Felix Miata <mrmazda@ij.net> writes:
On 2006/11/28 16:56 (GMT+0100) Andreas Jaeger apparently typed:
Felix Miata <mrmazda@ij.net> writes:
On 2006/11/28 07:30 (GMT+-0500) Felix Miata apparently typed:
I did a FTP install from the beta2 mirror state IIRC. Shortly after I did some partition rearranging, moving the factory from hda21 to hda10. After editing the appropriate files and running grub-install, it would not boot when hda10 selected from IBM Boot Manager or when hda10 chainloaded from grub on another partition. Loading the kernel and initrd from grub elsewhere worked fine.
After updating to the rc1 mirror state I reran grub-install once again, and now it boots from BM, and presumably also from chainload.
The rc1 graphical grub boot menu is way cool!
I just stumbled onto a clue why my problem may have occurred, but don't know what to make of or do about it.
I just grabbed the newest default kernel rpm via ftp and ran rpm -ivh on it. After mkinitrd's bootsplash output came these two lines:
ERROR: Core::SplidDevPath: Cannot get device for (hd0,20)/boot/grub/stage1 ERROR: Core::SplidDevPath: Cannot get device for (hd0,20)/boot/grub/stage2
The error is understandable, as when initially installed factory was on hda21. Now hda21 no longer exists. I moved factory to hda10 over a week ago.
What triggered the errors? What is remembering the original installation target?
Reboot with the new 2.6.18.2-31-default kernel seemed totally normal.
Please file a bugreport for this!
What product and component? IOW, whose fault is it likely to be?
So, is this a 10.2 install? In that case against 10.2, component kernel,
/etc/grub.conf was left untouched from its original installation creation state (the old partition originally installed to) after I used YaST 2 System Update to upgrade to RC1.
Andreas -- Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj/ SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126