Date: Thu, 11 Jul 1996 23:19:21 +0200 (MET DST) From: J Wunsch <j@uriah.heep.sax.de> To: freebsd-scsi@freebsd.org Cc: jbh@labyrinth.net.au (John Hartley) Subject: Re: tandberg scsi tape + FreeBSD 2.1/2.0.5 Message-ID: <199607112119.XAA16265@uriah.heep.sax.de> In-Reply-To: <199607102352.JAA25478@minotaur.labyrinth.net.au> from John Hartley at "Jul 11, 96 09:52:38 am"
next in thread | previous in thread | raw e-mail | index | archive | help
As John Hartley wrote: > Jul 10 21:42:02 qwiff /kernel: st0(ahc0:2:0): command: 1a,0,0,0,c,0-[12 bytes] > Jul 10 21:42:02 qwiff /kernel: ------------------------------ > Jul 10 21:42:02 qwiff /kernel: 000: 2f 25 10 08 00 00 00 00 00 00 02 00 > Jul 10 21:42:02 qwiff /kernel: ------------------------------ > Jul 10 21:42:02 qwiff /kernel: st0(ahc0:2:0): ststart st0(ahc0:2:0): density > code 0x0, 512-byte blocks, write-enabled, st0(ahc0:2:0): buffered > Present Mode: Density = ECMA TC17 Blocksize variable > ---------available modes--------- > Mode 0: Density = 0x00 Blocksize variable > Mode 1: Density = 0x00 Blocksize variable > Mode 2: Density = 0x00 Blocksize variable > Mode 3: Density = 0x00 Blocksize variable This looks self-contradicionary. The kernel message claims density code 0, while the mt status output claims density code 0x15. The kernel message says 512-byte blocks... What's the actual medium type in the drive while you captured the above? (Just curious.) > Jul 10 21:42:03 qwiff /kernel: st0(ahc0:2:0): command: 15,0,0,0,c,0-[12 bytes] > Jul 10 21:42:03 qwiff /kernel: ------------------------------ > Jul 10 21:42:04 qwiff /kernel: 000: 00 00 10 08 00 00 00 00 00 00 00 00 > Jul 10 21:42:04 qwiff /kernel: ------------------------------ > Jul 10 21:42:04 qwiff /kernel: code70 valid0 seg0 key5 ili0 eom0 fmark0 > Jul 10 21:42:04 qwiff /kernel: info: 0 0 0 0 followed by 10 extra bytes > Jul 10 21:42:04 qwiff /kernel: extra: 0 8 0 0 24 0 2 0 0 0 > Jul 10 21:42:04 qwiff /kernel: st0(ahc0:2:0): ILLEGAL REQUEST csi:0,8,0,0 > asc:24,0 Invalid field in CDB field replaceable unit: 2 I cannot find anything about the meaning of the CSI for sequential- access devices in the standard, so the 0,8,0,0 seems unintelligible to me. The above SCSI command looks reasonable if the cartridge in question is QIC-525 or higher. Basically, it requests the drive to use variable blocksize. This is IMHO invalid for QIC-150. So if your cartridge is QIC-150, does the following help you in any way? mt -f /dev/st0ctl.1 blocksize 512 mt -f /dev/st0ctl.1 density 0x10 tar tvf /dev/rst0.1 I *think* we need the same quirk record for the Tandberg 4100 that we're using for the 42xx series. Basically, it forces the drive to do one IO operation on the cartridge before retrieving the density and blocksize information. After this first IO operation, the returned data are valid, and can be used in subsequent IO. -- 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. ;-)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199607112119.XAA16265>