Initializing a Linux Data Disk (Less Than or Equal to 2 TiB)

Scenarios

After a newly purchased data disk is attached to a server, you must log in to the server and initialize the disk before you can use the disk. This section describes how to initialize a Linux data disk. The operations may vary depending on the server OS.

The maximum disk size that MBR supports is 2 TiB, and that GPT supports is 18 EiB. If your disk capacity is less than 2 TiB, you can use either the MBR or GPT partition style. However, if your disk is greater than 2 TiB or you may expand it to over 2 TiB later, use GPT when initializing disks.

**Table 1** Initialization instructions
Operation	OS Requirements	Partition Style	Common File Systems	Example Configuration
Initializing a Data Disk Using fdisk	None	MBR	ext* (such as ext2, ext3, and ext4), xfs, and btrfs	OS: CentOS 7.4 64-bit Partitioning tool: fdisk Device name: /dev/vdb File system format: ext4 Mount points: /mnt/sdc and /mnt/sdd Partition 1: /dev/vdb1 Size: 40 GiB Partition style: MBR Partition 2: /dev/vdb2 Size: 60 GiB Partition style: MBR
Initializing a Data Disk Using parted	None	GPT MBR	ext* (such as ext2, ext3, and ext4), xfs, and btrfs	See Initializing a Data Disk Using parted.

Prerequisites

You have attached the disk to a server.

Constraints

A disk created from a data source does not need to be initialized. Such a disk contains the source data in the beginning. Initializing the disk may clear the initial data on it. If you need to re-initialize the disk, you are advised to back up the disk data first. For how to back up data using CBR, see Backing Up EVS Disks. For how to back up data using snapshots, see Creating an EVS Snapshot.
Initializing a disk does not delete the snapshots created for the disk, so you can still use snapshots to roll back data to the source disk after the disk is initialized.

The following example shows you how to use fdisk to create two primary partitions (/dev/vdb1: 40 GiB; /dev/vdb2: 60 GiB) on the /dev/vdb data disk and set the partition style to MBR.

Log in to the server.

For how to log in to an ECS, see Logging In to an ECS.

For how to log in to a BMS, see Logging In to a BMS.

Create partitions. In this example, create two primary partitions, /dev/vdb1 and /dev/vdb2 for data disk /dev/vdb.

Check that the capacity of the /dev/vdb data disk is 100 GiB.
lsblk
```
[root@ecs-centos76 ~]# lsblk
NAME   MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
vda    253:0    0   40G  0 disk
└─vda1 253:1    0   40G  0 part /
vdb    253:16   0  100G 0 disk
```
The command output shows that there are two EVS disks. /dev/vda is the system disk, and /dev/vdb is the new data disk, whose 100 GiB is not partitioned.

Create the first primary partition /dev/vdb1 for /dev/vdb.

fdisk /dev/vdb

[root@ecs-test-0001 ~]# fdisk /dev/vdb
Welcome to fdisk (util-linux 2.23.2).

Changes will remain in memory only, until you decide to write them.
Be careful before using the write command.

Device does not contain a recognized partition table
Building a new DOS disklabel with disk identifier 0x38717fc1.

Command (m for help): n
Partition type:
   p   primary (0 primary, 0 extended, 4 free)
   e   extended
Select (default p): p
Partition number (1-4, default 1): 1

Entering p for Partition type creates a primary partition, and entering e creates an extended partition. The default value is p.
Partition number indicates the partition serial number. Enter a value ranging from 1 to 4.

Set First sector to 2048 and Last sector to 83886079 for partition /dev/vdb1 (40 GiB). For details about the calculation method, see Table 2.

First sector (2048-209715199, default 2048): 2048
Last sector, +sectors or +size{K,M,G} (2048-209715199, default 209715199):83886079
Partition 1 of type Linux and of size 40 GB is set

**Table 2** First and last sectors in this example are calculated as follows
Sector	/dev/vdb1 (40 GiB)	/dev/vdb2 (60 GiB)	Formula for Calculating the Value of sectors
First sector	2048 (The first sector of the /dev/vdb data disk is used.)	Last sector of /dev/vdb1 + 1 = 83886079 + 1 = 83886080	Value of sectors = Capacity x 1073741824/512
Last sector	Value of sectors – 1 = (40 x 1073741824/512) – 1 = 83886079	First sector + Value of sectors – 1 = 83886080 + (60 x 1073741824/512) – 1 = 209715199	Value of sectors = Capacity x 1073741824/512

Create the second primary partition /dev/vdb2 for /dev/vdb.

Command (m for help): n
Partition type:
   p   primary (0 primary, 0 extended, 4 free)
   e   extended
Select (default p): p
Partition number (1-4, default 2): 2

Set the First sector to 83886080 and Last sector to 209715199 for partition /dev/vdb2.

First sector (83886080-209715199, default 83886080): 83886080
Last sector, +sectors or +size{K,M,G} (83886080-209715199, default 209715199):209715199
Partition 2 of type Linux and of size 60 GB is set

Check the sizes and partition styles of the new partitions.

Command (m for help): p

Disk /dev/vdb: 107.4 GB, 107374182400 bytes, 209715200 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk label type: dos
Disk identifier: 0x994727e5

   Device Boot       Start         End      Blocks   Id  System
/dev/vdb1             2048    83886079    41942016   83  Linux
/dev/vdb2         83886080   209715199    62914560   83  Linux

Command (m for help):

Disk label type: dos indicates the MBR partition style.

In case that you want to discard the changes made before, you can exit fdisk by entering q and press Enter. Then, re-create the partitions by referring to step 1 and step 2.

Write the changes to the partition table.
w
```
Command (m for help): w
The partition table has been altered!

Calling ioctl() to re-read partition table.
Syncing disks.
```
The partition is created.

If error message -bash: partprobe: command not found is returned, the system cannot identify the command. In this case, run yum install -y parted to install the command. Then, run the command again.
Synchronize the new partition table to the OS.
partprobe

Create file systems.

Create ext4 file systems for partitions /dev/vdb1 (40 GiB) and /dev/vdb2 (60 GiB).

mkfs -t ext4 /dev/vdb1

mkfs -t ext4 /dev/vdb2

mkfs -t <file-system-format> <disk-partition-name>: To create an xfs file system, the command is mkfs -t xfs <disk-partition-name>. To create a btrfs file system, the command is mkfs -t btrfs <disk-partition-name>.
It takes some time to create file systems. Do not exit before the system returns the following information:

[root@ecs-test-0001 ~]# mkfs -t ext4 /dev/vdb1
mke2fs 1.42.9 (28-Dec-2013)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
2621440 inodes, 10485504 blocks
524275 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=2157969408
320 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
        32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
        4096000, 7962624

Allocating group tables: done
Writing inode tables: done
Creating journal (32768 blocks): done
Writing superblocks and filesystem accounting information: done

Check whether the file system format is ext4.

blkid

[root@ecs-test-0001 ~]# blkid /dev/vdb
/dev/vdb1: UUID="0b3040e2-1367-4abb-841d-ddb0b92693df" TYPE="ext4"
/dev/vdb2: UUID="0d6769k2-1745-9dsf-453d-hgd0b34267dj" TYPE="ext4"

Create directories (mount points) and mount the new partitions on the created mount points.
1. Mount /dev/vdb1 on /mnt/sdc.
  mkdir -p /mnt/sdc
  
  mount /dev/vdb1 /mnt/sdc
1. Mount /dev/vdb2 on /mnt/sdd.
  mkdir -p /mnt/sdd
  
  mount /dev/vdb2 /mnt/sdd
2. View the mount results.
  lsblk
```
[root@ecs-test-0001 ~]# lsblk
NAME   MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
vda    253:0    0  40G  0 disk
├vda1 253:1    0  40G  0 part /
vdb    253:16   0 100G  0 disk
├vdb1 253:17   0  40G  0 part /mnt/sdc
├vdb2 253:18   0  60G  0 part /mnt/sdd
```
You should now see that partitions /dev/vdb1 and /dev/vdb2 are mounted on /mnt/sdc and /mnt/sdd.

(Optional) Use the partition UUIDs to configure auto mount at startup.

UUIDs are the unique character strings for identifying partitions in Linux. Mounts become invalid after a system reboot. You can configure auto mount at startup by adding information of the new partitions into the /etc/fstab file.

You are advised not to use device names to identify disks in the /etc/fstab file because device names are assigned dynamically and may change (for example, from /dev/vdb1 to /dev/vdb2) after a server stop or start. This can even prevent your server from booting up.
This operation will not affect the existing data on the ECS.

Query the partition UUIDs.
blkid /dev/vdb1

blkid /dev/vdb2
```
[root@ecs-test-0001 ~]# blkid /dev/vdb
/dev/vdb1: UUID="0b3040e2-1367-4abb-841d-ddb0b92693df" TYPE="ext4"
/dev/vdb2: UUID="0d6769k2-1745-9dsf-453d-hgd0b34267dj" TYPE="ext4"
```
Take note of the partition UUIDs, which will be used in the next step. In this example, the partition UUIDs are 0b3040e2-1367-4abb-841d-ddb0b92693df and 0d6769k2-1745-9dsf-453d-hgd0b34267dj.

Configure auto mount at startup.

vi /etc/fstab

Press i to enter the editing mode, move the cursor to the end of the file, press Enter, and add the following content:

UUID=0b3040e2-1367-4abb-841d-ddb0b92693df    /mnt/sdc     ext4    defaults     0    2
UUID=0d6769k2-1745-9dsf-453d-hgd0b34267dj    /mnt/sdd     ext4    defaults     0    2

Press Esc, enter :wq, and press Enter to save the settings and exit the vi editor.

**Table 3** Content description
Example Value	Description
UUID=0b3040e2-1367-4abb-841d-ddb0b92693df	The UUID of the partition.
/mnt/sdc	The mount point of the partition.
ext4	The file system format of the partition.
defaults	The partition mount option. Normally, this parameter is set to defaults.
0	The Linux dump backup option. 0: Linux dump backup is not used. Usually, dump backup is not used, and you can set this parameter to 0. 1: Linux dump backup is used.
2	The fsck option, which means whether to use fsck to check the disk during startup. 2: The check starts from the partitions whose mount points are non-root directories. / is the root directory. 1: The check starts from the partitions whose mount points are root directories. 0: The fsck option is not used.

(Optional) Verify that auto mount takes effect.

You can restart the server to check whether auto mount takes effect. Alternatively, you can perform the following steps to simulate auto mount.
1. To verify auto mount, unmount the partitions first.
umount /dev/vdb1

umount /dev/vdb2
1. Reload all the content in the /etc/fstab file. /etc/fstab is a static file system table that contains the list of file systems that need to be automatically mounted during system startup.
mount -a
1. Query the file system mount information.
mount | grep /mnt/sdc

mount | grep /mnt/sdd

If information similar to the following is displayed, auto mount has taken effect:
```
root@ecs-test-0001 ~]# mount | grep /mnt/sdc
/dev/vdb1 on /mnt/sdc type ext4 (rw,relatime,data=ordered)
root@ecs-test-0001 ~]# mount | grep /mnt/sdd
/dev/vdb2 on /mnt/sdd type ext4 (rw,relatime,data=ordered)
```

You can use parted to create either MBR or GPT partitions. The only difference is that the command used to set the partition style is different. All other operations are the same. For the initialization instructions with parted, see Initializing a Data Disk Using parted.

Commands for setting partition styles:

MBR: