The meaning of "/dev/md0: No md super block found"?

PhillipHuang · 01-09-2007, 10:40 PM

This article is long, please read with kind patience.

Two questions about creating soft RAID by mdadm:

1. Need I to restart system after fdisk, and when start up again, do I have to format the partitions before create RAID?

I'm now reading some books about RHCE exam written by Michal Jang. He said, "When you configure RAID, you will format same spaces twice, the first format the patitions that make up the array, while the second format the RAID as if they were new partitions".

From my practices in lab, it seems that I must reboot the OS to activate new partitions. Later, I issue mdadm to create array while skip the first formation step above, the array is also built successfully.

I've googled for many pages, and do not confirm whether I did right things yet.

2. What's the meaning of "/dev/md0: No md super block found, not an md component."?

Here's my process:
fdisk (new partitions assigned fd)--> reboot OS --> mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/sda5 /dev/sda6 --spare-devices=1 /dev/sda7
The array is created successfully without error, then check the status:

Code:

[root@Muyang ~]# mdadm --query /dev/md0
/dev/md0: 101.88MiB raid1 2 devices, 1 spare. Use mdadm --detail for more detail.
/dev/md0: No md super block found, not an md component.
[root@Muyang ~]# mdadm --detail /dev/md0
/dev/md0:
        Version : 00.90.01
  Creation Time : Wed Jan 10 10:03:33 2007
     Raid Level : raid1
     Array Size : 104320 (101.88 MiB 106.82 MB)
    Device Size : 104320 (101.88 MiB 106.82 MB)
   Raid Devices : 2
  Total Devices : 3
Preferred Minor : 0
    Persistence : Superblock is persistent

    Update Time : Wed Jan 10 10:03:37 2007
          State : clean
 Active Devices : 2
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 1


    Number   Major   Minor   RaidDevice State
       0       8        5        0      active sync   /dev/sda5
       1       8        6        1      active sync   /dev/sda6
       2       8        7       -1      spare   /dev/sda7
           UUID : 345a71c9:76dc0d4c:8e753193:96fde17c
         Events : 0.2

Superblock records the array information, and I think /dev/md0 means meta device of array, why does it say "not found md super block"? Is the concept of "superblock" based on hardware, not devices as "/dev/md0"?

Please give me some hints.
Thanks in advance.
--Phillip

macemoneta · 01-09-2007, 10:59 PM

First, the only reason you need to restart the system is because you are altering the partition table on a dsisk that the system is using, probably for the root filesystem. When using real devices as opposed to the nonsensical configuration, you would not experience this. It would be better to experiment with loopback devices (see this).

Second, use the command:

mdadm --examine /dev/sda{5,6,7}

to examine the superblock on each device. I suspect the spare does not (yet) have a superblock.

PhillipHuang · 01-09-2007, 11:15 PM

Please see the following:

Code:

[root@Muyang ~]# mdadm --examine /dev/sda5
/dev/sda5:
          Magic : a92b4efc
        Version : 00.90.00
           UUID : 345a71c9:76dc0d4c:8e753193:96fde17c
  Creation Time : Wed Jan 10 10:03:33 2007
     Raid Level : raid1
    Device Size : 104320 (101.88 MiB 106.82 MB)
   Raid Devices : 2
  Total Devices : 3
Preferred Minor : 0

    Update Time : Wed Jan 10 10:05:23 2007
          State : clean
 Active Devices : 2
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 1
       Checksum : 51f0cb9 - correct
         Events : 0.4


      Number   Major   Minor   RaidDevice State
this     0       8        5        0      active sync   /dev/sda5
   0     0       8        5        0      active sync   /dev/sda5
   1     1       8        6        1      active sync   /dev/sda6
   2     2       8        7        2      spare   /dev/sda7
[root@Muyang ~]# mdadm --examine /dev/sda6
/dev/sda6:
          Magic : a92b4efc
        Version : 00.90.00
           UUID : 345a71c9:76dc0d4c:8e753193:96fde17c
  Creation Time : Wed Jan 10 10:03:33 2007
     Raid Level : raid1
    Device Size : 104320 (101.88 MiB 106.82 MB)
   Raid Devices : 2
  Total Devices : 3
Preferred Minor : 0

    Update Time : Wed Jan 10 10:05:23 2007
          State : clean
 Active Devices : 2
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 1
       Checksum : 51f0cbc - correct
         Events : 0.4


      Number   Major   Minor   RaidDevice State
this     1       8        6        1      active sync   /dev/sda6
   0     0       8        5        0      active sync   /dev/sda5
   1     1       8        6        1      active sync   /dev/sda6
   2     2       8        7        2      spare   /dev/sda7
[root@Muyang ~]# mdadm --examine /dev/sda7
/dev/sda7:
          Magic : a92b4efc
        Version : 00.90.00
           UUID : 345a71c9:76dc0d4c:8e753193:96fde17c
  Creation Time : Wed Jan 10 10:03:33 2007
     Raid Level : raid1
    Device Size : 104320 (101.88 MiB 106.82 MB)
   Raid Devices : 2
  Total Devices : 3
Preferred Minor : 0

    Update Time : Wed Jan 10 10:05:23 2007
          State : clean
 Active Devices : 2
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 1
       Checksum : 51f0cb9 - correct
         Events : 0.4


      Number   Major   Minor   RaidDevice State
this     2       8        7        2      spare   /dev/sda7
   0     0       8        5        0      active sync   /dev/sda5
   1     1       8        6        1      active sync   /dev/sda6
   2     2       8        7        2      spare   /dev/sda7

macemoneta · 01-09-2007, 11:35 PM

OK, good. If the output of:

mdadm --detail /dev/md0

Is also good, then the error is incorrect. I'm running mdadm-2.5.4-2 and not seeing the error, so it may be version related.

You should also check your /etc/mdadm.conf. In particular, make sure that the UUID, if specified matches that in the superblocks.

PhillipHuang · 01-10-2007, 12:25 AM

Thank you. The output of "mdadm --detail /dev/md0" also run well as shown:

Code:

[root@Muyang ~]# mdadm --detail /dev/md0
/dev/md0:
        Version : 00.90.01
  Creation Time : Wed Jan 10 10:03:33 2007
     Raid Level : raid1
     Array Size : 104320 (101.88 MiB 106.82 MB)
    Device Size : 104320 (101.88 MiB 106.82 MB)
   Raid Devices : 2
  Total Devices : 3
Preferred Minor : 0
    Persistence : Superblock is persistent

    Update Time : Wed Jan 10 10:05:23 2007
          State : clean
 Active Devices : 2
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 1


    Number   Major   Minor   RaidDevice State
       0       8        5        0      active sync   /dev/sda5
       1       8        6        1      active sync   /dev/sda6
       2       8        7       -1      spare   /dev/sda7
           UUID : 345a71c9:76dc0d4c:8e753193:96fde17c
         Events : 0.4

My Operating System is CentOS4.4, mdadm-1.6.0-3.

Quote:

Is also good, then the error is incorrect.

Do you mean this is bug of mdadm-1.6.0-3, it reports wrong while the array is OK?

I've checked /etc/mdadm.conf, and sure it is right to assemble array.

Code:

[root@Muyang ~]# cat /etc/mdadm.conf 
DEVICE /dev/sda5 /dev/sda6 /dev/sda7
ARRAY /dev/md0 UUID=345a71c9:76dc0d4c:8e753193:96fde17c

Update, same thing happens on other machine with mdadm-1.7.0.

Code:

-bash-2.05b# mdadm --query /dev/md0
/dev/md0: 1535.94MiB raid1 2 devices, 0 spares. Use mdadm --detail for more detail.
/dev/md0: No md super block found, not an md component.
-bash-2.05b# mdadm -V
mdadm - v1.7.0 - 11 August 2004

Regards,
--Phillip

macemoneta · 01-10-2007, 09:26 AM

mdadm is currently at 2.6. There have been many changes to messages since 1.7. I think the message you are seeing can be safely ignored.

PhillipHuang · 01-10-2007, 08:37 PM

Macemoneta, thanks for your kind help. I decide to update the lastest mdadm-2.6.

Regards,
Phillip