netmask 255.255.255.0 I'm usually running X and the machine will just totally stop *everything*. After about 3 seconds, it just reboots. No kernel messages in /var/log/* or anything. I've since put things back the way they were at first with the Tyan/AMD-K5 AHA-2940U, but that didn't help anything. I built even built a new kernel from priesteen 2.2-BETA sources with Justin's fixes to: src/sys/dev/aic7xxx/aic7xxx.seq src/sys/dev/aic7xxx/aic7xxx_reg.h src/sys/i386/scsi/aic7xxx.c src/sys/pci/aic7870.c But that didn't change anything. I've got / and /usr on /dev/sd0s1. As an experiment, I also installed 2.2-BETA on an IDE drive and then mount /dev/sd0s1b for additional swap and /dev/sd0s1e for home dirs. This setup is much more stable, but I still get some spontainious reboots. I get non of this from 2.2-961014-SNAP with /, /usr and some swap on an IDE drive and addition swap from sd0s1b and homes from sd0s1e. ============================================================================== KERNEL PROBES: ahc0 rev 0 int a irq 11 on pci0:18 **( this was on pci0:19 )*** ahc0: aic7880 Single Channel, SCSI Id=7, 16 SCBs (ahc0:0:0): "TANDEM 4265-1 1011" type 0 fixed SCSI 2 sd0(ahc0:0:0): Direct-Access 4303MB (8813870 512 byte sectors) (ahc0:4:0): "EXABYTE EXB-85058SQANXR1 07J0" type 1 removable SCSI 2 st0(ahc0:4:0): Sequential-Access density code 0x0, drive empty (ahc0:6:0): "TEAC CD-ROM CD-56S 1.0D" type 5 removable SCSI 2 cd0(ahc0:6:0): CD-ROM cd present [3910 x 2048 byte records] ============================================================================== DISK INFORMATION ******* Working on device /dev/rsd0 ******* parameters extracted from in-core disklabel are: cylinders=4303 heads=64 sectors/track=32 (2048 blks/cyl) Figures below won't work with BIOS for partitions not in cyl 1 parameters to be used for BIOS calculations are: cylinders=4303 heads=64 sectors/track=32 (2048 blks/cyl) Warning: BIOS sector numbering starts with sector 1 Information from DOS bootblock is: The data for partition 0 is: sysid 165,(FreeBSD/NetBSD/386BSD) start 32, size 8812512 (4302 Meg), flag 0 beg: cyl 0/ sector 1/ head 1; end: cyl 1023/ sector 32/ head 63 The data for partition 1 is: The data for partition 2 is: The data for partition 3 is: # /dev/rsd0c: type: SCSI disk: sd0s1 label: flags: bytes/sector: 512 sectors/track: 32 tracks/cylinder: 64 sectors/cylinder: 2048 cylinders: 4302 sectors/unit: 8812512 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: # size offset fstype [fsize bsize bps/cpg] a: 1024000 0 4.2BSD 0 0 0 # (Cyl. 0 - 499) b: 262144 1024000 swap # (Cyl. 500 - 627) c: 8812512 0 unused 0 0 # (Cyl. 0 - 4302*) e: 4096000 1286144 4.2BSD 0 0 0 # (Cyl. 628 - 2627) f: 1024000 5382144 4.2BSD 0 0 0 # (Cyl. 2628 - 3127) g: 1789952 6406144 4.2BSD 0 0 0 # (Cyl. 3128 - 4001) h: 616416 8196096 4.2BSD 0 0 0 # (Cyl. 4002 - 4302*) ============================================================================== Enjoy! -- David (obrien@NUXI.com -or- obrien@FreeBSD.org) From owner-freebsd-scsi Fri Jan 24 05:11:27 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA27675 for freebsd-scsi-outgoing; Fri, 24 Jan 1997 05:11:27 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA27658; Fri, 24 Jan 1997 05:11:21 -0800 (PST) Received: from chain-work.iafrica.com (khetan@chain-work.iafrica.com [196.31.1.66]) by who.cdrom.com (8.7.5/8.6.11) with ESMTP id FAA18583 ; Fri, 24 Jan 1997 05:11:15 -0800 (PST) Received: from localhost (khetan@localhost) by chain-work.iafrica.com (8.8.4/8.6.12) with SMTP id PAA00266; Fri, 24 Jan 1997 15:09:53 +0200 (SAT) X-Authentication-Warning: chain-work.iafrica.com: khetan owned process doing -bs Date: Fri, 24 Jan 1997 15:09:52 +0200 (SAT) From: Khetan Gajjar To: hardware@freebsd.org cc: scsi@freebsd.org Subject: DPT PM2024 Message-ID: X-Alternate-Address: gjjkhe01@sonnenberg.uct.ac.za X-PGP-Fingerprint: FF F9 1C B8 39 06 1E CD 60 4C E8 57 2D A3 46 E7 MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi there. I just wanted to know what the progress of the DPT SCSI card support in FreeBSD was. I saw quite a lot of mail in the archives around May 1996, but nothing since then. I've asked Bob Willcox, but he hasn't been able to work on it. If anyone's got anything, I'd really appreciate it. TIA, --- Khetan Gajjar [ http://www.iafrica.com/~khetan] I'm a FreeBSD User! [ http://www.freebsd.org ] PGP Key [finger khetan@chain.iafrica.com] UUNet Internet Africa [0800-030-002 & help@iafrica.com] From owner-freebsd-scsi Fri Jan 24 06:36:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA00998 for freebsd-scsi-outgoing; Fri, 24 Jan 1997 06:36:14 -0800 (PST) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA00993 for ; Fri, 24 Jan 1997 06:36:11 -0800 (PST) Received: from narnia (localhost [127.0.0.1]) by narnia.plutotech.com (8.8.4/8.7.3) with ESMTP id GAA01007; Fri, 24 Jan 1997 06:36:08 -0800 (PST) Message-Id: <199701241436.GAA01007@narnia.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: deobrien@ucdavis.edu cc: freebsd-scsi@freebsd.org Subject: Re: unstable 2.2-BETA In-reply-to: Your message of "Fri, 24 Jan 1997 01:50:38 PST." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 24 Jan 1997 06:36:08 -0800 From: "Justin T. Gibbs" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Hi all, > >I'm sorry to bother you, but I've got a very unstable 2.2-BETA box right now. >I'm believe my problems to be SCSI subsystem related, but I could be >wrong. > >Is the below something to to worry about? (ie, is my SCSI disk going >bad, or is this a kernel problem? Your problem is termination related and the bug in the driver that causes this should be fixed later today. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-scsi Fri Jan 24 15:37:33 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA26510 for freebsd-scsi-outgoing; Fri, 24 Jan 1997 15:37:33 -0800 (PST) Received: from cabri.obs-besancon.fr (cabri.obs-besancon.fr [193.52.184.3]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA26421; Fri, 24 Jan 1997 15:35:45 -0800 (PST) Received: by cabri.obs-besancon.fr (5.57/Ultrix3.0-C) id AA08766; Sat, 25 Jan 97 00:38:13 +0100 Date: Sat, 25 Jan 97 00:38:13 +0100 Message-Id: <9701242338.AA08766@cabri.obs-besancon.fr> From: Jean-Marc Zucconi To: se@freebsd.org Cc: freebsd-scsi@freebsd.org In-Reply-To: Subject: Re: 2 NCR controllers, problem. X-Mailer: Emacs Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >>>>> Stefan Esser writes: > No error or probe messages at all from the second controller ? > This is RATHER strange! In fact this is not a problem because that this is the second controller. I tried to boot after having removed the 1st NCR and the bios does not recognize the DAT (the machine does not boot however because the drive has no OS on it). Of course this does not explain why a 1542 see the DAT (they both use the same scsi attach routine?) nor why I can make the DAT to appear when I use scsi(8) Jean-Marc _____________________________________________________________________________ Jean-Marc Zucconi Observatoire de Besancon F 25010 Besancon cedex PGP Key: finger jmz@cabri.obs-besancon.fr ============================================================================= From owner-freebsd-scsi Fri Jan 24 22:13:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA20460 for freebsd-scsi-outgoing; Fri, 24 Jan 1997 22:13:28 -0800 (PST) Received: from rover.village.org (rover.village.org [204.144.255.49]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id WAA20455 for ; Fri, 24 Jan 1997 22:13:22 -0800 (PST) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 0.56 #1) id E0vo1Ma-00025B-00; Fri, 24 Jan 1997 23:13:16 -0700 To: scsi@freebsd.org Subject: Question about the scsi sub system. Date: Fri, 24 Jan 1997 23:13:16 -0700 From: Warner Losh Message-Id: Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk OK. I'm trying to track down why my Jaz drive spins up when it is supposed to on file system access, but not when the raw device is accessed.... In tracing down code paths, I would assume that when the raw device is access, it is done via the character device. I was half expecting there to be a cdevsw struct in sd.c that would tell me what I needed to know, however, it appears that one is just blank, filled in elsewhere. Where is this elsewhere? And what is it filled in with? Also, does anybody have any theories about why this would have stopped working sometime during the late summer and fall while my tape drive was on walkabout? What I'm seeing is the following on the console and /var/log/messages: Jan 23 05:48:43 rover /kernel: sd1(uha0:3:0): NOT READY asc:4,0 Jan 23 05:48:43 rover /kernel: sd1(uha0:3:0): Logical unit not ready, cause not reportable whenever I touch the raw device (say with dump) when it has spun down. I don't get these messages when I go through the file system. Once I get this error, it is the kiss of death for that device and all further attempts return I/O error, even through the file system. Pointers to things are fine (and appreciated), since I'm also doing this to learn parts of the kernel better than I've had the chance to to date. Thank you for any help that you can render in my efforts. Warner From owner-freebsd-scsi Sat Jan 25 01:46:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA26793 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 01:46:32 -0800 (PST) Received: from uhura.cc.rochester.edu (pinc_cif@uhura.cc.rochester.edu [128.151.224.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA26788 for ; Sat, 25 Jan 1997 01:46:29 -0800 (PST) Received: (from pinc_cif@localhost) by uhura.cc.rochester.edu (8.8.5/8.8.5) id EAA18319; Sat, 25 Jan 1997 04:46:25 -0500 (EST) From: Joshua Pincus Message-Id: <199701250946.EAA18319@uhura.cc.rochester.edu> Subject: DISASTER! To: freebsd-scsi@freebsd.org (scsi) Date: Sat, 25 Jan 1997 04:46:25 -0500 (EST) Cc: pinc_cif@uhura.cc.rochester.edu (Joshua Pincus) X-Mailer: ELM [version 2.4 PL25] 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 1-25-97 I operate a P6-200 running FreeBSD 2.1.6-RELEASE. It crashed this evening. Whenever I try to run fsck on one particular partition, I get the following message: ** /dev/rsd0s1e BAD SUPER BLOCK: NCYL LESS THAN NCG*CPG pid 161 (fsck), Uid 0: exited on signal 8 floating exception - core dumped First of all, why does fsck dump core when performing this operation? Second of all, is there any way I can retrieve the information off of this partition? Our backup system is on the blink, and it is imperative that the information in this partition be extracted. Can I relabel the disk in order to validate the superblock information, or is that a newfs operation only? The problem is that newfs will erase all of the old info. I am desperate to restore as much information as I can. Is there anyway that I can use the raw disk device as an input file and dump the information onto another partition, or into a file? Any help would be warmly accepted. Josh Pincus From owner-freebsd-scsi Sat Jan 25 02:55:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA28449 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 02:55:47 -0800 (PST) Received: from dfw-ix5.ix.netcom.com (dfw-ix5.ix.netcom.com [206.214.98.5]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id CAA28443; Sat, 25 Jan 1997 02:55:45 -0800 (PST) Received: from baloon.mimi.com (ncr-us9-06.ix.netcom.com [204.31.236.198]) by dfw-ix5.ix.netcom.com (8.6.13/8.6.12) with ESMTP id CAA15382; Sat, 25 Jan 1997 02:55:11 -0800 Received: (from asami@localhost) by baloon.mimi.com (8.8.4/8.6.12) id CAA29748; Sat, 25 Jan 1997 02:55:09 -0800 (PST) Date: Sat, 25 Jan 1997 02:55:09 -0800 (PST) Message-Id: <199701251055.CAA29748@baloon.mimi.com> To: joe@via.net CC: hardware@freebsd.org, scsi@freebsd.org In-reply-to: <199701250816.AAA15585@monk.via.net> (message from Joe McGuckin on Sat, 25 Jan 1997 00:16:27 -0800) Subject: Re: Symbios (NCR) '875 SCSI card question... From: asami@vader.cs.berkeley.edu (Satoshi Asami) Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * From: Joe McGuckin Moved to "scsi".... * With both sets of drives connected, the board cannot probe out any * drives. If I disconnect either the internal WIDE drive or the external * narrow drives, the card will probe out the remaining drives that are * connected. Have you turned off termination on the card? Or maybe the external wide to narrow cable is not handling some wires right, try terminating only the high bits on the card. Satoshi From owner-freebsd-scsi Sat Jan 25 05:30:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA02509 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 05:30:41 -0800 (PST) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA02504 for ; Sat, 25 Jan 1997 05:30:38 -0800 (PST) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id AAA25410; Sun, 26 Jan 1997 00:23:33 +1100 Date: Sun, 26 Jan 1997 00:23:33 +1100 From: Bruce Evans Message-Id: <199701251323.AAA25410@godzilla.zeta.org.au> To: imp@village.org, scsi@freebsd.org Subject: Re: Question about the scsi sub system. Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >whenever I touch the raw device (say with dump) when it has spun >down. I don't get these messages when I go through the file system. >Once I get this error, it is the kiss of death for that device and all >further attempts return I/O error, even through the file system. Mounting a file system keeps the device open. So does swapping. This stops the sd_close() routine from allowing ejection of the disk. Other than that, there is nothing special about raw i/o at the device driver level. Unmounting the file system should break things just as well as closing the raw device :-). Turning off swapping is not supported :-(. Ejection isn't allowed even after shutdown. Bruce From owner-freebsd-scsi Sat Jan 25 06:09:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA03474 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 06:09:56 -0800 (PST) Received: from perki0.connect.com.au (perki0.connect.com.au [192.189.54.85]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA03467 for ; Sat, 25 Jan 1997 06:09:51 -0800 (PST) Received: (from Unemeton@localhost) by perki0.connect.com.au id BAA17336 (8.7.6h/IDA-1.6); Sun, 26 Jan 1997 01:09:42 +1100 (EST) X-Authentication-Warning: perki0.connect.com.au: Unemeton set sender to giles@nemeton.com.au using -f >Received: from localhost.nemeton.com.au (localhost.nemeton.com.au [127.0.0.1]) by nemeton.com.au (8.8.5/8.8.5) with SMTP id BAA06874; Sun, 26 Jan 1997 01:07:52 +1100 (EST) Message-Id: <199701251407.BAA06874@nemeton.com.au> To: Joshua Pincus cc: freebsd-scsi@freebsd.org Subject: Re: DISASTER! In-reply-to: <199701250946.EAA18319@uhura.cc.rochester.edu> Date: Sun, 26 Jan 1997 01:07:52 +1100 From: Giles Lean Content-Type: text Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 25 Jan 1997 04:46:25 -0500 (EST) Joshua Pincus wrote: > ** /dev/rsd0s1e > BAD SUPER BLOCK: NCYL LESS THAN NCG*CPG > pid 161 (fsck), Uid 0: exited on signal 8 > floating exception - core dumped (First note: none of this is SCSI specific.) The following suggestions are pretty much in order of desirability. Try running fsck with an alternate superblock. If you don't know the number of an alternate superblock, try 32. To find superblocks you can read the partition looking for blocks that have FS_MAGIC in the right place -- I've included some code for this below. (Whether this code compiles untweaked on FreeBSD right now I don't know, but it has saved me a couple of partitions that I decided I wanted *after* I trashed the disklabel.) Dig into fsck and fix the (probable) divide by zero. There was a PR or mail item recently about a fsck core dump; might be the same one. > Second of all, is there any way I can retrieve the information off of > this partition? Our backup system is on the blink, and it is imperative > that the information in this partition be extracted. Ouch. Yes, you should still be able to read some of the data. I suspect you'll need to get a sorta-kinda-valid superblock before you can do much. Corrupted filesystems can be mounted (sometimes) and if all your filesystems are mounted read only no more harm will occur even if the machine panics. If you can't get a mountable filesystem but have some sort of superblock Tom Christiansen's "icat" program which retrieves files by inode number might be useful. (Don't know if I've compiled this on FreeBSD, either, but I've included it below as well.) The filesystem might be totally toast, of course. :-( > Can I relabel the disk in order to validate the superblock No. Someone familiar with filesystem hacking should be able to get some of the data off even if the above tools fail, or even write a program (or use a debugger -- I don't know gdb well) to manually reconstruct the superblock. Failing that, if you just need a particular file or two 'strings' on the filesystem (onto a *big* scratch file) might be enough. I've pulled data out of a trashed database in this way. (This was nasty -- I had no data on the internal data structures, so couldn't do anything much smarter.) Good luck! Don't do anything rash without seeking further opinions -- some of the data recovery services could doubtless help, at a price. Giles #text/plain; name=findsb.c src/findsb.c #text/plain; name=icat.c src/icat/icat.c #text/plain; name=icat.man src/icat/icat.man From owner-freebsd-scsi Sat Jan 25 07:06:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA04710 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 07:06:11 -0800 (PST) Received: from perki0.connect.com.au (perki0.connect.com.au [192.189.54.85]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA04704 for ; Sat, 25 Jan 1997 07:06:05 -0800 (PST) Received: (from Unemeton@localhost) by perki0.connect.com.au id CAA21051 (8.7.6h/IDA-1.6); Sun, 26 Jan 1997 02:05:55 +1100 (EST) X-Authentication-Warning: perki0.connect.com.au: Unemeton set sender to giles@nemeton.com.au using -f >Received: from localhost.nemeton.com.au (localhost.nemeton.com.au [127.0.0.1]) by nemeton.com.au (8.8.5/8.8.5) with SMTP id CAA08199; Sun, 26 Jan 1997 02:05:01 +1100 (EST) Message-Id: <199701251505.CAA08199@nemeton.com.au> cc: Joshua Pincus , freebsd-scsi@freebsd.org Subject: Re: DISASTER! In-reply-to: <199701251407.BAA06874@nemeton.com.au> MIME-Version: 1.0 Content-ID: <8192.854204687.0@nemeton.com.au> Date: Sun, 26 Jan 1997 02:04:59 +1100 From: Giles Lean Content-Type: multipart/mixed; boundary="----- =_aaaaaaaaaa0" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk ------- =_aaaaaaaaaa0 Content-Type: text/plain; charset="us-ascii" Content-ID: <8192.854204687.1@nemeton.com.au> On Sun, 26 Jan 1997 01:07:52 +1100 Giles Lean wrote: > #text/plain; name=findsb.c src/findsb.c > #text/plain; name=icat.c src/icat/icat.c > #text/plain; name=icat.man src/icat/icat.man Dammit! How embarassing. Here they are. Giles ------- =_aaaaaaaaaa0 Content-Type: text/plain; name="findsb.c"; charset="us-ascii" Content-ID: <8192.854204687.2@nemeton.com.au> /* * Find superblocks on a BSD FFS filesystem. * * On NetBSD, need only to check for FS_MAGIC. HP-UX has about three * flavours of FS_MAGIC. You'll need something else for a Veritas * file system. */ #include #include #include #include #include /* * Usually larger on HP-UX -- too small will only result in the odd * false match. */ #define BLOCKSIZE (4 * 1024) int main(int argc, char *argv[]) { char block[BLOCKSIZE]; int fd; struct fs *filesys; int n; int blockno; blockno = 0; filesys = (struct fs *) block; if (argc != 2) { fprintf(stderr, "usage: findsb raw_device\n"); exit(1); } if ((fd = open(argv[1], O_RDONLY)) < 0) { perror(argv[1]); exit(1); } while ((n = read(fd, block, BLOCKSIZE)) == BLOCKSIZE) { if (filesys->fs_magic == FS_MAGIC) printf("%d\n", blockno * (BLOCKSIZE / 512)); blockno++; } } ------- =_aaaaaaaaaa0 Content-Type: text/plain; name="icat.c"; charset="us-ascii" Content-ID: <8192.854204687.3@nemeton.com.au> /* * icat -- "cat" some inodes on a given device to standard out. * WARNING: this program is the biggest security hole in the * world, it allows you to completely bypass the file System. * Keep this code under your hat. */ #include #include #include #include #include #include #include #include #include #include /* 4.4BSD changes */ #ifndef itod #define itod ino_to_fsba #define itoo ino_to_fsbo #endif #define ISIZ (sizeof(struct dinode)) #define MAXINOPB (MAXBSIZE / ISIZ) #define MAXNINDIR (MAXBSIZE / sizeof(daddr_t)) union { char dummy[SBSIZE]; struct fs sblk; } sb_un; #define sblock sb_un.sblk struct dinode *iget(); void bread(); void indir(); char *progname; int status; long dev_bsize = 1; struct dinode *ip; char bbuf[MAXBSIZE]; main(argc, argv) int argc; char *argv[]; { register int i; int fd; progname = argv[0]; if (argc < 3) { fprintf(stderr, "Usage: %s filsys inumber ...\n", progname); exit(1); } fd = open(argv[1], O_RDONLY); if (fd < 0) { fprintf(stderr, "%s: cannot open ", progname); perror(argv[1]); exit(1); } #ifndef tahoe bread(fd, SBSIZE, (char *)&sblock, SBSIZE); #else bread(fd, SBOFF, (char *)&sblock, SBSIZE); #endif if (sblock.fs_magic != FS_MAGIC) { fprintf(stderr, "%s: bad super-block magic number 0x%x\n", progname, sblock.fs_magic); exit(1); } dev_bsize = sblock.fs_fsize / fsbtodb(&sblock, 1); for (i = 2; i < argc; i++) { if (icatit(fd, argv[i]) == 0) status++; } exit(status); } icatit(fd, inumber) int fd; char *inumber; { register int i; unsigned int n; int rsize; long size; if (!isnumber(inumber)) { fprintf(stderr, "%s: %s is not a number\n", progname, inumber); return (0); } n = atoi(inumber); if (n == 0) { fprintf(stderr, "%s: 0 is an invalid inode number\n", progname); return (0); } ip = iget(fd, n); if (ip == (struct dinode *)0) { fprintf(stderr, "%s: cannot read inode number %u\n", progname, n); return (0); } if ((ip->di_mode & IFMT) != IFREG) { fprintf(stderr, "%s: inode %u has mode 0%o; not a regular file\n", progname, n, ip->di_mode); return (0); } /* direct blocks */ for (i = 0, size = ip->di_size; i < NDADDR && size > 0; i++) { rsize = MIN(size, sblock.fs_bsize); if (ip->di_db[i]) { bread(fd, fsbtodb(&sblock, ip->di_db[i]), bbuf, (int) sblock.fs_bsize); } else { bzero((char *) bbuf, rsize); } if (write(1, bbuf, rsize) != rsize) { fprintf(stderr, "%s: failed write to ", progname); perror("stdout"); return (0); } size -= rsize; } /* indirect blocks */ for (i = 0; i < NIADDR && size > 0; i++) { if (ip->di_ib[i]) indir(fd, ip->di_ib[i], i, &size); } return (1); } isnumber(s) char *s; { register c; while(c = *s++) if (c < '0' || c > '9') return(0); return(1); } struct dinode * iget(fd, inum) int fd; unsigned int inum; { struct ino { char junk[ISIZ]; }; static struct ino buf[MAXINOPB]; struct dinode *ip; bread(fd, fsbtodb(&sblock, itod(&sblock, inum)), (char *)buf, (int) sblock.fs_bsize); ip = (struct dinode *) &buf[itoo(&sblock, inum)]; return (ip); } void bread(fd, bno, buf, cnt) int fd; daddr_t bno; char *buf; int cnt; { off_t lseek(); register int n; off_t off; off = bno * dev_bsize; if (lseek(fd, off, L_SET) != off) { fprintf(stderr, "%s: cannot read block 0x%x\n", progname, bno); exit(1); } if ((n = read(fd, buf, cnt)) < 0) { fprintf(stderr, "%s: premature EOF; ", progname); fprintf(stderr, "bno = %ld expected = %d count = %d\n", bno, cnt, n); exit(1); } } void indir(fd, blk, lvl, size) int fd; daddr_t blk; int lvl; long *size; { int i, rsize; daddr_t idblk[MAXNINDIR]; if (blk != 0) bread(fd, fsbtodb(&sblock, blk), (char *)idblk, (int)sblock.fs_bsize); else bzero((char *)idblk, (int)sblock.fs_bsize); if (lvl <= 0) { for (i = 0; i < NINDIR(&sblock) && *size > 0; i++) { rsize = MIN(*size, sblock.fs_bsize); if (idblk[i]) { bread(fd, fsbtodb(&sblock, idblk[i]), bbuf, (int) sblock.fs_bsize); } else { bzero((char *) bbuf, rsize); } if (write(1, bbuf, rsize) != rsize) { fprintf(stderr, "%s: failed write to ", progname); perror("stdout"); return; } *size -= rsize; } return; } lvl--; for (i = 0; i < NINDIR(&sblock); i++) { indir(fd, idblk[i], lvl, size); if (*size <= 0) return; } } ------- =_aaaaaaaaaa0 Content-Type: text/plain; name="icat.man"; charset="us-ascii" Content-ID: <8192.854204687.4@nemeton.com.au> ICAT(8) ICAT(8) NAME icat - cat contents of inode SYNOPSIS icat filesys inumber ... DESCRIPTION Icat works like cat except that rather than working with file names, it works with (device, inumber) pairs. This is particularly useful in looking at files that have been unlinked from the file system but are still open. fstat(8) and lsof(8L) can be used to identify such files. SEE ALSO cat(1), fstat(8), lsof(8L) WARNING Icat should only be runnable by the superuser. Under no circumstances should this program be installed SUID root or SGID the group that owns the raw disks, because if so, file system security can be easily circumvented. ------- =_aaaaaaaaaa0-- From owner-freebsd-scsi Sat Jan 25 07:22:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA05003 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 07:22:03 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA04997 for ; Sat, 25 Jan 1997 07:21:59 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id QAA28038; Sat, 25 Jan 1997 16:21:37 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.4/8.6.9) id PAA16651; Sat, 25 Jan 1997 15:58:28 +0100 (MET) Message-ID: Date: Sat, 25 Jan 1997 15:58:26 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: chris@bb.cc.wa.us (Chris Coleman) Cc: freebsd-scsi@FreeBSD.org (FreeBSD SCSI list) Subject: Re: Tape Backup Drive Not working. References: X-Mailer: Mutt 0.55-PL10 Mime-Version: 1.0 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 Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Chris Coleman on Jan 23, 1997 18:36:01 -0800 Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk As Chris Coleman wrote: > > (Btw., your mail was mis-referenced. I've also moved this to the > > SCSI list.) > Thanks, i didnt even think of that, I only subscribe to -hackers. No deal, we can Cc you. > > Which one? (Probe string) > > "Conner CTT8000-S 1.17" Type 1 removable SCSI2 > st0(ahc:0:3:0):Sequential-Access Density Code 0x45, Drive Empty Never heard of such a beast. The density code is far beyond those mentioned in the SCSI-2 specs. I assume this is something only compatible to itself? > > > /kernel st0 (ahc:0:3:0): ILLEGAL REQUEST asc: 24,0 Invalid field in CDB > > > /kernel st0: bad request, must be between 0 and 0 > > > /kernel st0 (ahc:0:3:0): ILLEGAL REQUEST asc: 24,0 Invalid field in CDB > > > > Which blocksize setting? (output of `mt status') > > -- > mt status: > > Present Mode: Density = 0x00 Blocksize Variable Do you have any kind of doc for it? Is variable blocking even supported for that model? If not, try to wire down the drive to 512 or 1024 bytes per block, e.g. mt blocksize 512 (This will only last for one mount-session.) I'm not sure how recent your system is. If it's not from the 2.2 line, you might give the 2.2 boot+fixit combo a try as well. I remember that the ``must be between 0 and 0'' bogosity happend often previously, and seemingly has been fixed since. If all this is of no help, you need to turn on SCSIDEBUG (kernel option), and dive into all of this. Unfortunately, SCSI debug messages are fairly verbose, and i haven't found a good model yet to trim them down to something useful, you best turn everything on and just filter ``by brain'' to spot the really important information. You need a bit of understanding about the SCSI command structure. At least, we need to learn which command the drive is complaining about, and what exactly is the invalid field in the CDB (command descriptor block) -- the latter information is called `sks' (sense key specific information). -- 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 Sat Jan 25 07:40:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA05568 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 07:40:53 -0800 (PST) Received: from deacon.cogsci.ed.ac.uk (deacon.cogsci.ed.ac.uk [129.215.144.7]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA05563 for ; Sat, 25 Jan 1997 07:40:50 -0800 (PST) Received: (from richard@localhost) by deacon.cogsci.ed.ac.uk (8.6.10/8.6.12) id PAA16318; Sat, 25 Jan 1997 15:40:40 GMT Date: Sat, 25 Jan 1997 15:40:40 GMT Message-Id: <199701251540.PAA16318@deacon.cogsci.ed.ac.uk> From: Richard Tobin Subject: Re: Tape Backup Drive Not working. To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch), chris@bb.cc.wa.us (Chris Coleman) In-Reply-To: J Wunsch's message of Sat, 25 Jan 1997 15:58:26 +0100 Organization: just say no Cc: freebsd-scsi@FreeBSD.org (FreeBSD SCSI list) Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > > "Conner CTT8000-S 1.17" Type 1 removable SCSI2 > > st0(ahc:0:3:0):Sequential-Access Density Code 0x45, Drive Empty > Never heard of such a beast. The density code is far beyond those > mentioned in the SCSI-2 specs. I assume this is something only > compatible to itself? Looks like Conner's version of the HP T4000s, which is actually a QIC standard (3095). I posted fairly trivial patches for the HP a few months ago. Essentially all that was required was setting the PF bit in mode select, but there's no reason to suppose that Conner's SCSI implementation will have the same quirks as HP's. Incidentally, I would interested if anyone else is successfully using the T4000s, since I having trouble reading tapes which is most likely a hardware problem, but I'm not certain. -- Richard From owner-freebsd-scsi Sat Jan 25 08:20:40 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA06865 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 08:20:40 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id IAA06858 for ; Sat, 25 Jan 1997 08:20:34 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id RAA29151; Sat, 25 Jan 1997 17:20:32 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.4/8.6.9) id QAA25557; Sat, 25 Jan 1997 16:59:51 +0100 (MET) Message-ID: Date: Sat, 25 Jan 1997 16:59:49 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: freebsd-scsi@FreeBSD.org (FreeBSD SCSI list) Cc: chris@bb.cc.wa.us (Chris Coleman) Subject: Re: Tape Backup Drive Not working. References: <199701251540.PAA16318@deacon.cogsci.ed.ac.uk> X-Mailer: Mutt 0.55-PL10 Mime-Version: 1.0 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 Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199701251540.PAA16318@deacon.cogsci.ed.ac.uk>; from Richard Tobin on Jan 25, 1997 15:40:40 +0000 Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk As Richard Tobin wrote: > Essentially all that was required was setting the PF bit in mode > select, but there's no reason to suppose that Conner's SCSI > implementation will have the same quirks as HP's. Hmm. I wonder if there's any reason to _not_ set the PF bit by default? Julian? Justin? Are there any tape drives known to choke on this? For the curious, here's the patch (offhand, untested). I'd vote that, if we need it at all, we should unset the bit based on some drive's quirks, but turn it on by default (in the assumption that most drives adhere to SCSI-2). I'd be particularly interested in getting feedback from people owning an old drive that doesn't claim to be SCSI-2 (like an old Archive Viper 150 or Wangtek 5150ES). Index: scsi/st.c =================================================================== RCS file: /home/ncvs/src/sys/scsi/st.c,v retrieving revision 1.74 diff -u -u -r1.74 st.c --- st.c 1997/01/14 06:54:22 1.74 +++ st.c 1997/01/25 15:56:34 @@ -1405,6 +1405,7 @@ bzero(&dat, dat_len); bzero(&scsi_cmd, sizeof(scsi_cmd)); scsi_cmd.op_code = MODE_SELECT; + scsi_cmd.byte2 |= SMS_PF; scsi_cmd.length = dat_len; dat.header.blk_desc_len = sizeof(struct blk_desc); dat.header.dev_spec |= SMH_DSP_BUFF_MODE_ON; -- 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 Sat Jan 25 09:24:16 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA09047 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 09:24:16 -0800 (PST) Received: from rover.village.org (rover.village.org [204.144.255.49]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA09036 for ; Sat, 25 Jan 1997 09:24:12 -0800 (PST) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 0.56 #1) id E0voBpb-0002zM-00; Sat, 25 Jan 1997 10:23:55 -0700 To: Bruce Evans Subject: Re: Question about the scsi sub system. Cc: scsi@freebsd.org In-reply-to: Your message of "Sun, 26 Jan 1997 00:23:33 +1100." <199701251323.AAA25410@godzilla.zeta.org.au> References: <199701251323.AAA25410@godzilla.zeta.org.au> Date: Sat, 25 Jan 1997 10:23:54 -0700 From: Warner Losh Message-Id: Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <199701251323.AAA25410@godzilla.zeta.org.au> Bruce Evans writes: : Mounting a file system keeps the device open. So does swapping. : This stops the sd_close() routine from allowing ejection of the disk. : Other than that, there is nothing special about raw i/o at the device : driver level. Unmounting the file system should break things just as well : as closing the raw device :-). Turning off swapping is not supported :-(. : Ejection isn't allowed even after shutdown. Hmmm, so is there code in the file system to handle scsi disks that need to be spun up? Also, in trying to find the code path through the kernel in question, I was unable to find the cdevsw's table for sd. I'd still like to know where that is being initialized or how bdevsw is being used instead. To recap, if the Jaz drive is spinning when I touch the raw device, then it works and I get a dump. If the drive has gone into power down mode, then I don't. I'm not trying to eject the disk or anything like that, and there is no swapping being done to this device. The file system is mounted the whole time. That is # mount | egrep /jaz /dev/sd1c on /jaz (local) # ls /jaz I hear the drive spin up FreeBSD OpenBSD mips # dump 1f /dev/null /jaz # Wait 1/2 hour for drive to spin down. # dump 1f /dev/null /jaz ... sd1(uha0:3:0): NOT READY asc:4,0 sd1(uha0:3:0): Logical unit not ready, cause not reportable DUMP: Cannot open /dev/rsd1c # ls /jaz ls: I/O error # At this point I can't even unmount the disk and remount it. I have to reboot. I've not interested in ejecting the disk, but it won't eject (so sd_close isn't being called). So why would a file system access spin the disk up, while a raw device access wouldn't? I'm confused. Warner From owner-freebsd-scsi Sat Jan 25 09:52:06 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA09895 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 09:52:06 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA09883 for ; Sat, 25 Jan 1997 09:52:02 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id SAA01452; Sat, 25 Jan 1997 18:51:50 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.4/8.6.9) id SAA24863; Sat, 25 Jan 1997 18:34:06 +0100 (MET) Message-ID: Date: Sat, 25 Jan 1997 18:34:05 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: imp@village.org (Warner Losh) Cc: scsi@freebsd.org Subject: Re: Question about the scsi sub system. References: X-Mailer: Mutt 0.55-PL10 Mime-Version: 1.0 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 Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Warner Losh on Jan 24, 1997 23:13:16 -0700 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Warner Losh wrote: > Also, does anybody have any theories about why this would have stopped > working sometime during the late summer and fall while my tape drive > was on walkabout? > > What I'm seeing is the following on the console and /var/log/messages: > Jan 23 05:48:43 rover /kernel: sd1(uha0:3:0): NOT READY asc:4,0 > Jan 23 05:48:43 rover /kernel: sd1(uha0:3:0): Logical unit not ready, cause not reportable Move the START UNIT command earlier in the sdopen() function. I've already did this for the `cd' and `od' driver, and at least in the latter case, it works fine. (I'm running the `od' driver with an automatic spindown at odclose().) -- 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 Sat Jan 25 10:07:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA10477 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 10:07:55 -0800 (PST) Received: from rover.village.org (rover.village.org [204.144.255.49]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA10470 for ; Sat, 25 Jan 1997 10:07:48 -0800 (PST) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 0.56 #1) id E0voCVq-00031I-00; Sat, 25 Jan 1997 11:07:34 -0700 To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) Subject: Re: Question about the scsi sub system. Cc: scsi@freebsd.org In-reply-to: Your message of "Sat, 25 Jan 1997 18:34:05 +0100." References: Date: Sat, 25 Jan 1997 11:07:34 -0700 From: Warner Losh Message-Id: Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message J Wunsch writes: : Move the START UNIT command earlier in the sdopen() function. I've : already did this for the `cd' and `od' driver, and at least in the : latter case, it works fine. (I'm running the `od' driver with an : automatic spindown at odclose().) I noticed that in the sd and cd drivers, the start unit command are placed in the same location, relative to eachother. Is this in your private copy? In a branch that I'm missing? Anyway, I'm going to try moving scsi_start_unit() to before the scsi_test_unit_ready() call and see if that helps or not. Thanks for the suggestion! Warner From owner-freebsd-scsi Sat Jan 25 10:15:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA11067 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 10:15:39 -0800 (PST) Received: from DNS.Lamb.net (root@DNS.Lamb.net [207.90.181.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA11062 for ; Sat, 25 Jan 1997 10:15:37 -0800 (PST) Received: from PacBell.TelcoSucks.org (ulf@PacBell.TelcoSucks.org [207.90.181.5]) by DNS.Lamb.net (8.8.5/20.74.3.14) with SMTP id KAA19424 for ; Sat, 25 Jan 1997 10:15:43 -0800 (PST) Message-Id: <3.0.32.19970125101729.00a6c834@Gatekeeper-3.Lamb.net> X-Sender: ulf@Gatekeeper-3.Lamb.net X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 25 Jan 1997 10:17:30 -0800 To: scsi@freebsd.org From: Ulf Zimmermann Subject: Archive DDS2 Autoload, anyone tried ? Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello. I got an Archive (Connor/Seagate) 4586NP (SCSI probe: ARCHIVE Python 288429-XXX 4.CM). After adding ch to the kernel config, adding the probe message with wildcards to /sys/scsi/scsiconf.c the changer was still not found. So I changed the maybe_more in scsiconf.c to 1 and it finds it. Now the problem, if I do a test_unit_ready (like the changer program does, which I got from Bill) the changer says "NOT READY, Medium not present". Because of that, I can't load the first tape, I always need to load 1 tape manual, then the rest works. Anyone who can give me some more tips ? I will call Seagate on monday to get the operation handbook from them. Regards, Ulf. ----------------------------------------------------------- Alameda Networks, Inc. | Ulf Zimmermann (ulf@Alameda.net) 1525 Pacific Avenue | Phone: (510)769-2936 Alameda, CA 94501 | Fax : (510)521-5073 From owner-freebsd-scsi Sat Jan 25 10:36:08 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA11949 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 10:36:08 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA11931 for ; Sat, 25 Jan 1997 10:36:00 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id TAA02953; Sat, 25 Jan 1997 19:35:52 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.4/8.6.9) id TAA06229; Sat, 25 Jan 1997 19:23:56 +0100 (MET) Message-ID: Date: Sat, 25 Jan 1997 19:23:55 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: imp@village.org (Warner Losh) Cc: scsi@freebsd.org Subject: Re: Question about the scsi sub system. References: X-Mailer: Mutt 0.55-PL10 Mime-Version: 1.0 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 Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Warner Losh on Jan 25, 1997 11:07:34 -0700 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Warner Losh wrote: > I noticed that in the sd and cd drivers, the start unit command are > placed in the same location, relative to eachother. Is this in your > private copy? In a branch that I'm missing? I thought i committed a patch to cd.c recently? (rev. 1.75) -- 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 Sat Jan 25 13:52:01 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA19747 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 13:52:01 -0800 (PST) Received: from rover.village.org (rover.village.org [204.144.255.49]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA19726 for ; Sat, 25 Jan 1997 13:51:55 -0800 (PST) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 0.56 #1) id E0voG0o-0000Bz-00; Sat, 25 Jan 1997 14:51:46 -0700 To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) Subject: Re: Question about the scsi sub system. Cc: scsi@freebsd.org In-reply-to: Your message of "Sat, 25 Jan 1997 19:23:55 +0100." References: Date: Sat, 25 Jan 1997 14:51:46 -0700 From: Warner Losh Message-Id: Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message J Wunsch writes: : I thought i committed a patch to cd.c recently? (rev. 1.75) Opopps. You are correct. The cd driver does a scsi_test_init_ready() followed by a scsi_start_unit() followed by another scsi_test_unit_ready(). I found that so far with my meager testing that moving the start unit to before the scsi_test_unit_ready() the sd.c driver that my jaz drive no longer buys the far when the raw device is accessed when spun down. I want to test this more, but so far it looks good. I'll go ahead and add the similar logic in cd.c as well to make sure things are cool. Thanks for your suggestions. I think that I may have a fix for this soon. Who would be up (or best) for reviewing the change? I'll commit unreviewed changes to userland, but not to the kernel just yet :-). Warner From owner-freebsd-scsi Sat Jan 25 17:29:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA07298 for freebsd-scsi-outgoing; Sat, 25 Jan 1997 17:29:14 -0800 (PST) Received: from deacon.cogsci.ed.ac.uk (deacon.cogsci.ed.ac.uk [129.215.144.7]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id RAA07291 for ; Sat, 25 Jan 1997 17:29:10 -0800 (PST) Received: from pitcairn.cogsci.ed.ac.uk (pitcairn.cogsci.ed.ac.uk [129.215.197.19]) by deacon.cogsci.ed.ac.uk (8.6.10/8.6.12) with ESMTP id BAA08138; Sun, 26 Jan 1997 01:29:05 GMT Date: Sun, 26 Jan 1997 01:29:03 GMT Message-Id: <10798.199701260129@pitcairn.cogsci.ed.ac.uk> From: Richard Tobin Subject: Re: Tape Backup Drive Not working. To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch), freebsd-scsi@FreeBSD.org (FreeBSD SCSI list) In-Reply-To: J Wunsch's message of Sat, 25 Jan 1997 16:59:49 +0100 Organization: just say no Cc: chris@bb.cc.wa.us (Chris Coleman) Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > Hmm. I wonder if there's any reason to _not_ set the PF bit by > default? Well, as you say, all SCSI-2 devices should be happy with it. On the other hand, I regard it as a bug that the HP T4000s complains if it's not set, since there are no parameters being passed that it affects. Why should it care about the format of zero bytes of data? I conjecture that Linux sets it by default, since I have heard that the HP drive worked with Linux without change. > I'd vote that, > if we need it at all, we should unset the bit based on some drive's > quirks, but turn it on by default (in the assumption that most drives > adhere to SCSI-2). I agree. -- Richard