Raid 5 array won't assemble after 1 drive's partition table was wiped

nezumi

11/1/23, 10:00 AM

This is my first post/question here, so please bear with me.

I have an issue with my system where my existing raid5 array won't assemble. This occurred after I inadvertently wiped the partition table of the first disk in the array.

The system was set up a while ago with two raid arrays across 4 identical 4Tb disks - a 2Tb effective/4Tb disk space raid 1 array (md0), intended for home backups, and an ~8Tb effective/12Tb disk space raid 5 array (md1) for media storage.

I was trying to extract all the data from the system to wipe the arrays and start again, so I had removed all data from the Raid 1 array and deleted the array in webmin. I then went into partition manager in webmin intending to change the 1Tb partition on the first drive from part of the now-gone md0 to a usable drive. Unfortunately, doing so wiped the partition table for the whole disk.

So now I am left with a scenario where md1 is showing as inactive in webmin, and I don't know what the best way forward is. Can I "recreate" the raid partitions on sda and have it add back to md1? or can I force md1 to assemble from the remaining 3 drives, and continue the existing data transfer process? the output of

sudo mdadm -D /dev/md1

/dev/md1:
       Version : 1.2
    Raid Level : raid5
 Total Devices : 3
   Persistence : Superblock is persistent

         State : inactive
Working Devices : 3

          Name : miranda:1  (local to host miranda)
          UUID : 3d6bf0c4:16037750:01681844:95415c3d
        Events : 2086

Number   Major   Minor   RaidDevice

   -       8       50        -        /dev/sdd2
   -       8       34        -        /dev/sdc2
   -       8       18        -        /dev/sdb2

Using lsblk

gives

sda       8:0    0   3.6T  0 disk
sdb       8:16   0   3.6T  0 disk
├─sdb1    8:17   0 931.5G  0 part
└─sdb2    8:18   0   2.7T  0 part
  └─md1   9:1    0     0B  0 md
sdc       8:32   0   3.6T  0 disk
├─sdc1    8:33   0 931.5G  0 part
└─sdc2    8:34   0   2.7T  0 part
  └─md1   9:1    0     0B  0 md
sdd       8:48   0   3.6T  0 disk
├─sdd1    8:49   0 931.5G  0 part
└─sdd2    8:50   0   2.7T  0 part
  └─md1   9:1    0     0B  0 md

The data on these drives isn't critical, but it would be preferable to be able to recover/access them than not.

1 + 2

raid5

partition-table

Aleksandr Makhov

11/1/23, 10:45 AM

Hi, can you add results of "cat /proc/mdstat" and also what happen if you run mdam --start /dev/md1 ? You probably just play around with the mdraid, but for real systems it's not a good idea to use one physical disk in two arrays, because it reduces raids redundancy and can be a reason of performance issues if your configuration is based on HDDs.

nezumi

11/2/23, 2:08 AM

@AleksandrMakhov So ```cat /proc/mdstat``` gives the following ```Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] md1 : inactive sdd2[3](S) sdc2[2](S) sdb2[1](S) 8790398976 blocks super 1.2``` using ```mdadm --start /dev/md1``` returns "unrecognized option '--start' "

Score:0

Server

nezumi

11/11/23, 12:37 AM

So in the end I was able to get the array to work by using the command

mdadm --run /dev/md1

Once I confirmed that it was all happy and accessible, I used the partition information from the 3 good disks to "repartition" the disk that I accidentally wiped the table from, then added it back into the array. I must have got the apportionment slightly off, because it had to rebuild, but it did so successfully overnight without any data loss.

+ 0

Elon Musk

I sit in a Tesla and translated this thread with Ai:

EN: Raid 5 array won't assemble after 1 drive's partition table was wiped

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.