Update WARNING: GRUB reconfigured INcorrectly.

No doubt about it - it’s got new items (1 wrong, 1 right), and it changed the disk enumeration for the primary GRUB entry (11.1).

In my case, GRUB/SuSE is on disk hd0, which is on a PCI Raidbus card. ALL Linux’es call this disk hd2, but BIOS does NOT. BIOS calls it hd0.

I did the updates last night. Where is ‘Update History’, to look further into this?

Should this be submitted to Development as a problem? Or should it be ignored, since all Linux’es have this enumeration problem?

Does SuSE have a system restore function, in case problems arise more serious than this?

You can overrule anything during install. Please check the posts in the forum about GRUB.
For your info:
in /boot/grub:
device.map contains the disk order.
menu.lst contains the booting options

I don’t think this is a specific openSUSE issue, I do think it’s got something to do with the 2.6.27 kernels.

This is device map in my box:
(hd2) /dev/disk/by-id/ata-ST3802110A_5LR74M8V
(hd0) /dev/disk/by-id/ata-Maxtor_33073U4_N4178KBC
(hd1) /dev/disk/by-id/ata-ST3802110A_5LRADN27

The Maxtor is on IDE secondary Master. That’s why it says hd0.

The Seagate shown as hd2 is the boot device in BIOS. In other words, BIOS calls it hd0. Set BIOS to boot from it, and it boots first. Since it is on a PCI Raidbus card, I can also set the first boot device as SCSI, and it boots first. To BIOS, it is hd0 any way it is configured.

Set boot device as hd1, and the Maxtor boots.

There are 3 Linux’es on ‘hd2’. All 3, when they were installed, called themselves ‘hd0’.

I can’t explain why. But that’s the fact, Jack!

This thread entailed diggin’ into this enumeration problem. It was not solved; a work-around through boot device change was achieved…

ALL configuration attempts are documented.

Any GRUB experts? I clicked somethin’ wrong and one won’t boot… - LinuxQuestions.org

It might not be an enumeration difference. It is a fault in code.

I just updated SuSE GRUB last night, fresh install 1st update for 11.1.

Today when I booted it up, it would not boot. The update did a disk enumeration check, which included GRUB, and it changed the menu entry from hd(correct),x, to hd(incorrect),x.

I changed it back, and no problem. It also added an entry for failsafe 11.1, and gave it the wrong hd#.

All 3 Linux’es on this PCI Raidbus disk, when installed, read disk configuration incorrectly. There is no code for checking enumeration when a PCI Raidbus card is installed.

To really throw in convolution, my Raidbus card has only one disk. The card can handle 4 disks - primary master and slave, and secondary master and slave.

BIOS WILL boot from my ODD’s first, since they are first boot device. They are both on ONBOARD IDE primary master and slave. BUT, set BIOS second boot device as hd0, and what boots? NOT the ONBOARD IDE secondary master, but the the RAIDbus card!!!

Why does BIOS boot device ‘read’ jump from ONBOARD, to PCI RAIDbus card?

AND, what if there are MULTIPLE disks on the RAIDbus card, in the RAIDbus’ card primary/secondary master/slave configuration???