3

I had a raid 5 with 3 disks in my old computer, but the motherboard died. I put all the disks in a New computer, but the raid doesn't want the start anymore. When booting the pc, i can see all disks, and mdadm -E /dev/sd [bcd]1 tells me all three disks are clean. Cat /proc/mdstat says /dev/md0 : inactive sdc13 sdd11 sdb10 I am running fedora 15 on this box. Any tips to get this raid started are greatly appreciated!

--Edited-- In the meantime I reinstalled the OS (on a new, separate disk, the os was never part of the raid), and used CentOS 7.

The outputs:

mdadm -E /dev/sda1                                                                                                           
/dev/sda1:                                                                                                                                        
          Magic : a92b4efc                                                                                                                        
        Version : 1.1                                                                                                                             
    Feature Map : 0x1                                                                                                                             
     Array UUID : aa1739a9:abf29975:85909ad4:7ce90400                                                                                             
           Name : HappyFiles:0
  Creation Time : Mon Jun  6 19:07:27 2011
     Raid Level : raid5
   Raid Devices : 3

 Avail Dev Size : 3907024896 (1863.01 GiB 2000.40 GB)
     Array Size : 3907023872 (3726.03 GiB 4000.79 GB)
  Used Dev Size : 3907023872 (1863.01 GiB 2000.40 GB)
    Data Offset : 2048 sectors
   Super Offset : 0 sectors
          State : clean
    Device UUID : 822d3a00:cf68046c:45de1427:d65beb68

Internal Bitmap : 8 sectors from superblock
    Update Time : Mon Nov 10 22:19:29 2014
       Checksum : 195e3a86 - correct
         Events : 1527192

         Layout : left-symmetric
     Chunk Size : 512K

   Device Role : Active device 1
   Array State : .AA ('A' == active, '.' == missing)


mdadm -E /dev/sdb1
/dev/sdb1:
          Magic : a92b4efc
        Version : 1.1
    Feature Map : 0x1
     Array UUID : aa1739a9:abf29975:85909ad4:7ce90400
           Name : HappyFiles:0
  Creation Time : Mon Jun  6 19:07:27 2011
     Raid Level : raid5
   Raid Devices : 3

 Avail Dev Size : 3907024896 (1863.01 GiB 2000.40 GB)
     Array Size : 3907023872 (3726.03 GiB 4000.79 GB)
  Used Dev Size : 3907023872 (1863.01 GiB 2000.40 GB)
    Data Offset : 2048 sectors
   Super Offset : 0 sectors
          State : clean
    Device UUID : 7546d111:71f87ce2:8b5e9cfb:05ed0bc4

Internal Bitmap : 8 sectors from superblock
    Update Time : Tue Nov 11 19:01:52 2014
       Checksum : d11417ca - correct
         Events : 1527199

         Layout : left-symmetric
     Chunk Size : 512K

   Device Role : Active device 2
   Array State : ..A ('A' == active, '.' == missing)


cat /proc/mdstat
Personalities :
md127 : inactive sdb1[3](S) sda1[1](S)
      3907024896 blocks super 1.1

unused devices: <none>

mdadm --assemble --scan
mdadm: No arrays found in config file or automatically

Apparently at this moment he lost one of his disks (not in /dev list either), probable a cable got loose after moving the computer. I'm accessing it via SSH from at the office right now (where I'm spending waaaaay too many time). Normally, mdadm -E /dev/sdX1 of the third disk gives similar output as the other two disks.

----EDIT----

The port of the third disk appeared to be broken, after connecting it to another card, I get to mdadm -E this one too!

     Magic : a92b4efc
        Version : 1.1
    Feature Map : 0x1
     Array UUID : aa1739a9:abf29975:85909ad4:7ce90400
           Name : HappyFiles:0
  Creation Time : Mon Jun  6 19:07:27 2011
     Raid Level : raid5
   Raid Devices : 3

 Avail Dev Size : 3907024896 (1863.01 GiB 2000.40 GB)
     Array Size : 3907023872 (3726.03 GiB 4000.79 GB)
  Used Dev Size : 3907023872 (1863.01 GiB 2000.40 GB)
    Data Offset : 2048 sectors
   Super Offset : 0 sectors
          State : clean
    Device UUID : 3c5c8512:49ba8111:bd936c82:00cb6b67

Internal Bitmap : 8 sectors from superblock
    Update Time : Fri May  4 09:51:16 2012
       Checksum : 262a346f - correct
         Events : 82967

         Layout : left-symmetric
     Chunk Size : 512K

   Device Role : Active device 0
   Array State : AAA ('A' == active, '.' == missing)

cat /proc/mdstat now says

     # cat /proc/mdstat
Personalities : 
md127 : inactive sdc1[3](S) sdb1[1](S) sda1[0](S)
      5860537344 blocks super 1.1

unused devices: <none>

So, how do I now get this raid back online, at least long enough to copy as much as possible of the data off of it?

  • show full output – frostschutz Nov 16 '14 at 16:44
  • I have added the output. – Borniet Nov 27 '14 at 12:24
  • output for third disk is missing? Basically it's a two disk failure, if you have logs, check what happened at sda1's Update Time:. Might be a case for --assemble --force but corruption is possible. – frostschutz Nov 27 '14 at 19:37
  • It appeared one of the port for the HD was broken. I added another pci card with ports, and connected the drive to that port. Now it's recognized again. Yet, the raid still doesn't come up. Your help is greatly appreciated!!! – Borniet Dec 5 '14 at 19:45
  • according to that data the disk was missing since may 2012 so the data on it should be wholly useless. You should find out why sda1 failed, ddrescue it if necessary, and then see if you can assemble force with sda1 and sdb1 – frostschutz Dec 5 '14 at 20:07
3

I assume you did a fresh install and added the drives to the new computer.

First do a:

 mdadm --assemble --scan

and look in /proc/mdstat if the array has been activated. Then run:

mdadm --examine --scan >> /etc/mdadm/mdadm.conf

to get the config information in your new mdadm.conf again.

  • I'm sorry, been out with the flu... Yes, it is a new install now, but mdadm --assemble --scan says it can't start the device, because it's assembled from only 1 drive. – Borniet Nov 25 '14 at 10:23
  • @Borniet can you include the output of the mdadm and cat commands in your Q at the end of the Q as frostschutz asked? – Anthon Nov 25 '14 at 10:51
0

It's clean, that's good.

It's inactive, that's not bad.

Take a look e.g. here

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.