Automated storage provisioning scripts

In this lesson, you will learn how to design robust shell scripts that automatically assemble, format, and mount complex Btrfs storage stacks upon system boot or drive insertion.

Why Automate Storage?

When dealing with complex storage—such as multi-device Btrfs pools or specific subvolume configurations—manually running commands like mkfs.btrfs or mount is prone to human error. Automation ensures that your environment remains consistent, especially in dynamic scenarios like container hosts or systems using hot-plugged storage hardware.

Structuring Your Provisioning Script

A robust provisioning script should follow a "check-act-verify" workflow. Before attempting to mount a Btrfs pool, your script must confirm the presence of the block devices and verify that the filesystem is healthy.

Consider this simplified example for mounting a multi-device pool:

#!/bin/bash
# Define target devices and mount point
DEVICES="/dev/sdb /dev/sdc"
MOUNT_POINT="/mnt/data_pool"

# 1. Ensure devices exist
for dev in $DEVICES; do
    if [ ! -b "$dev" ]; then
        echo "Error: Device $dev not found."
        exit 1
    fi
done

# 2. Mount with subvolume and compression settings
mount -t btrfs -o compress=zstd,subvol=logs "$DEVICES" "$MOUNT_POINT"

# 3. Verify success
if mountpoint -q "$MOUNT_POINT"; then
    echo "Storage stack assembled successfully."
else
    echo "Mount failed. Initiating error recovery check..."
    btrfs device scan
fi

Key Considerations

When building these scripts, keep three principles in mind:

1. Idempotency: Your script should handle cases where the filesystem is already mounted without throwing errors or attempting to re-format an existing pool. Use mountpoint -q to check state before taking action.
2. Error Handling: Always verify the exit status of your commands. If mount fails, your script should trigger a btrfs device scan or alert the administrator rather than silently failing.
3. Path Resolution: Rely on /dev/disk/by-uuid/ or /dev/disk/by-id/ instead of /dev/sdX labels, as device identifiers can change upon reboot or hot-plugging.

By wrapping these manual storage management tasks into a script, you turn a tedious, error-prone process into a single, predictable operation.

Key Takeaway: Automated storage provisioning scripts reduce configuration drift and human error by enforcing consistent, idempotent, and verifiable assembly of your Btrfs storage stacks.