From owner-freebsd-scsi Wed Jan 17 10: 7: 8 2001 Delivered-To: freebsd-scsi@freebsd.org Received: from boco.fee.vutbr.cz (boco.fee.vutbr.cz [147.229.9.11]) by hub.freebsd.org (Postfix) with ESMTP id 4F92137B401 for ; Wed, 17 Jan 2001 10:06:49 -0800 (PST) Received: from kazi.dcse.fee.vutbr.cz (kazi.dcse.fee.vutbr.cz [147.229.8.12]) by boco.fee.vutbr.cz (8.11.2/8.11.2) with ESMTP id f0HI6ao61760 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK) for ; Wed, 17 Jan 2001 19:06:37 +0100 (CET) Received: (from cejkar@localhost) by kazi.dcse.fee.vutbr.cz (8.11.0/8.11.0) id f0HI6au29163 for freebsd-scsi@freebsd.org; Wed, 17 Jan 2001 19:06:36 +0100 (CET) Date: Wed, 17 Jan 2001 19:06:36 +0100 From: Cejka Rudolf To: freebsd-scsi@freebsd.org Subject: Troubles with Mammoth2 if there is a tape error Message-ID: <20010117190636.A28937@dcse.fee.vutbr.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I have Exabyte EZ17 library (message in dmesg is bogus) and Mammoth II tape drive. If there is an tape error, I can not kill writing process, because it is in "D" ps-state and tons of error messages are generated in /var/log/messages. Details follow: -- FreeBSD 4.2-STABLE #0: Thu Dec 7 16:08:27 CET 2000 CPU: Pentium III/Pentium III Xeon/Celeron (737.02-MHz 686-class CPU) real memory = 134131712 (130988K bytes) ahc0: port 0xb000-0xb0ff mem 0xf1800000-0xf1800fff irq 5 at device 13.0 on pci2 aic7890/91: Wide Channel A, SCSI Id=7, 32/255 SCBs sa0 at ahc0 bus 0 target 1 lun 0 sa0: Removable Sequential Access SCSI-2 device sa0: 80.000MB/s transfers (40.000MHz, offset 31, 16bit) ch0 at ahc0 bus 0 target 0 lun 0 ch0: Removable Changer SCSI-2 device ch0: 3.300MB/s transfers ch0: 7 slots, 1 drive, 1 picker, 0 portals -- At present, I test tapes with following script (Exabyte has still problems with Mammoth2 reliability): -- mt -f /dev/nrsa0 comp off mt -f /dev/nrsa0 rewind dd if=/dev/zero of=/dev/nrsa0 bs=32k -- If I use error-free tape, tests pass: -- # time dd if=/dev/zero of=/dev/nrsa0 bs=32k dd: /dev/nrsa0: No space left on device 1856046+0 records in 1856045+0 records out 60818882560 bytes transferred in 4814.912483 secs (12631358 bytes/sec) real 83m3.857s user 0m2.228s sys 1m44.261s -- Until now, we have had Mammoth II with firmware v02p or lower. If I use bad tape with this firmware, there is still no problem: dd ends with I/O error and in syslog I can find: -- WRITE(06). CDB: a 0 0 80 0 0 MEDIUM ERROR csi:0,0,0,6 asc:9,0 Track following error WRITE FILEMARKS. CDB: 10 0 0 0 2 0 MEDIUM ERROR csi:0,0,0,6 asc:9,0 Track following error failed to write terminating filemark(s) tape is now frozen- use an OFFLINE, REWIND or MTEOM command to clear this state -- But by now, we are trying a new firmware v02r and if I try a bad tape, dd locks in "D" ps-state and tons of messages are generated in /var/log/messages: -- WRITE(06). CDB: a 0 0 80 0 0 Deferred Error: MEDIUM ERROR asc:9,0 Track following error -- I can only reboot server. I tested Mammoth2 with a new firmware v02r and with a bad tape on Novell NetWare and on Solaris too. I have to say that these systems have not any problems. Both stop writing process and write reasonable error message. Do you have any idea, what to do? I'm not SCSI guru and do not understand meaning of "deferred error". Thanks. -- Rudolf Cejka (cejkar@dcse.fee.vutbr.cz; http://www.fee.vutbr.cz/~cejkar) Brno University of Technology, Faculty of El. Engineering and Comp. Science Bozetechova 2, 612 66 Brno, Czech Republic To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message