From owner-freebsd-scsi Wed Feb 21 17:18:10 2001 Delivered-To: freebsd-scsi@freebsd.org Received: from dan.emsphone.com (dan.emsphone.com [199.67.51.101]) by hub.freebsd.org (Postfix) with ESMTP id 16B5437B401 for ; Wed, 21 Feb 2001 17:18:07 -0800 (PST) (envelope-from dan@dan.emsphone.com) Received: (from dan@localhost) by dan.emsphone.com (8.11.1/8.11.1) id f1M1I4V03017; Wed, 21 Feb 2001 19:18:04 -0600 (CST) (envelope-from dan) Date: Wed, 21 Feb 2001 19:18:03 -0600 From: Dan Nelson To: Matthew Jacob Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: "NO SENSE ILI" errors in 4.2? Message-ID: <20010221191803.A5967@dan.emsphone.com> References: <20010221160938.A27805@dan.emsphone.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.14i In-Reply-To: ; from "Matthew Jacob" on Wed Feb 21 16:47:11 GMT 2001 X-OS: FreeBSD 5.0-CURRENT Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In the last episode (Feb 21), Matthew Jacob said: > On Wed, 21 Feb 2001, Dan Nelson wrote: > > I just upgraded a bunch of 4.0 machines to 4.2 and I started getting > > the following messages scrolling on the console: > > > > media7 /kernel: (sa0:ahc0:0:1:0): READ(06). CDB: 8 0 0 ff 0 0 > > media7 /kernel: (sa0:ahc0:0:1:0): NO SENSE ILI (length mismatch): 32632 > > > > .. one per block read. The message is generated by scsi_sense_print, > > but I don't think it should be called in this case. ILI is not an > > error, and in my case it's supposed to happen (I always do 64k reads). > > Have you set the device to variable length mode? Well, I'm using the default tape settings: (dan@media7) /mnt/emssrv5/home/dan> mt -f /dev/rsa0.ctl status Mode Density Blocksize bpi Compression Current: 0x28:X3.224 variable 37871 disabled I've never had to explicitly set my tapes to variable before. I'm seeing this on a couple models of tape drive, and different controllers, so it's not a quirk entry or adapter bug biting me. Here's a bit more info. media2 and media7 exhibit the problem; media3 and media5 apparently don't. media2 and media5 have identical tape drives; they all have slightly different adapters. media2: aha1: at port 0x334-0x337 iomem 0xc8000-0xcbfff irq 10 drq 5 on isa0 aha1: AHA-1542CP FW Rev. F.0 (ID=46) SCSI Host Adapter, SCSI ID 7, 16 CCBs sa0 at aha1 bus 0 target 3 lun 0 sa0: Removable Sequential Access SCSI-2 device sa0: 10.000MB/s transfers (10.000MHz, offset 8) (sa0:aha1:0:3:0): READ(06). CDB: 8 0 0 ff 0 0 (sa0:aha1:0:3:0): NO SENSE ILI (length mismatch): 36330 media3: ahc0: port 0x1000-0x10ff mem 0xf4201000-0xf4201fff irq 11 at device 13.0 on pci0 sa0 at ahc0 bus 0 target 4 lun 0 sa0: Removable Sequential Access SCSI-2 device sa0: 4.032MB/s transfers (4.032MHz, offset 8) media5: aha0 at port 0x130-0x133 irq 11 drq 5 on isa0 aha0: AHA-1542CF FW Rev. E.0 (ID=45) SCSI Host Adapter, SCSI ID 7, 16 CCBs sa0: Removable Sequential Access SCSI-2 device sa0: 10.000MB/s transfers (10.000MHz, offset 8) media7: ahc0: port 0x1000-0x10ff mem 0xf4009000-0xf4009fff irq 9 at device 16.0 on pci0 sa0 at ahc0 bus 0 target 1 lun 0 sa0: Removable Sequential Access SCSI-2 device sa0: 20.000MB/s transfers (10.000MHz, offset 15, 16bit) ch0 at ahc0 bus 0 target 1 lun 1 (sa0:ahc0:0:1:0): READ(06). CDB: 8 0 0 7e e1 0 (sa0:ahc0:0:1:0): NO SENSE ILI (length mismatch): 1 -- Dan Nelson dnelson@emsphone.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message