On Mon, Sep 15, 2008 at 05:04:39PM +0100, Russell Howe wrote:
> Johan Strvm wrote, sometime around 15/09/08 16:39:
: :
> >
> >Yep, thats my plan too (or well 250G since 250G is almost as cheap as 
> >80G, and we are using 250G in other machines, no need for different 
> >spares), and use software raid. One thing I'm worried about though is if 
> >one disk fails, will the BIOS be able to boot from the other disk with a 
> >broken/empty disk in the first slot? I haven't seen any indications in 
> >the BIOS about being able to change, and I've had similar problems 
> >before (empty disk in slot1, disk with OS in slot2, box refusing to boot 
> >since disk1 is empty).
> 
> I don't think this will work with the way I have it set up at present. 
> The trick on Linux is to install the bootloader on disk 2 so that it is 
> configured to boot from disk 1 (as disk #2 will become disk #1 when disk 
> #1 is no longer there or operational). I haven't tried to figure out the 
> necessary magic for that as yet.

I have done this on a DL140 G3 with OpenBSD 4.1. I hope now with
OpenBSD 4.4 that RAIDFrame autoconfigured root disk is up and
working again. In 4.3 I recall it was broken, 

There is not much magic involved. Both disks have an partition 'a'
that contain /boot and /bsd (RAID enabled). You have to run
/usr/mdec/installboot on them (there is a bit of magic)
to install the boot sector and point it to /boot on the disk.

If you remove disk #1 the BIOS finds disk #2 and uses its boot
sector. It loads /boot that finds the disk it is on and loads
/bsd, which in its turn is RAID autoconfig enabled and
finds the broken rootdisk mirror that the lonely disk #2 is.

If you put a clean disk as #1 I am not dead sure what the BIOS does.
It should try to boot disk #2 if there is no boot sector on #1.
Then you can partition the disk and restore the mirror,
plus recreate the 'a' partition and boot sector.

That was only the minimal way. I have in fact
complete small installations on both 'a' partitions
i.e the installation used to compile the RAID kernel,
plus a /etc/boot.conf file, and /bsd, /bsd.old /bsd.raid
/obsd kernels, so it is possible to keyboard control /boot
and order it to boot a non-RAID-aware system just in case.

In the RAID drive the /etc/fstab has mount points for
/dev/wd0a and /dev/wd1a so they can be mounted to
copy new kernels and run /usr/mdec/installboot on them 
whenever /boot changes or moves.

> 
> -- 
> Russell Howe, IT Manager. <[EMAIL PROTECTED]>
> BMT Marine & Offshore Surveys Ltd.

-- 

/ Raimo Niskanen, Erlang/OTP, Ericsson AB

Reply via email to