Hello:
In a new installation of amd64 (raid 1 two devices md0 md1) from the recent debian amd64 testing installer, I made some mistake as to md0, whereby the system does not boot. The situation is illustrated below (booting live SystemRescueCD):

root@sysresccd /root % fsarchiver probe simple gig64

[======DISK======] [=============NAME==============] [====SIZE====] [MAJ] [MIN]

[sda ] [WDC WD1002FAEX-0 ] [ 931.51 GB] [ 8] [ 0]

[sdb ] [WDC WD1002FAEX-0 ] [ 931.51 GB] [ 8] [ 16]

[sr0 ] [DVD RW AD-7260S ] [ 380.83 MB] [ 11] [ 0]

[=====DEVICE=====] [==FILESYS==] [======LABEL======] [====SIZE====] [MAJ] [MIN]

[loop0 ] [squashfs ] [<unknown> ] [ 266.84 MB] [ 7] [ 0]

[sda1 ] [linux_raid_] [gig64:0 ] [ 285.00 MB] [ 8] [ 1]

[sda2 ] [linux_raid_] [gig64:1 ] [ 931.23 GB] [ 8] [ 2]

[sdb1 ] [linux_raid_] [gig64:0 ] [ 285.00 MB] [ 8] [ 17]

[sdb2 ] [linux_raid_] [gig64:1 ] [ 931.23 GB] [ 8] [ 18]

[md0 ] [LVM2_member] [<unknown> ] [ 284.69 MB] [ 9] [ 0]

[md1 ] [LVM2_member] [<unknown> ] [ 931.11 GB] [ 9] [ 1]

[dm-0 ] [ext3 ] [<unknown> ] [ 3.72 GB] [253] [ 0]

[dm-1 ] [ext3 ] [<unknown> ] [ 55.88 GB] [253] [ 1]

[dm-2 ] [ext3 ] [<unknown> ] [ 10.24 GB] [253] [ 2]

[dm-3 ] [ext3 ] [<unknown> ] [ 11.18 GB] [253] [ 3]

[dm-4 ] [ext3 ] [<unknown> ] [ 3.72 GB] [253] [ 4]

[dm-5 ] [swap ] [<unknown> ] [ 37.25 GB] [253] [ 5]

[dm-6 ] [ext3 ] [<unknown> ] [ 809.11 GB] [253] [ 6]



Actually, instead of LVM2-member, it would be like below, in another amd 64 (wheezy) installation on another box:

root@tya64:/home/francesco# fsarchiver probe simple tya64

[======DISK======] [=============NAME==============] [====SIZE====] [MAJ] [MIN]

[sda ] [ST500DM002-1BD14 ] [ 465.76 GB] [ 8] [ 0]

[sdb ] [ST500DM002-1BD14 ] [ 465.76 GB] [ 8] [ 16]

[=====DEVICE=====] [==FILESYS==] [======LABEL======] [====SIZE====] [MAJ] [MIN]

[sda1 ] [linux_raid_] [tya64:0 ] [ 187.00 MB] [ 8] [ 1]

[sda2 ] [linux_raid_] [tya64:1 ] [ 465.29 GB] [ 8] [ 2]

[sdb1 ] [linux_raid_] [tya64:0 ] [ 187.00 MB] [ 8] [ 17]

[sdb2 ] [linux_raid_] [tya64:1 ] [ 465.29 GB] [ 8] [ 18]

[md0 ] [ext2 ] [<unknown> ] [ 186.81 MB] [ 9] [ 0]

[md1 ] [LVM2_member] [<unknown> ] [ 465.16 GB] [ 9] [ 1]

[dm-0 ] [ext3 ] [<unknown> ] [ 952.00 MB] [253] [ 0]

[dm-1 ] [swap ] [<unknown> ] [ 13.97 GB] [253] [ 1]

[dm-2 ] [ext3 ] [<unknown> ] [ 27.94 GB] [253] [ 2]

[dm-3 ] [ext3 ] [<unknown> ] [ 9.31 GB] [253] [ 3]

[dm-4 ] [ext3 ] [<unknown> ] [ 9.31 GB] [253] [ 4]

[dm-5 ] [ext3 ] [<unknown> ] [ 2.79 GB] [253] [ 5]

[dm-6 ] [ext3 ] [<unknown> ] [ 400.47 GB] [253] [ 6]


Grateful for indications how to remedy (from live SystemrescueCD, or otherwise). I have no experience with such manipulations

Examining with the used amd testing installer, the error followed from errors in configured partitions:

LVM VG vg1 etc for md1
.......................................
........................................
RAID1 device #0 - 298.5 MB Linux Software RAID Array
#1 298.5 MB F lvm
RAID1 device #1 - 999.8 GB Linux Software RAID device
#1 999.8 GB K lvm
512.0 B unusable
SCSI1 (0,0,0) (sda) 1.0 TB ATA ....
1MB FREE SPACE
#1 298.8 MB K raid
#2 999.9 GB K raid
728.6 kB FREE SPACE

SCSI2 (0,0,0) (sdb) 1.0 TB ATA ....
1MB FREE SPACE
#1 298.8 MB K raid
#2 999.9 GB K raid
728.6 kB FREE SPACE

Well, with the Debian amd64 testing installer that I downloaded there is no way to get a bootable partition under md0. I retried, when used as physical volume for raid, the bootable flag gets OFF.

I retried from scratch. The md0 would-be-boot partition was correctly set ext2, as indicated below (although it is not clear whether "boot" was accepted:


root@sysresccd /root % fsarchiver probe simple
[======DISK======] [=============NAME==============] [====SIZE====] [MAJ] [MIN]
[sda ] [WDC WD1002FAEX-0 ] [ 931.51 GB] [ 8] [ 0]
[sdb ] [WDC WD1002FAEX-0 ] [ 931.51 GB] [ 8] [ 16]
[sr0 ] [DVD RW AD-7260S ] [ 380.83 MB] [ 11] [ 0]

[=====DEVICE=====] [==FILESYS==] [======LABEL======] [====SIZE====] [MAJ] [MIN]
[loop0 ] [squashfs ] [<unknown> ] [ 266.84 MB] [ 7] [ 0]
[sda1 ] [linux_raid_] [gig64:0 ] [ 285.00 MB] [ 8] [ 1]
[sda2 ] [linux_raid_] [gig64:1 ] [ 931.23 GB] [ 8] [ 2]
[sdb1 ] [linux_raid_] [gig64:0 ] [ 285.00 MB] [ 8] [ 17]
[sdb2 ] [linux_raid_] [gig64:1 ] [ 931.23 GB] [ 8] [ 18]
[md0 ] [ext2 ] [<unknown> ] [ 284.69 MB] [ 9] [ 0]
[md1 ] [LVM2_member] [<unknown> ] [ 931.11 GB] [ 9] [ 1]
[dm-0 ] [ext3 ] [<unknown> ] [ 3.72 GB] [253] [ 0]
[dm-1 ] [ext3 ] [<unknown> ] [ 55.88 GB] [253] [ 1]
[dm-2 ] [ext3 ] [<unknown> ] [ 10.24 GB] [253] [ 2]
[dm-3 ] [ext3 ] [<unknown> ] [ 11.18 GB] [253] [ 3]
[dm-4 ] [ext3 ] [<unknown> ] [ 3.72 GB] [253] [ 4]
[dm-5 ] [swap ] [<unknown> ] [ 37.25 GB] [253] [ 5]
[dm-6 ] [ext3 ] [<unknown> ] [ 809.11 GB] [253] [ 6]
root@sysresccd /root %


With live SystemRescueDC, /dev/md0 partition /dev/md0p1 has problems:

e2label: No such file or directory while trying to open /dev/md0p1
Couldn't find valid filesystem superblock.

tune2fs 1.42.9 (28-Dec-2013)

tune2fs: No such file or directory while trying to open /dev/md0p1
Couldn't find valid filesystem superblock.

Couldn't find valid filesystem superblock.

dumpe2fs 1.42.9 (28-Dec-2013)
dumpe2fs: No such file or directory while trying to open /dev/md0p1

Unable to read the contents of this file system!
Because of this some operations may be unavailable.
The cause might be a missing software package.
The following list of software packages is required for ext2 file system
support: e2fsprogs.
xxxxxxxxxxxxxxxxxxxxxxx

I paid attention to the installation progress, the would-be-boot partition was formatted ext2, however, grub was installed (efi or pc, not clear) automatically, not at request as in the good days of the past.

Debian maintainers should recognize that many users in the science area have no time (and no interest) to learn all subtleties of the OS, while they may need updated libraries, and therefore go to testing. They have to follow
the scientific literature in their area. That testing installer appears as a testing, which should not at that level.

ext2 not supported?

perhaps better to go to a reliable OS (Gentoo?)

SOLVED means that I stopped wasting my time, going to the wheezy installer, getting what I wanted, ie grub under md0 (which badly failed twice with the testing installer) and LV under md1. Now the long procedure of dist-upgrading.