Alan Cox wrote:
>
> > found on the net (I'm not sure but maybe 1.10?). And using kernel 2.1.102
> > in an all scsi system with an IBM 4.1 Gb UW, booting from it with lilo
> > *will* trash any partition greater than 1.9 Gb, I mean e2fsck will try to
> > access far beyond the end of the device and will trash all data.
>
> Thats strange. You don't say what driver you are using.
>
> For reference
>
> /dev/sda1 1012847 936321 24201 97% /
> /dev/sda6 2985176 1902146 928663 67% /usr/src
>
> Host: scsi0 Channel: 00 Id: 03 Lun: 00
> Vendor: IBM Model: DCAS-34330 Rev: S65A
> Type: Direct-Access ANSI SCSI revision: 02
>
> 2.1.105 #10 SMP Wed Jun 10 14:15:57 BST 1998 i586 unknown
>
> Buslogic SCSI
>
> and that seems to be behaving
>
I had the same problem when I was porting my own distribution
from libc5 to libc6. It killed my whole 6GB "/usr/src" partition...
It depends on the library version: with glibc, e2fsck has some
problems with llseek function declarations. Ted Tso released a
e2fsprogs-1.12-WIP version, which solves this problem.
>From the release-notes:
> E2fsprogs 1.12 (February 20, 1998)
> ==================================
>
> E2fsprogs now works with glibc (at least with the version shipped wtih
> RedHat 5.0). The ext2fs_llseek() function should now work even with
> i386 ELF shared libraries and if llseek() is not present. We also
> explicitly do a configure test to see if (a) llseek is in libc, and
> (b) if llseek is declared in the system header files. (See standard
> complaints about libc developers don't understand the concept of
> compatibility with previous versions of libc.)
[...]
Just another reason to be _very_ careful with glibc...!
- andreas
-- Andreas Haumer | email: andreas@xss.co.at | PGP key available *x Software + Systeme | phone: +43.1.6001508 | on request. Buchengasse 67/8 | +43.664.3004449 | A-1100 Vienna, Austria | fax: +43.1.6001507 |- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu