Raid5 not starting up after reboot

English Install/Boot/Login
#1

Running OpenSuse 11.4 and have setup 2 x Raid5 configs - raid created, disks format, everything working fine. I’ve just rebooted and the raid5 fails to initialize.

Getting these errors:

May 29 16:58:30 suse kernel:  1788.170692] md: md0 stopped.
May 29 16:58:30 suse kernel:  1788.197864] md: invalid superblock checksum on sdb1
May 29 16:58:30 suse kernel:  1788.197876] md: sdb1 does not have a valid v0.90 superblock, not importing!
May 29 16:58:30 suse kernel:  1788.197903] md: md_import_device returned -22
May 29 16:58:30 suse kernel:  1788.201565] md: bind<sde2>
May 29 16:58:30 suse kernel:  1788.207588] md: invalid superblock checksum on sda2
May 29 16:58:30 suse kernel:  1788.207701] md: sda2 does not have a valid v0.90 superblock, not importing!
May 29 16:58:30 suse kernel:  1788.207715] md: md_import_device returned -22
May 29 16:58:30 suse kernel:  1788.292422] md: md1 stopped.
May 29 16:58:30 suse kernel:  1788.334886] md: bind<sde1>
May 29 16:58:30 suse kernel:  1788.336734] md: bind<sdf1>
May 29 16:58:30 suse kernel:  1788.337498] md: invalid superblock checksum on sda1
May 29 16:58:30 suse kernel:  1788.337508] md: sda1 does not have a valid v0.90 superblock, not importing!
May 29 16:58:30 suse kernel:  1788.337516] md: md_import_device returned -22
May 29 16:58:30 suse kernel:  1788.385477] bio: create slab <bio-1> at 1
May 29 16:58:30 suse kernel:  1788.385526] md/raid:md1: device sdf1 operational as raid disk 2
May 29 16:58:30 suse kernel:  1788.385532] md/raid:md1: device sde1 operational as raid disk 1
May 29 16:58:30 suse kernel:  1788.386242] md/raid:md1: allocated 3179kB
May 29 16:58:30 suse kernel:  1788.389609] md/raid:md1: raid level 5 active with 2 out of 3 devices, algorithm 2
May 29 16:58:30 suse kernel:  1788.389619] RAID conf printout:
May 29 16:58:30 suse kernel:  1788.389623]  --- level:5 rd:3 wd:2
May 29 16:58:30 suse kernel:  1788.389630]  disk 1, o:1, dev:sde1
May 29 16:58:30 suse kernel:  1788.389634]  disk 2, o:1, dev:sdf1
May 29 16:58:30 suse kernel:  1788.389719] md1: detected capacity change from 0 to 3000595644416
May 29 16:58:30 suse kernel:  1788.413562]  md1: unknown partition table

I have managed to get it working using something I found on one of the forums:

mdadm --assemble --verbose --update summaries /dev/md1 /dev/sda1 /dev/sde1 /dev/sdf1

mdadm --assemble --verbose --update summaries /dev/md0 /dev/sda2 /dev/sdb1 /dev/sde2

My mdadm.conf file:

DEVICE /dev/sda2 /dev/sdb1 /dev/sde2
ARRAY /dev/md0 UUID=8d0e9eaa:ab89d7f1:8b94c90b:72da1a08

DEVICE /dev/sda1 /dev/sde1 /dev/sdf1
ARRAY /dev/md1 UUID=f77c10b6:da833e9c:8b94c90b:72da1a08

mdadm --detail /dev/md0

/dev/md0:
        Version : 0.90
  Creation Time : Fri May 27 18:35:58 2011
     Raid Level : raid5
     Array Size : 976718848 (931.47 GiB 1000.16 GB)
  Used Dev Size : 488359424 (465.74 GiB 500.08 GB)
   Raid Devices : 3
  Total Devices : 3
Preferred Minor : 0
    Persistence : Superblock is persistent

    Update Time : Sun May 29 17:13:41 2011
          State : clean
 Active Devices : 3
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 0

         Layout : left-symmetric
     Chunk Size : 512K

           UUID : 8d0e9eaa:ab89d7f1:8b94c90b:72da1a08 (local to host suse)
         Events : 0.21951

    Number   Major   Minor   RaidDevice State
       0       8        2        0      active sync   /dev/sda2
       1       8       17        1      active sync   /dev/sdb1
       2       8       66        2      active sync   /dev/sde2

mdadm --detail /dev/md1

/dev/md1:
        Version : 0.90
  Creation Time : Fri May 27 19:04:57 2011
     Raid Level : raid5
     Array Size : 2930269184 (2794.52 GiB 3000.60 GB)
  Used Dev Size : 1465134592 (1397.26 GiB 1500.30 GB)
   Raid Devices : 3
  Total Devices : 3
Preferred Minor : 1
    Persistence : Superblock is persistent

    Update Time : Sun May 29 17:13:49 2011
          State : clean
 Active Devices : 3
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 0

         Layout : left-symmetric
     Chunk Size : 512K

           UUID : f77c10b6:da833e9c:8b94c90b:72da1a08 (local to host suse)
         Events : 0.53851

    Number   Major   Minor   RaidDevice State
       0       8        1        0      active sync   /dev/sda1
       1       8       65        1      active sync   /dev/sde1
       2       8       81        2      active sync   /dev/sdf1

Any ideas? Its driving me nuts!

#2

What kind of RAID is this to start? Software, Real Hardware, FAKE (BIOS Assisted)?

#3

Its a software RAID.

#4

How did you create the RAID? Via YaST or with the help of rescue CDs and command line programs?

#5

Hi
Are they EARS Drives? Maybe you need to configure as 4K blocks/sectors first?

#6

I used the command line to create them. mdadm --create

EARS drives? They’re Samsung SATA Drives.

#7

Perhaps you should override the superblock version to 1.0 because 1.0 is what I got when I used YaST to create it instead of invoking mdadm --create manually.

#8

How do I got about doing that?

#9

-e, --metadata=, according to the mdadm man page. Or use YaST to create your array.

#10

I’ve already created the array though. Will changing the metadata on a working array corrupt it in any way?

#11

I don’t think you can change the metadata location non-destructively. You’ll have to start over.

In case you are avoiding YaST because you have only CLI, you do know that YaST will run in character cell graphics mode (ncurses)?

#12

Thanks - dont think I"ll be starting over. I’ll just add the fixup scripts to the rc.local :slight_smile: