Your message dated Tue, 13 Jul 2010 02:45:16 +0100
with message-id <[email protected]>
and subject line Re: Bug#588303: grub-common: device.map order wrong since 
using uuids
has caused the Debian Bug report #588303,
regarding grub-common: device.map order wrong since using uuids
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)


-- 
588303: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=588303
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: grub-common
Version: 1.98+20100706-1

Hello,

my device.map's device order changed since grub uses uuid device-names.
Before uuid device-names it was:
        (hd0)   /dev/sda
        (hd1)   /dev/sdb
        (hd2)   /dev/sdc
        (hd3)   /dev/sdd
        (hd4)   /dev/sde
        (hd5)   /dev/sdf
This was actually even correct.

Now it is:
        $ while read hd id; do echo "$hd\t$(realpath $id)"; done < 
/boot/grub/device.map
        (hd0)   /dev/sde
        (hd1)   /dev/sdd
        (hd2)   /dev/sdc
        (hd3)   /dev/sdf
        (hd4)   /dev/sda
        (hd5)   /dev/sdb
This is completely off the mark. It seems like grub-mkdevicemap just
sorts the /dev/disk/by-id/* names alphabetically - which was good before
but, of course, is not anymore.

Just for the records: my EDD device order is:
        $ ls -l /dev/disk/by-id/edd-* | sed -e 's%^.*/dev%/dev%'
        /dev/disk/by-id/edd-int13_dev80 -> ../../sda
        /dev/disk/by-id/edd-int13_dev81 -> ../../sdb
The other devices have no Volume ID (in fact, they have a zeroed MBR)
and thus cannot be assigned to EDD devices.


regards
   Mario
-- 
The social dynamics of the net are a direct consequence of the fact that
nobody has yet developed a Remote Strangulation Protocol.  -- Larry Wall

Attachment: signature.asc
Description: Digital signature


--- End Message ---
--- Begin Message ---
Source: grub2
Source-Version: 1.98+20100710-1

On Wed, Jul 07, 2010 at 11:24:05AM +0200, Mario 'BitKoenig' Holbe wrote:
> On Wed, Jul 07, 2010 at 09:24:50AM +0100, Colin Watson wrote:
> > On Wed, Jul 07, 2010 at 09:32:17AM +0200, Mario 'BitKoenig' Holbe wrote:
> > > my device.map's device order changed since grub uses uuid device-names.
> > If device.map order is relevant *at all*, then you are doing something
> > terribly wrong.
> ...
> > Did this actually break something for you?
> 
> No, it doesn't break something for me. In fact, on the machine in
> question I'm still using lilo (for several reasons) and just carefully
> watch the progress of grub.
> I just didn't know whether the device.map's device order was important
> or not. I noticed the change and thought it could be a good idea to tell
> you about this because the former device.map made sense to me while the
> current doesn't :)

Vadim Solomin submitted a patch which fixes this as a follow-up to
#554790, which I merged with some changes in 1.98+20100710-1.  I forgot
about this bug until a moment ago, but it can be closed now and I'm
doing so with this message.

grub2 (1.98+20100710-1) unstable; urgency=low

  * New Bazaar snapshot.
    - Handle degraded RAID arrays in grub-probe and grub-setup.
    - Fix gfxterm pager handling.

  [ Fabian Greffrath ]
  * Get value of correct debconf question when deciding whether to purge
    /boot/grub (closes: #588331).

  [ Colin Watson ]
  * Generate device.map in something closer to the old ordering (thanks,
    Vadim Solomin).

  [ Updated translations ]
  * Croatian (hr.po) by Josip Rodin, closes: #588350.
  * French (fr.po) by Christian Perrier (closes: #588695).

 -- Colin Watson <[email protected]>  Mon, 12 Jul 2010 11:46:53 +0100

Thanks,

-- 
Colin Watson                                       [[email protected]]


--- End Message ---

Reply via email to