Date: Wed, 12 Apr 2000 18:13:03 +0100 From: Adam Laurie <adam@algroup.co.uk> To: "Aleksandr A.Babaylov" <babolo@links.ru> Cc: freebsd-bugs@FreeBSD.ORG, Most Worshipful System Administrators <sysadmin@aldigital.co.uk> Subject: Re: freebsd 3.4 disklabel bug Message-ID: <38F4AE9F.54FC3672@algroup.co.uk> References: <200004121615.UAA09489@aaz.links.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
"Aleksandr A.Babaylov" wrote: > > > 8 partitions: > > # size offset fstype [fsize bsize bps/cpg] > > c: 40032696 0 unused 0 0 # (Cyl. 0 - > > 2491*) > > e: 40032696 0 4.2BSD 1024 8192 16 # (Cyl. 0 - > > 2491*) > ------^ asterick mean that not full cylinder Ah! > 0skiv~(1)>dc > 16065 2491 *p > 40017915 > 40032696 40017915 -p > 14781 > so 2491 cylinders full and part of one more cylinder. OK. > > > note that it thinks it has 2491 cylinders, but then uses 0 - 2491, i.e. > > 2492. > > > > This will go unnoticed if your offending partition is swap, as it > > doesn't try to write to swap until it errr... swaps. However, if you > > have a 2nd hard disk, and you allocate the entire disk, you will be > > unable to newfs it (from dmesg): > > > > wd1e: hard error writing fsbn 40031999 (wd1 bn 40031999; cn 2491 tn > > 223 sn 35) (status 51<rdy,seekdone,err> error 10<no_id>) > what dmesg say about disk size? > I often use disks in such a way - with not full last cylinder. > It needs apply right -t -u -c to use whole disk. > > > Setting it to this: > > > > # size offset fstype [fsize bsize bps/cpg] > > c: 40032696 0 unused 0 0 # (Cyl. 0 - > > 2491*) > > e: 40016631 0 4.2BSD 1024 8192 16 # (Cyl. 0 - > > 2490*) > what about > e: 40031999 0 ? > may be full disk size is 40032000? I think we tried lower numbers than that and the error followed to a lower number - I'm not sure what newfs is trying to write there, as it fails right at the beginning of the process, not at the end. Sorry to be vague, but we've shipped the machines now, so I don't have access to them to do any further testing... However, the automatic setup processes most definitely fail, so the block count is wrong even if the cylinder* is correct - I may have been able to squeeze some more space out of it by playing with block counts, but I was happy in this case to simply drop it by a whole cylinder. We're getting some more in shortly so I'll see if we can get more info... cheers, Adam -- Adam Laurie Tel: +44 (181) 742 0755 A.L. Digital Ltd. Fax: +44 (181) 742 5995 Voysey House Barley Mow Passage http://www.aldigital.co.uk London W4 4GB mailto:adam@algroup.co.uk UNITED KINGDOM PGP key on keyservers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-bugs" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?38F4AE9F.54FC3672>