From owner-freebsd-scsi@FreeBSD.ORG Tue Jun 15 19:45:54 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id F2F3016A4CE for ; Tue, 15 Jun 2004 19:45:53 +0000 (GMT) Received: from elvis.bardenent.com (elvis.bardenent.com [216.90.241.7]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8098B43D49 for ; Tue, 15 Jun 2004 19:45:53 +0000 (GMT) (envelope-from meb@bardenent.com) Received: from meb (meb.bardenent.com [216.90.241.123]) by elvis.bardenent.com (8.12.9p2/8.12.9) with ESMTP id i5FJjBZQ081445 for ; Tue, 15 Jun 2004 15:45:11 -0400 (EDT) (envelope-from meb@bardenent.com) Message-Id: <200406151945.i5FJjBZQ081445@elvis.bardenent.com> From: "Matthew Bertrand" To: Date: Tue, 15 Jun 2004 15:45:11 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook, Build 11.0.5510 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2096 Thread-Index: AcRTEUVkA72GgSncRES0Iw55HuRUIw== Subject: SCSI Tape problem after full write X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 15 Jun 2004 19:45:54 -0000 I am trying to use a Dell PV-122T LTO autoloader on FreeBSD 4.10-RELEASE, but am having some difficulties with it after writing a full tape. The tape drive is recognized properly and device entries are created. Here's the relevant information from my dmesg: sa0 at asr0 bus 0 target 6 lun 0 sa0: Removable Sequential Access SCSI-3 device ch0 at asr0 bus 0 target 5 lun 0 ch0: Removable Changer SCSI-2 device ch0: 8 slots, 1 drive, 1 picker, 0 portals (this is prior to having ch loaded into the kernel -- just trying to get the tape drive to work for right now) I'm attempting to use syutils/bacula to perform network backups, so following the instructions in the port, I used tapetest to write a full tape to verify I wouldn't lose any data after filling a tape. It took 2 hours and 11 minutes to fill up the 100GB tape, and it provided me this output: crazyeddie# c++ -g -O2 -Wall -pthread -c tapetest.c crazyeddie# c++ -g -O2 -Wall -pthread tapetest.o -o tapetest crazyeddie# ./tapetest /dev/nsa0 *rewind Rewound /dev/nsa0 *rawfill Begin writing blocks of 64512 bytes. +++++++ [...] weof_dev ioctl MTWEOF error on /dev/nsa0. ERR=Device not configured. Bad status from weof -1. ERR=Device not configured Write failed. Last block written=1656999. stat=-1 ERR=Input/output error *rewind Rewind error on /dev/nsa0. ERR=Device not configured. Bad status from rewind. ERR=Device not configured * After exiting out of that, it I would get the same Device not configured error trying anything else (mt, tar). Ran the following commands to try and get the tape drive back online: crazyeddie# camcontrol devlist at scbus0 target 1 lun 0 (pass0,da0) at scbus0 target 5 lun 0 (pass1) at scbus0 target 6 lun 0 (pass2,sa0) crazyeddie# camcontrol tur sa0 -v Unit is not ready CAM status is 0xa crazyeddie# camcontrol reset 0:6:0 Reset of 0:6:0 was successful crazyeddie# camcontrol devlist at scbus0 target 1 lun 0 (pass0,da0) at scbus0 target 5 lun 0 (pass1) crazyeddie# After that, the only way I could get the tape drive recognized again was to power cycle the autoloader and then restart FreeBSD. I've enabled CAMDEBUG in the kernel ( which is generic other than CAMDEBUG & ch), but the output from that doesn't seem to provide me with anything useful (I don't know what to look for). Does anyone have any suggestions for this problem or any tips on troubleshooting it further? Thanks, -Matthew Bertrand