On Tue, 30 Aug 2011, martin f krafft wrote:
> also sprach Cristian Ionescu-Idbohrn <cristian.ionescu-idbo...@axis.com>
> [2011.08.30.1922 +0200]:
> > My experience is that everything boils down to device names.
> >
> > /dev/md0 and /dev/md/0 is the same device, AFAICT.
> > They should be handled as equaly. This trivial patch:
>
> These are handled the same by mdadm already. Please try to narrow
> down the bug more.
I'd very much like to do so. Any ideas what I should look for?
$INITRDSTART is empty and the output I see comming out of
'update-initramfs' is:
warn "I am supposed to start $i from the initial ramdisk,"
warn "yet I cannot find the array in the configuration file."
warn "I am thus reverting to starting all arrays."
$INITRDSTART gets set to 'all' and I see:
info "will start all available MD arrays from the initial ramdisk."
Still, 'lsmod' does not show modules 'md_mod' and 'raid1' loaded.
Without actively adding/uncommenting the /etc/initramfs-tools/modules
line:
raid1
I get the 'initramfs' prompt.
At that prompt, if I execute:
mdadm --assemble --scan
and ^D after that, the named modules are loaded (directory
/dev/disk/by-uuid shows up) and the boot process continues and finishes as
expected.
Where should I look, what should I look for?
Cheers,
--
Cristian
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
