Replace a disk in a RAID

Last update: October 05 2026Open in Markdown

1. Remove the disk from the array.
2. Determine the disk partitioning scheme.
3. Copy the partition table to the new disk.
4. Add the disk to the RAID array.
5. Install the bootloader.

1. Remove the disk from the array

1. Connect to the server via SSH or via the KVM console.

2. Output current RAID array configuration:

   cat /proc/mdstat

   The output will contain information about the RAID array configuration. Example:

   Personalities : [raid1]
   md1 : active raid1 sda3[0] sdb3[1]
       975628288 blocks super 1.2 [2/2] [UU]
       bitmap: 3/8 pages [12KB], 65536KB chunk
   
   md0 : active raid1 sda2[2] sdb2[1]
       999872 blocks super 1.2 [2/2] [UU]
   
   unused devices: <none>

   Where:

   • md0 — the name of the RAID array assembled from partitions sda2 and sdb2;
   • md1 — the name of the RAID array assembled from partitions sda3 and sdb3.

3. Output information about partitions on the disks and RAID array configuration:

   lsblk

   The response will contain a list of disks with partitions. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0     1M  0 part
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sda3             8:3    0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /
   sdb                8:16   0 931.5G  0 disk
   ├─sdb1             8:17   0     1M  0 part
   ├─sdb2             8:18   0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sdb3             8:19   0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /

   Where:

   • sda — the name of the disk with working RAID arrays;
   • sdb — the name of the disk to be replaced;
   • sdb2 — the name of the partition on disk sdb, which is a component of the first RAID array;
   • md0 — the name of the first RAID array assembled from partitions sda2 and sdb2;
   • sdb3 — the name of the partition on disk sdb, which is a component of the second RAID array;
   • md1 — the name of the second RAID array assembled from partitions sda3 and sdb3.

4. Mark the disk you need to replace as faulty:

   mdadm /dev/<raid_array_1> -f /dev/<partition_1>
   mdadm /dev/<raid_array_2> -f /dev/<partition_2>

   Specify:

   • <raid_array_1> — the name of the first RAID array that you obtained in step 3, in the example it is md0;
   • <partition_1> — the name of the first RAID array partition on the disk to be replaced that you obtained in step 3, in the example it is sdb2;
   • <raid_array_2> — the name of the second RAID array that you obtained in step 3, in the example it is md1;
   • <partition_2> — the name of the second RAID array partition on the disk to be replaced that you obtained in step 3, in the example it is sdb3.

5. Remove the disk you marked as faulty in step 4 from all devices:

   mdadm /dev/<raid_array_1> --remove /dev/<partition_1>
   mdadm /dev/<raid_array_2> --remove /dev/<partition_2>

   Specify:

   • <raid_array_1> — the name of the first RAID array that you obtained in step 3, in the example it is md0;
   • <partition_1> — the name of the first RAID array partition on the disk to be replaced that you obtained in step 3, in the example it is sdb2;
   • <raid_array_2> — the name of the second RAID array that you obtained in step 3, in the example it is md1;
   • <partition_2> — the name of the second RAID array partition on the disk to be replaced that you obtained in step 3, in the example it is sdb3.

2. Determine the disk partition scheme

1. Connect to the server via SSH or via the KVM console.

2. Install the utility for working with the GPT (GUID Partition Table) partition table on disks — gdisk:

   apt-get install gdisk -y

3. Output information about partitions on the disks:

   lsblk

   The response will contain information about partitions on the disks. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0     1M  0 part
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sda3             8:3    0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /
   sdb                8:16   0 931.5G  0 disk

   Where:

   • sda — the name of the disk with working RAID partitions;
   • sdb — the name of the new disk.

4. Determine the disk partition scheme:

   gdisk -l /dev/<source_disk>

   Specify <source_disk> — the name of the disk with working RAID partitions that you obtained in step 3, in the example it is sda.

   The answer depends on the disk partition scheme:

   • GPT:

     Partition table scan:
     MBR: protective
     BSD: not present
     APM: not present
     GPT: present

   • MBR:

     Partition table scan:
     MBR: MBR only
     BSD: not present
     APM: not present
     GPT: not present

3. Copy the partition table to the new disk

The process of copying the partition table depends on the disk partitioning scheme — GPT or MBR.

carefully

When copying the layout, it is important to specify the disks in the correct order in the command. If the disk order is specified incorrectly, the healthy disk will be overwritten, and data will be deleted. Data cannot be recovered.

GPT

1. Connect to the server via SSH or via the KVM console.

2. Output information about partitions on the disks:

   lsblk

   The response will contain information about partitions on the disks. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0     1M  0 part
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sda3             8:3    0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /
   sdb                8:16   0 931.5G  0 disk

   Where:

   • sda — the name of the disk with working RAID partitions;
   • sdb — the name of the new disk.

3. Copy the layout to the new disk. It is important to specify the disks in the correct order in the command; otherwise, the layout and data on the initially healthy disk will be destroyed.

   sgdisk -R /dev/<new_disk> /dev/<source_disk>

   Specify:

   • <new_disk> — the name of the new disk to which the partition table is copied, you obtained it in step 2, in the example it is sdb;
   • <source_disk> — the name of the disk with working RAID partitions from which the partition table is copied, you obtained it in step 2, in the example it is sda.

4. Assign a random UUID to the new disk:

   sgdisk -G /dev/<new_disk>

   Specify <new_disk> — the name of the new disk to which the partition table was copied, you obtained it in step 2, in the example it is sdb.

5. Ensure the partition schemes on the disks are identical:

   lsblk

   The response will contain information about partitions on the disks. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0     1M  0 part
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sda3             8:3    0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /
   sdb                8:16   0 931.5G  0 disk
   ├─sdb1             8:17   0     1M  0 part
   ├─sdb2             8:18   0   977M  0 part
   └─sdb3             8:19   0 930.6G  0 part

MBR

1. Connect to the server via SSH or via the KVM console.

2. Output information about partitions on the disks:

   lsblk

   The response will contain information about partitions on the disks. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0     1M  0 part
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sda3             8:3    0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /
   sdb                8:16   0 931.5G  0 disk

   Where:

   • sda — the name of the disk with working RAID partitions;
   • sdb — the name of the new disk.

3. Copy the layout to the new disk. It is important to specify the disks in the correct order in the command; otherwise, the layout and data on the initially healthy disk will be destroyed.

   sfdisk -d /dev/<source_disk> | sfdisk /dev/<new_disk>

   Specify:

   • <source_disk> — the name of the disk with working RAID partitions from which the partition table is copied, you obtained it in step 2, in the example it is sda;
   • <new_disk> — the name of the new disk to which the partition table is copied, you obtained it in step 2, in the example it is sdb.

4. If partitions are not visible in the system, refresh partition information:

   partprobe /dev/<new_disk>

   Specify <new_disk> — the name of the new disk to which the partition table was copied, you obtained it in step 2, in the example it is sdb.

5. Ensure the partition schemes on the disks are identical:

   lsblk

   The response will contain information about partitions on the disks. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0     1M  0 part
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sda3             8:3    0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /
   sdb                8:16   0 931.5G  0 disk
   ├─sdb1             8:17   0     1M  0 part
   ├─sdb2             8:18   0   977M  0 part
   └─sdb3             8:19   0 930.6G  0 part

4. Add the disk to the RAID array

1. Connect to the server via SSH or via the KVM console.

2. Output information about partitions on the disks and RAID array configuration:

   lsblk

   The response will contain information about partitions on the disks and RAID array configuration. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0     1M  0 part
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sda3             8:3    0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /
   sdb                8:16   0 931.5G  0 disk
   ├─sdb1             8:17   0     1M  0 part
   ├─sdb2             8:18   0   977M  0 part
   └─sdb3             8:19   0 930.6G  0 part

   Where:

   • md0 — the name of the first RAID array assembled from partition sda2;
   • md1 — the name of the second RAID array assembled from partition sda3;
   • sdb2 — the name of the first partition on the new disk to be added to the RAID array;
   • sdb3 — the name of the second partition on the new disk to be added to the RAID array.

3. Add the partitions of the new disk to the RAID array:

   mdadm /dev/<raid_array_1> -a /dev/<partition_1>
   mdadm /dev/<raid_array_2> -a /dev/<partition_2>

   Specify:

   • <raid_array_1> — the name of the first RAID array that you obtained in step 2, in the example it is md0;
   • <partition_1> — the name of the first partition on the new disk that you obtained in step 2, in the example it is sdb2;
   • <raid_array_2> — the name of the second RAID array that you obtained in step 2, in the example it is md1;
   • <partition_2> — the name of the second partition on the new disk that you obtained in step 2, in the example it is sdb3.

   After adding the disk to the RAID array, synchronization will start. The synchronization speed depends on the disk size and type — SSD or HDD.

4. Wait for synchronization to complete.

5. Ensure that the new disk has been added to the RAID array:

   lsblk

   The response will contain information about partitions on the disks and RAID array configuration. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0     1M  0 part
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sda3             8:3    0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /
   sdb                8:16   0 931.5G  0 disk
   ├─sdb1             8:17   0     1M  0 part
   ├─sdb2             8:18   0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1
   └─sdb3             8:19   0 930.6G  0 part
     └─md1            9:1    0 930.4G  0 raid1  /

5. Install the bootloader

After adding the disk to the array, you need to install the operating system bootloader on it.

You can install the bootloader both in the OS and in Rescue mode.

Server booted into the OS

1. Connect to the server via SSH or via the KVM console.

2. Output information about partitions on the disks:

   lsblk -o +FSTYPE

   The response will contain information about partitions on the disks. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT  FSTYPE
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0   500M  0 part              vfat
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1             ext4
   └─sda3             8:3    0 930.1G  0 part
     └─md1            9:1    0 929.9G  0 raid1  / 
   sdb                8:16   0 931.5G  0 disk
   ├─sdb1             8:17   0   500M  0 part          
   ├─sdb2             8:18   0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1             ext4
   └─sdb3             8:19   0 930.1G  0 part
     └─md1            9:1    0 929.9G  0 raid1  /          ext4 

   Where:

   • sda — the name of the disk with working RAID arrays;
   • sdb — the name of the new disk.

3. Install the GRUB bootloader:

   grub-install /dev/<new_disk>

   Specify <new_disk> — the name of the new disk that you obtained in step 2, in the example it is sdb.

Server booted in Rescue mode

1. Boot the server in Rescue recovery and diagnostics mode.

2. Connect to the server via SSH or via the KVM console.

3. Output information about the OS boot mode:

   [ -d /sys/firmware/efi ] && echo "UEFI" || echo "BIOS"

   The output will show information about the OS boot mode — BIOS or UEFI.

4. Check if the system has detected RAID arrays by outputting information about partitions on the disks:

   lsblk -o +FSTYPE

   The response will contain information about partitions on the disks and RAID array configuration. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT  FSTYPE
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0   500M  0 part              vfat
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1             ext4
   └─sda3             8:3    0 930.1G  0 part
     └─md1            9:1    0 929.9G  0 raid1  
   sdb                8:16   0 931.5G  0 disk
   ├─sdb1             8:17   0   500M  0 part          
   ├─sdb2             8:18   0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1             ext4
   └─sdb3             8:19   0 930.1G  0 part
     └─md1            9:1    0 929.9G  0 raid1             ext4 

   Here md0, md1 are the names of the RAID arrays on disk partitions.

5. If there are RAID arrays in the lsblk output in step 4, skip this step. If there are no RAID arrays, assemble the RAID arrays:

   mdadm --assemble --scan

6. Output information about partitions on the disks and RAID array configuration:

   lsblk -o +FSTYPE

   The response will contain information about partitions on the disks and RAID array configuration. For example:

   NAME             MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT  FSTYPE
   loop0              7:0    0   985M  1 loop
   sda                8:0    0 931.5G  0 disk
   ├─sda1             8:1    0   500M  0 part              vfat
   ├─sda2             8:2    0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1             ext4
   └─sda3             8:3    0 930.1G  0 part
     └─md1            9:1    0 929.9G  0 raid1  
   sdb                8:16   0 931.5G  0 disk
   ├─sdb1             8:17   0   500M  0 part          
   ├─sdb2             8:18   0   977M  0 part
   │ └─md0            9:0    0 976.4M  0 raid1             ext4
   └─sdb3             8:19   0 930.1G  0 part
     └─md1            9:1    0 929.9G  0 raid1             ext4 

   Where:

   • sda — the name of the disk with working RAID partitions;
   • sdb — the name of the new disk;
   • md0, md1 are the names of the RAID arrays on disk partitions.

7. Identify the partitions to be mounted:

   • root partition / — usually located on the largest disk partition, in the example in step 6 it is md1;
   • boot partition /boot — usually has the ext4 file system and a size of 512 MB – 1 GB, in the example in step 6 it is md0;
   • EFI partition /boot/efi — a partition with the vfat file system, used when booting the OS in UEFI mode, in the example in step 6 these are partitions sda1 and sdb1.

8. Mount the root file system to the /mnt directory:

   mount /dev/<raid_array> /mnt

   Here <system_raid_array> is the name of the root partition that you determined in step 7, in the example it is md1.

9. Mount the boot partition:

   mount /dev/<boot_partition> /mnt/boot

   Specify <boot_raid_array> — the name of the /boot boot partition that you determined in step 7, in the example it is md0.

10. If the OS uses a UEFI bootloader, mount the EFI partition:

    mount /dev/<efi_partition> /mnt/boot/efi

    Specify <efi_partition> — the name of the EFI partition /boot/efi on the new disk that you determined in step 7, in the example it is sdb1.

11. Mount the service file systems:

    mount --bind /sys /mnt/sys
    mount --bind /proc /mnt/proc
    mount --bind /dev /mnt/dev
    mount -t devpts devpts /mnt/dev/pts

12. If in step 3 you determined the OS boot mode to be UEFI, mount the efivars file system:

mount -t efivarfs efivarfs /mnt/sys/firmware/efi/efivars

13. Connect to the environment:

    chroot /mnt /bin/bash

14. Export the PATH variable:

    export PATH=/bin:/usr/bin:/usr/local/bin:/sbin:/usr/sbin:/usr/local/sbin

15. Install the GRUB bootloader. The installation command depends on the OS boot mode you determined in step 3:

    • UEFI:

    grub-install --target=x86_64-efi --efi-directory=/boot/efi --bootloader-id=GRUB

    • BIOS:

    grub-install /dev/<new_disk>

    Specify <new_disk> — the name of the new disk that you obtained in step 6, in the example it is sdb.

16. Create the GRUB configuration file:

    /sbin/grub-mkconfig -o /boot/grub/grub.cfg

17. After the GRUB installation is finished, exit the environment:

    exit

18. Unmount the service file systems:

    mount -t devpts devpts /mnt/dev/pts
    mount --bind /dev /mnt/dev
    mount --bind /proc /mnt/proc
    mount --bind /sys /mnt/sys

19. Unmount the file system:

    umount /dev/<system_raid_array> /mnt

    Specify <system_raid_array> — the name of the / root partition that you mounted in step 8, in the example it is md1.

20. Restore the previous server boot template or reboot the server from the OS. When rebooting the server from the OS, the boot template will automatically revert to the one set before booting the server in Rescue mode.