From owner-freebsd-stable Mon Feb 9 14:49:10 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA13201 for stable-outgoing; Mon, 9 Feb 1998 14:49:10 -0800 (PST) (envelope-from owner-freebsd-stable@FreeBSD.ORG) Received: from us.net (laurel.us.net [198.240.72.4]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA13172 for ; Mon, 9 Feb 1998 14:48:56 -0800 (PST) (envelope-from jjw@us.net) Received: from q.jjw.us.net (q.jjw.us.net [207.244.202.2]) by us.net (8.8.8/8.8.8) with SMTP id QAA04166 for ; Mon, 9 Feb 1998 16:24:33 -0500 (EST) X-Provider: US Net - Advanced Internet Services - (301) 572-5926 - info@us.net Where Business Connects! (tm) -- http://www.us.net/ Message-ID: <34DF740C.41C67EA6@us.net> Date: Mon, 09 Feb 1998 16:24:28 -0500 From: John Woodruff Organization: US Net X-Mailer: Mozilla 3.01Gold (X11; I; FreeBSD 2.2.5-RELEASE i386) MIME-Version: 1.0 To: freebsd-stable@FreeBSD.ORG Subject: Question: BSDI unsliced disks w/ FBSD Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk I'm trying to mount a SCSI disk from a good BSDI 1.1 system on FreeBSD-2.2.6. The disk was unsliced, but it looks like FreeBSD insists on applying the slice offsets from what it thinks is an fdisk slice table, thus getting preposterous sizes and offsets. The result is I can't mount the disk, and I *really* don't want to loose what's on it. How can I convince FreeBSD that this is really a unsliced disk? Details follow. Thanks in advance... The disk probes as: (ahc0:2:0): "QUANTUM FIREBALL1080S 1Q09" type 0 fixed SCSI 2 sd2(ahc0:2:0): Direct-Access 1042MB (2134305 512 byte sectors) disklabel says: disklabel /dev/rsd2c > "DIOCGDINFO: Invalid argument" disklabel -r /dev/rsd2c > "Undefined error: 0" disklabel /dev/rsd2 > Decent-looking synthesized label disklabel -r /dev/rsd2 > Looks like actual label from BSDI In *all 4* cases, I also get log messages: sd2s2: raw partition size != slice size sd2s2: start 0, end 4294901759, size 4294901760 # preposterous sd2s2c: start 0, end 2128895, size 2128896 # correct part c "Decent-looking" means the synthesized c partition size matched what the disk probe said, and offset was 0. The only other diff between the synthesized and actual label (besides only 'c' being defined in the synthesized one) was the geometry info. The sds2c log message reports what the actual label 'c' partiton has. I tried creating partition nodes with dkslice() = WHOLE_DISK_SLICE: mknod /dev/sd2wa b 4 0x00010010 mknod /dev/rsd2wa c 13 0x00010010 but "fsck -n /dev/rsd2wa" got nothing (and the same kernel megs) Reading sd(4), it sounds like I could use DIOCSDINFO to jam in a decent label, since I think I know what it is. Is there a disklabel operation (or other program) that calls this, or should I try to write one? As I said, I don't want to screw this disk... -- John Woodruff, Sr. Network Engineer, US Net - 301-572-5926 Washington/Baltimore/Richmond ISP - $6.95/month for full PPP! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe stable" in the body of the message