Bug 6751

Summary: System crash when starting RAID1 partition
Product: Mageia Reporter: Jean-Marc Boudeau <mageia>
Component: InstallerAssignee: Mageia Bug Squad <bugsquad>
Status: RESOLVED FIXED QA Contact:
Severity: critical    
Priority: Normal    
Version: 2   
Target Milestone: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Source RPM: systemd ? CVE:
Status comment:

Description Jean-Marc Boudeau 2012-07-11 12:01:04 CEST 
Description of problem:
After installation of a small server (2 disks in Raid1):
/boot ext4 /dev/md0 150Mb   (disks /dev/sda1 /dev/sdb1)
swap  swap /dev/md1 4000Mb  (disks /dev/sda2 /dev/sdb2)
/     ext4 /dev/md2 18000Mb (disks /dev/sda3 /dev/sdb3)
/data ext4 /dev/md3 30000Mb (disks /dev/sda4 /dev/sdb4)

Often (not always) the /data partition can't be started (sometimes CTRL-D is sufficient to continue the normal process and /data will be mounted - sometimes it is necessary to restart the server by CTRL-ALT-DEL or push power button). And I don't know if data on /data are or are not corrupted.

The error messages are:
systemd-fsck[1227]: fsck.ext4: Argument invalide lors de la tentative d'ouverture de /dev/md3
/dev/md3: Le superbloc n'a pu être lu ou ne contient pas un système de fichiers ext2 correct. Si le périphérique est valide et qu'il contient réellement un système de fichiers ext2 (et non pas de type swap, ufs ou autre), alors le superbloc est corrompu, et vous pourriez tenter d'exécuter...

The same installation with Mageia1 worked fine.

It seems to be a problem introduced by systemd.

It is impossible to use Mageia2 as small data server for me at this time.

The size of /data seems not to be the problem (I tested with 250000Mb, same problem).
The type of partition on disk (primary sdx4)) seems not to be the problem (I tested with extended sdx5, same problem).
The hardware seems not to be the problem (I tested with 3 different PCs AMD and Intel).
The Raid1 seems not to be the problem (I tested with 3 disks and /data in Raid5 with more problems - not only crash but data lost). 

Version-Release number of selected component (if applicable):
Mageia 2 x86_64 (not tested with i586)
Comment 1 Manuel Hiebel 2012-10-07 12:03:46 CEST 
hi can you try a netinstall from cauldron or from the coming alpha2 when you have time to see if it's fixed ?
(as the mageia2 will not be updated anymore iirc)

thanks

Source RPM: (none) => systemd ?

Comment 2 Jean-Marc Boudeau 2012-11-02 16:27:45 CET 
Hello,
Thank you for your interest to the bug with Raid system partition.
I just tested with Mageia 3 alpha2... and there is another major bug:
Linux refuses to start !

The system starts with LILO 23 - Boot Menu
I wait until the timeout (10s) and then the system displays
boot:
Loading linux

And nothing else... (the cursor stays just after the x of linux...).

I made the test just after the launch of alpha2 (with the same problem) but I wanted to be sure before to answer you.

Now I just retest with a simple configuration :
Blank disk(s)
2 partitions (swap 4000 Mb and / Ext4 20000 Mb)
Standard installation KDE desktop
First install: 1 disk (no raid => Grub) => no problem
Second install: 2 disks (raid1 => Lilo) => no boot

I can't answer to your question...
I will retest if you give me a way to test...

Jean-Marc
Comment 3 Jean-Marc Boudeau 2012-11-23 16:06:06 CET 
With Mageia 3 alpha3, it seems the 2 problems disappeared.

Thanks a lot.

Jean-Marc Boudeau

Status: NEW => RESOLVED
Resolution: (none) => FIXED