Shrink AWS EBS volumes using EC2 ubuntu

---

1. Launch an instance
2. Create a new volume
3. Attach the new 3GB volume
4. Format the new volume
5. Mount the new volume
6. Copy data from the old volume to the new one
7. Using Elastio, back up the data to s3 bucket
8. Mount the volume as a device on your instance
9. Clean up your resources in the AWS Account

1. Launching an instance

• In the AWS dashboard (EC2 service), create an instance of 8GB

2. Creating new volume

• Create Volume of 3G and name it new-volume

3. Attaching new 8GB vol to instance making it the two volumes

• Attach the new 3GB volume to the instance (from 1. above) to make two volumes (8GB and 3GB)

a. The original volume is /dev/sda1 b. The new-vol of 3G to the will be attached as /dev/sdg

4. Format the new volume

a. Login to the EC2 instance via SSH b. List all available volumes: lsblk

c. Run the $ df -hT command: of the 8GB EBS volume space, only 2.2 GB is used and the rest is free space. Next, the free space will be eliminated.

d. Reminder: the original volume is at /dev/xvda1 and the new volume at**/dev/xvdg.** Create a partition on /dev/xvdg using the $ fdisk command and format the partition.

• The new partition is xvdg1

e. To format, check if the volume has any data or not by the following command: sudo file -s /dev/xvdg1

If the output is /dev/xvdg1: data, it means the volume is empty and it can be formatted.

f. Format the volume with the following command: sudo mkfs.ext4 /dev/xvdg1

5. Mounting the new volume

Mount the new volume to the EC2 instance.

a. Create a directory to mount the new volume: $ sudo mkdir /new-volume

b. Mount the new volume into this directory: $ sudo mount /dev/xvdg1 /new-volome

c. Verify with: $ df -hT

6. Copying data from the old volume to the new one

Copy the data from the original volume to the new volume

sudo rsync -axv / /dev/root /new-volume

7. Using Elastio, back up the data to s3 bucket (very fast process

• Using Elastio, back up the data to s3 bucket (very fast process

8. Mount the volume as a device on your instance

sudo modprobe nbd sudo -E elastio mount rp --id r-dged6mmwk3sq5ssfpwlweqol

Note: r-... is the recovery point

9. The same process can be followed for multiple partitions on a volume

10. Clean up your resources in the AWS Account.

NB: You can use cloud-nuke to cleean up your aws resources.