Couple of things re FC4 & yum

[Mike McCarty]

Gene Heskett wrote:

>On Monday 11 July 2005 11:24, Paul Howarth wrote:
>  
>
>>Gene Heskett wrote:
>>    
>>
>>>This machine is a bit odd in that its actual boot partition isn't
>>>even mounted by FC4.  Somehow, in rigging it for dual booting of
>>>FC4 and emc's bdi, its now booting from (hd1,0) instead of
>>>(hd0,0), so that when a new kernel is installed as was the case
>>>this morning, I have to hand modify the /dev/hdb1/grub/menu.lst
>>>and copy all the new stuffs from /dev/hda1 to /dev/hdb1, which
>>>when its booted to FC4, can be hand mounted as /mnt/bdi-boot. 
>>>/dev/hda1 is mounted by FC4 as /boot, but thats not where it boots
>>>from.
>>>      
>>>
>>I'd suggest adding an entry:
>>
>>title Fedora Core 4
>>        rootnoverify (hd0,1)
>>        chainloader +1
>>
>>to /dev/hdb1/grub/menu.lst
>>
>>You should then be able to pick FC4 from the OS's boot menu, and
>>never have to fiddle with grub entries again.
>>
>>Paul.
>>    
>>
>
>What would this do to the grub choice of boots menu?  And shouldn't 
>that be (hd0,0) which is the 'other' /boot partition that FC4 would 
>normally use if somehow grub hadn't been pointed at hdb?
>  
>
It will add a new entry.

>ISTR that at some point in screwing around with the hardware in that 
>box, I swapped the master/slave jumpering of the drives, which may be 
>why its now booting from hdb.  One drive is a 60GB and one is a 46GB.
>And I don't recall theres an option to control the booting device in 
>the bios...
>  
>
That would do it.

Here's how it works, from the BIOS point of view:

For each entry in the BOOT ENABLED DEVICE LIST do
    case DEVICE TYPE of
        CDROM: (I don't know what it does exactly)
        FLOPPY: read the absolute first sector
                         check that the last two bytes are 0x55 0xAA
                         if they are then break out of loop, we
                         found the boot sector
        HARDDISC:
                        read the absolute first sector
                        check that the last two bytes are 0x55 0xAA
                        if they are then break out of the loop, we
                        found the MBR.
     end case
end for

if we broke out of the loop, then jump to the place we read
    the bootstrap into
else
    display an error message about being unable to boot
endif

What happens from here on out depends on what is installed
in the MBR or Boot Sector. I know what happens from here
on out for DOS machines, but not for others. Anyway, there
is supposed to be an entry in the PT which shows exactly
one partition marked as ACTIVE (some call it BOOT)
on hard discs.

Mike

-- 
p="p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
This message made from 100% recycled bits.
I can explain it for you, but I can't understand it for you.
I speak only for myself, and I am unanimous in that!