I'm having a really odd problem with grub. The history is that I have a server that has gone through some odd gyrartions with LVM on RAID1 and such. The Debian Sarge kernels woudln't boot properly because they wouldn't start raid properly and LVM would start up using the underlying devices instead of the raid onces. Using a second machine, I reproduced the problem and verified that using a stock kernel works (rather than the debian kernel-image-* packaged versions).
So... I'm trying to get the new kernel installed on the sick server. Using a stock kernel and set of libs that I have verified works on an identical box, I have tried to get the sick box to boot up. The problem is that though Grub properly sees the new kernel and loads it into memory, it doesn't seem to be able to kick it off. My boot screen stops with this: Booting 'Debian GNU/Linux, kernel 2.6.8.1 ' root (hd0,1) Filesystem type is reiserfs, partition type 0xfd kernel /vmlinuz-2.6.8.1 root=/dev/md1 ro console=tty1 console=ttyS0,9600 [Linux-bzImage, setup=0x1400, size=0x1cf7fb] savedefault boot Nothing happens after this. Only a power cycle seems to work. The really odd thing is that kernels on the sick server that have booted previously still work just fine. Any new kernel that is installed (including Debian kernel-image-* versions) has this problem. Grub doesn't seem to kick off the kernel init. I have tried reinstalling grub and reinstalling the boot sector stuff from the grub shell. Any suggestions on how to debug? --Ian _______________________________________________ Bug-grub mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/bug-grub