On Tue, 6 Jan 2004 09:48:40 -0800 (PST) "Scott I. Remick" <[EMAIL PROTECTED]> probably wrote:
> > --- Sergey 'DoubleF' Zaharchenko <[EMAIL PROTECTED]> wrote: > > > I can't find a zero-bad floppy in this place! It's all the holidays! > > That's what AOL disks (vs. discs) used to be good for. :) > > > With `c', they're all offset by 63(why?). But still, you can mount the > > partitions on the ad4s1, so the disklabel should be ok... > > Yeah. Starts to suggest what we were thinking was a evidence related to the > problem is really unrelated and "normal" behavior (is disklabel/bsdlabel > only meant to be run on slices and not bsd-partitions?). Are we looking in > the wrong place? After trying out 5.2-RC2, it seems like the offsets reported with the `c' slice are from the beginning of the disk, not from the beginning of the slice. That accounts for the +63 difference. I guess it's documented somewhere, but as I don't use 5.x I haven't read its docs. > What about that potentially good superblock we found a > while ago? (the skip 16 one that contained "/data" in it) Should we be > saving that somewhere while we can? (how?) I think you already have a copy (the data at offset 32 seems to be it). If you want, do a # dd if=/dev/ad6s1 skip=16 count=16 of=/some/file Please tell me everything what you tried to use to mount/fsck the drive (and the results, of course). > Anyone out there know 5.x file-system dirtiness like the back of their hand? > C'mon, you know you wanna join the fun. :) Try booting from a 4.x floppy and doing it all over again... The FS is UFS1, isn't it? -- DoubleF Why does New Jersey have more toxic waste dumps and California have more lawyers? New Jersey had first choice.
Description: PGP signature