Adding new disk on Amazon EC2 instance Centos/RHEL 6

Amazon VPS instances, by default only have a single disk for the whole filesystem of the server. Unless you chose to add an additional disk when you provision the server, adding disk space to your VPS can be tricky. Here’s a run down on how to add more storage to your server on amazon by mounting a new drive.

In this tutorial, I’ve mounted the new drive to /var/lib/mysql since I need to increase the disk space used by the database. This tutorial is still applicable even if you want to increase a different folder. The process will be the same.

**NOTE: Do the steps here in sequence. Do not proceed to next steps if you get any error. Any errors when messing with hard drives can cause you to lose data. I suggest you make a snapshot of you server first before you make any changes.

1. Perform pre-check on the server

Login to the server via ssh
Check the current disks and partition layout using the below command. Don’t worry if your system has different layout. You just need to make check the disk now so you can identify the new disk that we will add later.

[root@goddard ~]# parted -l && lsblk  Model: Xen Virtual Block Device (xvd)  Disk /dev/xvda: 32.2GB          --> this is the only hard drive on the system  Sector size (logical/physical): 512B/512B  Partition Table: gpt  Disk Flags: Number Start End Size File system Name Flags  1 1049kB 2097kB 1049kB bios_grub  2 2097kB 32.2GB 32.2GB xfs NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT  xvda 202:0 0 30G 0 disk          --> the current hard drive has two partitions  ├─xvda1 202:1 0 1M 0 part  └─xvda2 202:2 0 30G 0 part / [root@goddard ~]# df –PTh         --> this is the current disk usage  Filesystem Type Size Used Avail Use% Mounted on  /dev/xvda2 xfs 30G 5.6G 25G 19% /  devtmpfs devtmpfs 477M 0 477M 0% /dev  tmpfs tmpfs 496M 0 496M 0% /dev/shm  tmpfs tmpfs 496M 13M 484M 3% /run  tmpfs tmpfs 496M 0 496M 0% /sys/fs/cgroup  tmpfs tmpfs 100M 0 100M 0% /run/user/1000

2. Create a disk EC2 web console

Volume Type: General Purpose SSD
Size: the size of the new disk.
Availability Zone: the same availability zone of the Server you want to attach it
Snapshot ID: (blank/none)
Encryption: Do not check

Click Create and wait for it until its state is “Available”. Take note of the Volume ID of the newly created disk.

3. Attach the New Disk to the Server

Before you attach the disk, go to > Instances on the amazon web console select the server instance and click Actions > Instance State > Stop to shut down the server.
Go to Elastic Block Store > Volumes again
Select the newly created disk and click Actions > Attach Volume.

On the dialog box:

Instance: input the Instance ID of the server in which you’ll attach this volume
Device: Leave this as the default

If the volume is successfully attached, you see the Attachment Information column filled with the Instance ID which uses this volume.

Now go to > Instances and select the server instance

You should see on the Description tab that a new block device is attached to it.

4. Create a Partition on the New Disk

We need to partition and format the disk before we can use it.
Login to the server via ssh.
Re-check the drives to identify the newly added disk

[root@goddard ~]# parted -l; lsblk  Model: Xen Virtual Block Device (xvd)  Disk /dev/xvda: 32.2GB  Sector size (logical/physical): 512B/512B  Partition Table: gpt  Disk Flags: Number Start End Size File system Name Flags  1 1049kB 2097kB 1049kB bios_grub  2 2097kB 32.2GB 32.2GB xfs Error: /dev/xvdf: unrecognised disk label  Model: Xen Virtual Block Device (xvd)  Disk /dev/xvdf: 10.7GB              --> this is the new disk  Sector size (logical/physical): 512B/512B  Partition Table: unknown  Disk Flags: NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT  xvda 202:0 0 30G 0 disk  ├─xvda1 202:1 0 1M 0 part  └─xvda2 202:2 0 30G 0 part /  xvdf 202:80 0 10G 0 disk             --> the new disk has no existing partition

Now we’ll need to create a partition on the new disk. Use fdisk utility as shown below:
** BE SURE TO USE THE CORRECT DISK OR DATA WILL BE LOST

 [root@goddard /]# fdisk /dev/xvdf        -->  new drive to be used by fdisk
 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 0xc0060bd5.

Command (m for help): p           --> p to print the current partitions of the disk

Disk /dev/xvdf: 10.7 GB, 10737418240 bytes, 20971520 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: 0xc0060bd5

Device Boot Start End Blocks Id System

Command (m for help): n        --> n to create a new partition
 Partition type:
 p primary (0 primary, 0 extended, 4 free)
 e extended
 Select (default p): p           -->  p for primary partition
 Partition number (1-4, default 1):       -->  default (just press Enter)
 First sector (2048-20971519, default 2048):     -->  default (just press Enter)
 Using default value 2048
 Last sector, +sectors or +size{K,M,G} (2048-20971519, default 20971519):   --> default (Enter)
 Using default value 20971519
 Partition 1 of type Linux and of size 10 GiB is set

Command (m for help): p        --> print again to see the new created partition

Disk /dev/xvdf: 10.7 GB, 10737418240 bytes, 20971520 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: 0xc0060bd5

Device Boot Start End Blocks Id System
 /dev/xvdf1 2048 20971519 10484736 83 Linux          --> NEW partition created

**IF you made a mistake on the above points, press q, then Enter to quit without saving. Then start fdisk again. If all is correct, then continue below:

Command (m for help): w       --> save and exit the changes you made
 The partition table has been altered!

Calling ioctl() to re-read partition table.
 Syncing disks.

Verify the new partition created:

 [root@goddard /]# lsblk
 NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
 xvda 202:0 0 30G 0 disk
 ├─xvda1 202:1 0 1M 0 part
 └─xvda2 202:2 0 30G 0 part /
 xvdf 202:80 0 10G 0 disk
 └─xvdf1 202:81 0 10G 0 part      --> This is the new partition

5. Format the New Partition as ext4

Now we’ll need to format the partition to ext4 (default for Centos 6).

[root@goddard /]# mkfs.ext4 /dev/xvdf1       -->  Be sure to input the correct partition
 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
 655360 inodes, 2621184 blocks
 131059 blocks (5.00%) reserved for the super user
 First data block=0
 Maximum filesystem blocks=2151677952
 80 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

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

6. Create a Temporary Mount for the New Partition

Now we can mount the drive and start the copying of data.
Create a temporary folder:
[root@goddard /]# mkdir /mysqlnew

Change the ownership of the folder to mysql:
[root@goddard /]# chown -R mysql.mysql /mysqlnew

Mount the new partition:
[root@goddard /]# mount /dev/xvdf1 /mysqlnew

Verify it is mounted:

 [root@goddard /]# df -PTh
 Filesystem Type Size Used Avail Use% Mounted on
 /dev/xvda2 xfs 30G 5.6G 25G 19% /
 devtmpfs devtmpfs 477M 0 477M 0% /dev
 tmpfs tmpfs 496M 0 496M 0% /dev/shm
 tmpfs tmpfs 496M 13M 484M 3% /run
 tmpfs tmpfs 496M 0 496M 0% /sys/fs/cgroup
 tmpfs tmpfs 100M 0 100M 0% /run/user/1000
 /dev/xvdf1 ext4 9.8G 37M 9.2G 1% /mysqlnew   --> This is the temporary mount

7. Copy the Database data to the Temporary Folder

Stop the myslqd service first:
[root@goddard /]# service mysqld stop

Verify mysqld is stopped successfully:
[root@goddard /]# service mysqld status

Start the copying of database data. This will take a long time depeding on how big you data is.

[root@goddard /]# cp -pav /var/lib/mysql/* /mysqlnew/
 ‘/var/lib/mysql/aria_log.00000001’ -> ‘/mysqlnew/aria_log.00000001’
 ‘/var/lib/mysql/aria_log_control’ -> ‘/mysqlnew/aria_log_control’
 ‘/var/lib/mysql/ibdata1’ -> ‘/mysqlnew/ibdata1’
 ‘/var/lib/mysql/ib_logfile0’ -> ‘/mysqlnew/ib_logfile0’
 ‘/var/lib/mysql/ib_logfile1’ -> ‘/mysqlnew/ib_logfile1’
 ‘/var/lib/mysql/mysql’ -> ‘/mysqlnew/mysql’
 ‘/var/lib/mysql/mysql/db.frm’ -> ‘/mysqlnew/mysql/db.frm’
 .
 .
[output truncated...]

Verify the temporary folder has the files. The contents should be the same as the original

[root@goddard /]# ls -l /mysqlnew
 total 28720
 -rw-rw---- 1 mysql mysql 16384 Apr 30 03:08 aria_log.00000001
 -rw-rw---- 1 mysql mysql 52 Apr 30 03:08 aria_log_control
 -rw-rw---- 1 mysql mysql 18874368 Apr 30 03:08 ibdata1
 -rw-rw---- 1 mysql mysql 5242880 Apr 30 03:08 ib_logfile0
 -rw-rw---- 1 mysql mysql 5242880 Apr 30 03:07 ib_logfile1
 drwx------ 2 root root 16384 Apr 30 04:27 lost+found
 drwx------ 2 mysql mysql 4096 Apr 30 03:07 mysql
 drwx------ 2 mysql mysql 4096 Apr 30 03:07 performance_schema
 drwx------ 2 mysql mysql 4096 Apr 30 03:07 test

This is the original folder for comparison:

[root@goddard /]# ls -l /var/lib/mysql/
 total 28700
 -rw-rw---- 1 mysql mysql 16384 Apr 30 03:08 aria_log.00000001
 -rw-rw---- 1 mysql mysql 52 Apr 30 03:08 aria_log_control
 -rw-rw---- 1 mysql mysql 18874368 Apr 30 03:08 ibdata1
 -rw-rw---- 1 mysql mysql 5242880 Apr 30 03:08 ib_logfile0
 -rw-rw---- 1 mysql mysql 5242880 Apr 30 03:07 ib_logfile1
 drwx------ 2 mysql mysql 4096 Apr 30 03:07 mysql
 drwx------ 2 mysql mysql 4096 Apr 30 03:07 performance_schema
 drwx------ 2 mysql mysql 6 Apr 30 03:07 test

Compare the size of the copied files to the original files if they are the same:

[root@goddard /]# du -sh /var/lib/mysql /mysqlnew
 30M /var/lib/mysql
 30M /mysqlnew

You can now umount the temporary folder:
[root@goddard /]# umount /mysqlnew

Verify it is unmounted:

[root@goddard /]# df -PTh      -->  /mysqlnew should not be displayed anymore
 Filesystem Type Size Used Avail Use% Mounted on
 /dev/xvda2 xfs 30G 5.6G 25G 19% /
 devtmpfs devtmpfs 477M 0 477M 0% /dev
 tmpfs tmpfs 496M 0 496M 0% /dev/shm
 tmpfs tmpfs 496M 13M 484M 3% /run
 tmpfs tmpfs 496M 0 496M 0% /sys/fs/cgroup
 tmpfs tmpfs 100M 0 100M 0% /run/user/1000

8. Create a Permanent Mount of the New Drive

Now that we have a new copy we can move the old /var/lib/mysql:
[root@goddard /]# mv /var/lib/mysql /var/lib/mysql.bak

Create the new folder and set the ownership:
[root@goddard /]# mkdir /var/lib/mysql
[root@goddard /]# chown -R mysql.mysql /var/lib/mysql

Next well need the UUID of the new Drive:

[root@goddard /]# blkid /dev/xvdf1 /dev/xvdf1: UUID="42e73888-c3fb-41c6-916d-85d3e4d50f32" TYPE="ext4"    -->  copy the UUID

Edit the fstab:
[root@goddard /]# nano /etc/fstab

Add the below line with the correct UUID of the drive: (this is a single only)
UUID=42e73888-c3fb-41c6-916d-85d3e4d50f32 /var/lib/mysql ext4 defaults 0 0

Save and exit the /etc/fstab file.

Mount the new disk on /var/lib/mysql:
[root@goddard /]# mount /var/lib/mysql

Change the ownership again just to make sure:
[root@goddard /]# chown -R mysql.mysql /var/lib/mysql

Verify you mounted /var/lib/mysql successfully:

[root@goddard /]# df -PTh
 Filesystem Type Size Used Avail Use% Mounted on
 /dev/xvda2 xfs 30G 5.6G 25G 19% /
 devtmpfs devtmpfs 477M 0 477M 0% /dev
 tmpfs tmpfs 496M 0 496M 0% /dev/shm
 tmpfs tmpfs 496M 13M 484M 3% /run
 tmpfs tmpfs 496M 0 496M 0% /sys/fs/cgroup
 tmpfs tmpfs 100M 0 100M 0% /run/user/1000
 /dev/xvdf1 ext4 9.8G 66M 9.2G 1% /var/lib/mysql    --> this is the new mysql disk

Verify the files are in there:

[root@goddard /]# ls -la /var/lib/mysql
 total 28728
 drwxr-xr-x 6 mysql mysql 4096 Apr 30 04:30 .
 drwxr-xr-x. 37 root root 4096 Apr 30 04:36 ..
 -rw-rw---- 1 mysql mysql 16384 Apr 30 03:08 aria_log.00000001
 -rw-rw---- 1 mysql mysql 52 Apr 30 03:08 aria_log_control
 -rw-rw---- 1 mysql mysql 18874368 Apr 30 03:08 ibdata1
 -rw-rw---- 1 mysql mysql 5242880 Apr 30 03:08 ib_logfile0
 -rw-rw---- 1 mysql mysql 5242880 Apr 30 03:07 ib_logfile1
 drwx------ 2 mysql mysql 16384 Apr 30 04:27 lost+found
 drwx------ 2 mysql mysql 4096 Apr 30 03:07 mysql
 drwx------ 2 mysql mysql 4096 Apr 30 03:07 performance_schema
 drwx------ 2 mysql mysql 4096 Apr 30 03:07 test

9. Start the mysqld Service and Test

After mounting the drive, start the mysqld service.
[root@goddard /]# service mysqld start
[root@goddard /]# service mysqld status

Now test the application and query the database if everything has been imported successfully.

10. Post-checks

To completely test the new drive is mounted during startup and no errors found, I suggest you REBOOT the server from the Amazon web console.

Once data is verified, we can now delete the old backup data to reclaim the space for the OS.
** BE SURE THAT YOU DELETE THE BACKUP FOLDER ONLY!!!
[root@goddard /]# rm -rf /var/lib/mysql.bak

———————————————

masterkenneth