On Mon, 2002-03-25 at 02:32, Ethan Benson wrote: > that is not possible, XFS has not left room at block zero of the > partition for as long as its existed, lilo when told you install on a > partition always puts it within the first 512 bytes, there is zero > chance that installing lilo will not result in the destruction of the > XFS superblock. you must be mistaken about how you had lilo configured. Previously i had this in /etc/lilo.conf (lilo-21.5-7) boot = /dev/sda2 prompt timeout = 300 default = Linux change-rules reset vga=792 # XFS image = /vmlinuz root = /dev/sda2 label = Linux read-only # LFS_3.1 image = /vmlinuz-lfs root = /dev/hde3 label = "LFS 3.1" read-only # Windows 2000 other = /dev/sda1 label = "Windows 2000" table = /dev/sda I bet if went back to XFS code from around Feb 20th i can still use this config without lilo overwriting the superblock on sda2...but i've put lilo on MBR and i'm happy with the way it works now :-). > it cannot be fixed, all that can be done is for lilo to check for XFS > magic before scribbling on partitions, and abort if it finds one. I would like to see this change in lilo -:) for one and it would save some boot suprises about VFS not being able to mout partitions... Regards, Paul -- _________________________________ ( http://www.linuxdiscussions.org )
Attachment:
signature.asc
Description: This is a digitally signed message part