>
> What is needed is a "raidhotremove --DOIT!!" to allow you to remove an active
> disk ... (people have said this should be possible in the kernel -- but
> tricky. Sounds pretty simple to me, but I'm not an expert ...)
>
> > now that everything lives in the kernel, i dont see how that would be
> > possible.
>
> I can think of HACKs ....
>
> > i tried using raidhot<remove|add> but since the corrupted disks are still
> > listed as active, removing fails.
>
> Quite :-((
>
When I first started playing with the new raid tools/kernel-patch, it was with
kernel 2.2.3, it was possible to do a
echo "scsi remove-single-device">/proc/scsi/scsi
on a running raid disk if there was no ordinary partition on the disk which was
mounted at the time. By touching all the partitions which were mirrored I could
flag them as failed and then remove them with raidhotremove. This seems not
to be possible with the 2.2.6 which I am using now, it seems that something has
been "fixed". I havent checked but something must have been changed to tell
the kernel if a partition on a disk is part of an md device then it is in use -
shame
it was a very nice feature. Perhaps we should try and find out what has changed and
make a patch to go back to the old situation.
Brian Murphy
