Tutorial - Manage Azure disks with the Azure CLI

Applies to: ✔️ Linux VMs ✔️ Flexible scale sets

Azure virtual machines (VMs) use disks to store the operating system, applications, and data. When you create a VM, it is important to choose a disk size and configuration appropriate to the expected workload. This tutorial shows you how to deploy and manage VM disks. You learn about:

Default Azure disks

When an Azure virtual machine is created, two disks are automatically attached to the virtual machine.

Operating system disk - The operating system (OS) disk hosts the VM operating system and can be up to 4,095 gibibytes (GiB), although many operating systems are partitioned with master boot records (MBRs) by default. An MBR limits the usable size to 2 TiB. If you need more than 2 TiB, create and attach data disks and use them for data storage. Device names vary by VM generation and disk controller type, so don't rely on a fixed path such as /dev/sda. OS disk caching is optimized for OS performance, so don't store application data on the OS disk. Use data disks for application and data storage.

Temporary disk - Temporary storage uses local storage on the same Azure host as the VM. Depending on the VM size and generation, this storage might appear as a temporary disk or local NVMe disk. It's high-performance, but nonpersistent. If the VM is moved to a new host, data on temporary storage is removed. The available size depends on the VM size.

Data disks

To install applications and store data, additional data disks can be added. Data disks should be used in any situation where durable and responsive data storage is desired. The size of the virtual machine determines how many data disks can be attached to a VM.

VM disk types

Azure managed disks offer five disk types:

• Ultra Disks
• Premium SSD v2
• Premium SSDs
• Standard SSDs
• Standard HDDs

In general, use Premium SSD v2 or Premium SSD for production workloads, Standard SSD for dev/test and less performance-sensitive workloads, and Ultra Disks for IO-intensive data workloads. Ultra Disks and Premium SSD v2 are data-disk only options and can't be used as OS disks. For detailed limits, region availability, and sizing guidance, see Azure managed disk types.

Launch Azure CLI

You can use Azure local Shell on your local PC to run the steps in this article.

Create and attach disks

Data disks can be created and attached at VM creation time or to an existing VM.

Attach disk at VM creation

Create a resource group with the az group create command.

az group create --name myResourceGroupDisk --location chinaeast2

Create a VM by using the az vm create command. The following example creates a VM named myVM, adds a user account named azureuser, and generates SSH keys if they don't already exist. The --data-disk-sizes-gb argument specifies extra data disks to create and attach. To create and attach more than one disk, use a space-delimited list of disk sizes. In the following example, you create a VM with two data disks, both 128 GiB. Because the disks are 128 GiB, you configure both as P10 disks, which provide a maximum of 500 IOPS per disk.

az vm create \
  --resource-group myResourceGroupDisk \
  --name myVM \
  --image Ubuntu2204 \
  --size Standard_DS2_v2 \
  --admin-username azureuser \
  --generate-ssh-keys \
  --data-disk-sizes-gb 128 128

Attach disk to existing VM

To create and attach a new disk to an existing virtual machine, use the az vm disk attach command. The following example creates a 128-GiB premium disk and attaches it to the VM created in the last step.

az vm disk attach \
    --resource-group myResourceGroupDisk \
    --vm-name myVM \
    --name myDataDisk \
    --size-gb 128 \
    --sku Premium_LRS \
    --new

Prepare a data disk in Linux

After you attach a disk to the virtual machine, configure the operating system to use the disk. The following examples manually configure the first 128-GiB data disk that was attached when you created the VM. The VM's disk controller determines how Linux exposes the disk. You can also automate this process by using cloud-init, which is covered in a later tutorial.

Create an SSH connection with the virtual machine. Replace the example IP address with the public IP of the virtual machine.

ssh azureuser@10.101.10.10

Verify that you identified the correct data disk before you format it. Formatting the wrong disk results in data loss.

lsblk -o NAME,HCTL,SIZE,MOUNTPOINT | grep -i "sd"

sudo parted /dev/sdc --script mklabel gpt mkpart xfspart xfs 0% 100%
sudo partprobe /dev/sdc

sudo mkfs.xfs /dev/sdc1

sudo mkdir /datadrive && sudo mount /dev/sdc1 /datadrive

df -h | grep -i "sd"

Filesystem      Size  Used Avail Use% Mounted on
/dev/sda1        29G  2.0G   27G   7% /
/dev/sda15      105M  3.6M  101M   4% /boot/efi
/dev/sdb1        14G   41M   13G   1% /mnt
/dev/sdc1       128G   52M  128G   1% /datadrive

sudo azure-nvme-id

sudo parted /dev/nvme0n2 --script mklabel gpt mkpart xfspart xfs 0% 100%
sudo partprobe /dev/nvme0n2

sudo mkfs.xfs /dev/nvme0n2p1

sudo mkdir /datadrive && sudo mount /dev/nvme0n2p1 /datadrive

df -h | grep -i "nvme"

To ensure that the drive is remounted after a reboot, it must be added to the /etc/fstab file. To do so, get the UUID of the disk with the blkid utility.

sudo -i blkid

The output displays the UUID of the drive. On a VM with a SCSI controller, the partition might appear as /dev/sdc1. On a VM with an NVMe controller, it might appear as /dev/nvme0n2p1.

/dev/sdc1: UUID="33333333-3b3b-3c3c-3d3d-3e3e3e3e3e3e" TYPE="xfs"

Open the /etc/fstab file in a text editor as follows:

sudo nano /etc/fstab

Add a line similar to the following to the /etc/fstab file, replacing the UUID value with your own.

UUID=33333333-3b3b-3c3c-3d3d-3e3e3e3e3e3e   /datadrive  xfs    defaults,nofail   1  2

When you are done editing the file, use Ctrl+O to write the file and Ctrl+X to exit the editor.

Now that the disk has been configured, close the SSH session.

exit

Take a disk snapshot

When you take a disk snapshot, Azure creates a read-only, point-in-time copy of the disk. Azure VM snapshots are useful to quickly save the state of a VM before you make configuration changes. In the event of an issue or error, you can restore a VM by using a snapshot. When a VM has more than one disk, a snapshot is taken of each disk independently. To take application-consistent backups, consider stopping the VM before you take disk snapshots. Alternatively, use Azure Backup, which supports automated backups while the VM is running.

Create snapshot

Before you create a snapshot, you need the ID or name of the disk. Use az vm show to show the disk ID. In this example, the disk ID is stored in a variable so that it can be used in a later step.

osdiskid=$(az vm show \
   -g myResourceGroupDisk \
   -n myVM \
   --query "storageProfile.osDisk.managedDisk.id" \
   -o tsv)

Now that you have the ID, use az snapshot create to create a snapshot of the disk.

az snapshot create \
    --resource-group myResourceGroupDisk \
    --source "$osdiskid" \
    --name osDisk-backup

Create disk from snapshot

This snapshot can then be converted into a disk using az disk create, which can be used to recreate the virtual machine.

az disk create \
   --resource-group myResourceGroupDisk \
   --name mySnapshotDisk \
   --source osDisk-backup

Restore virtual machine from snapshot

Before you delete the VM, use az vm show to store the resource IDs of all its data disks in the dataDiskIds variable.

dataDiskIds=$(az vm show \
   --resource-group myResourceGroupDisk \
   --name myVM \
   --query "storageProfile.dataDisks[].managedDisk.id" \
   --output tsv)

To demonstrate virtual machine recovery, delete the existing virtual machine using az vm delete.

az vm delete \
--resource-group myResourceGroupDisk \
--name myVM

Create a new virtual machine from the snapshot disk.

az vm create \
    --resource-group myResourceGroupDisk \
    --name myVM \
    --attach-os-disk mySnapshotDisk \
    --os-type linux

Reattach data disks

All data disks need to be reattached to the virtual machine.

Use az vm disk attach with each resource ID to reattach all the data disks.

for dataDiskId in $dataDiskIds; do
   az vm disk attach \
      --resource-group myResourceGroupDisk \
      --vm-name myVM \
      --name "$dataDiskId"
done

Next steps

In this tutorial, you learned about VM disks topics such as:

Advance to the next tutorial to learn about automating VM configuration.