[lug] SW Raid question

[dio2002 at indra.com]

>
>> I guess I'm confused. It seems that you are saying that the block
>> addresses
>> that get written into the mbr have some kind of device specifier. I have
>> moved
>> a disk from sdb to sda where sdb was prepped using root (hd1,0) in grub.
>> Booted the former sdb in the sda position just fine with only the one
>> disk.
>> Didn't even appear to look for an sdb device. I did adjust the root
>> statement
>> in grub.conf to (hd0,0), but did nothing to any device lines.
>
>
> Currently the vote is 2 against 1. :)  I suspect I'll learn something.
> I've got a CentOS 4.4 minimal image installed now w/ /boot on /dev/md0
> (which is /dev/sda1 and /dev/sdb1) and everything else in a volume group.
>
> I've copied off the first 512 bytes of sda and sdb after the install.
> I'm snapshotting now.  After that, I'll turn off the VM, disable sda and
> check to see if it boots.
>
> Next, I'll re-enable sda, boot, do the example with
> device (hd0) /dev/sdb
>
> dd off the first 512 bytes of sda and sdb, shutdown, disable sda and see
> if it boots
>
> Next, I'll revert to the snapshot, do the setup (hd1,0) version of the
> example, dd off the first 512 byes of sda and sdb, disable sda and see
> if it boots.
>
> Finally, I'll diff the two boot sectors for sdb between the two examples.
>
> Sound like a reasonable test?

yuh lost me there somewhere in the middle.  but i think simply trying both
methods with a pulled drive drive 0 will suffice.  and i'm not sure if a
diff of the MBR will be decisive, but i would imagine if your comparing
the MBR on sdb method 1 to sdb method 2 and not sda to sdb, they probably
should be identical.  the key is that the boot works identically.