Btrfs: Modern Copy-on-Write Filesystem

Btrfs: Where Your Data Gets Superpowers

Imagine a filesystem that could travel back in time. One that never loses data, even when you accidentally delete something. A filesystem that can detect and fix corruption before you even know it's there. Welcome to Btrfs—where science fiction meets your storage!

Btrfs (B-tree filesystem, pronounced "Butter FS" or "Better FS") isn't just another filesystem—it's a complete rethinking of how we store data. Born at Oracle in 2007 and now community-driven, Btrfs brings enterprise-grade features to everyone.

Think of Btrfs as Linux's Swiss Army knife for storage. While ext4 is your reliable daily driver, Btrfs is the transformer that can morph into whatever you need: a snapshot machine, a RAID array, a compression engine, or all of the above simultaneously!

Key Features

1. Copy-on-Write: The Magic Behind Everything

The Revolution: Traditional filesystems are like writing with a pen—once you overwrite something, it's gone forever. Btrfs is like having an infinite stack of transparent sheets. Every change creates a new layer, and you can always peek back at previous versions.

Watch Copy-on-Write in action—see how Btrfs creates snapshots instantly and preserves your data:

Copy-on-Write (CoW) in Action

Original File State

File contains blocks A, B, C, D at their original locations

Storage Blocks

Addr: 0x1000

Refs: 1

Addr: 0x2000

Refs: 1

Addr: 0x3000

Refs: 1

Addr: 0x4000

Refs: 1

Metadata Trees

Current File

1000

2000

3000

4000

Why Copy-on-Write is Brilliant

Instant Snapshots

Creating snapshots takes milliseconds, not minutes

Space Efficient

Only changed blocks use additional space

Data Protection

Original data never overwritten, always consistent

Time Travel

Access any previous version instantly

Why CoW Changes Everything

Copy-on-Write isn't just a feature—it's the foundation that enables:

Instant snapshots (no data copying needed)
Atomic transactions (all or nothing writes)
Data integrity (old data never corrupted)
Time travel (rollback to any snapshot)

2. Subvolumes

Subvolumes are like lightweight filesystems within Btrfs:

# Create subvolume
sudo btrfs subvolume create /mnt/data/projects

# List subvolumes
sudo btrfs subvolume list /mnt

# Subvolumes can be mounted independently
sudo mount -o subvol=projects /dev/sda1 /mnt/projects

# Each subvolume can have different mount options
sudo mount -o subvol=databases,nodatasum /dev/sda1 /mnt/db

3. Snapshots

Instant, space-efficient copies of subvolumes:

# Create snapshot
sudo btrfs subvolume snapshot /mnt/data /mnt/snapshots/data-$(date +%Y%m%d)

# Create read-only snapshot
sudo btrfs subvolume snapshot -r /mnt/data /mnt/snapshots/data-backup

# List snapshots
sudo btrfs subvolume list -s /mnt

# Rollback to snapshot
sudo btrfs subvolume delete /mnt/data
sudo btrfs subvolume snapshot /mnt/snapshots/data-backup /mnt/data

Btrfs Architecture

B-tree Structure

Everything in Btrfs is stored in B-trees:

# B-tree Structure:
#                     Root Tree
#                         |
#         ┌───────────────┼───────────────┐
#         |               |               |
#     Extent Tree    Chunk Tree    Filesystem Tree
#         |               |               |
#     [Extents]      [Devices]      [Files/Dirs]

Object Types

Subvolume: Independent filesystem tree
Snapshot: CoW copy of a subvolume
Extent: Contiguous data blocks
Chunk: Physical device allocation unit
Block Group: Collection of chunks

Creating and Managing Btrfs

Creating Btrfs Filesystem

# Single device
sudo mkfs.btrfs /dev/sdb1

# Multiple devices (RAID)
sudo mkfs.btrfs -d raid1 -m raid1 /dev/sdb1 /dev/sdc1

# With label and options
sudo mkfs.btrfs -L "MyData" \
                 -O compress-force \    # Force compression
                 -O no-holes \          # Efficient sparse files
                 /dev/sdb1

# For SSD with optimizations
sudo mkfs.btrfs -O ssd \
                 -O discard=async \
                 /dev/sdb1

Mount Options

# Basic mount
sudo mount /dev/sdb1 /mnt/btrfs

# With compression
sudo mount -o compress=zstd:3 /dev/sdb1 /mnt/btrfs

# SSD optimizations
sudo mount -o ssd,discard=async,noatime /dev/sdb1 /mnt/btrfs

# Space cache for performance
sudo mount -o space_cache=v2 /dev/sdb1 /mnt/btrfs

# Autodefrag for desktop use
sudo mount -o autodefrag /dev/sdb1 /mnt/btrfs

Advanced Features

1. Compression

Btrfs supports transparent compression:

# Mount with compression
sudo mount -o compress=zstd /dev/sdb1 /mnt

# Compression algorithms:
# - zlib: Best compression, slowest
# - lzo: Fast compression, moderate ratio
# - zstd: Best balance (recommended)

# Set compression level
sudo mount -o compress=zstd:3 /dev/sdb1 /mnt  # Level 1-15

# Force compression on all files
sudo mount -o compress-force=zstd /dev/sdb1 /mnt

# Check compression ratio
sudo compsize /mnt/

2. RAID Support

Built-in RAID without mdadm:

# RAID levels:
# - raid0: Striping (performance, no redundancy)
# - raid1: Mirroring (redundancy)
# - raid10: Striped mirrors
# - raid5/6: Parity (not production ready!)

# Create RAID1 filesystem
sudo mkfs.btrfs -d raid1 -m raid1 /dev/sdb /dev/sdc

# Add device to existing filesystem
sudo btrfs device add /dev/sdd /mnt

# Remove device
sudo btrfs device delete /dev/sdc /mnt

# Balance data across devices
sudo btrfs balance start /mnt

# Convert RAID level
sudo btrfs balance start -dconvert=raid10 -mconvert=raid10 /mnt

3. Deduplication

Remove duplicate data blocks:

# Online deduplication (not recommended for production)
# Requires third-party tools like duperemove

# Offline deduplication
sudo duperemove -dr /mnt/btrfs

# Check deduplication savings
sudo duperemove -d --hashfile=/tmp/hash.db /mnt/btrfs

4. Scrubbing

Verify data integrity using checksums:

# Start scrub
sudo btrfs scrub start /mnt

# Check scrub status
sudo btrfs scrub status /mnt

# Scrub specific device
sudo btrfs scrub start /dev/sdb1

# Cancel scrub
sudo btrfs scrub cancel /mnt

# Schedule regular scrubs (cron)
0 2 * * 0 /usr/bin/btrfs scrub start /mnt

5. Send/Receive

Efficient backup and replication:

# Create initial snapshot
sudo btrfs subvolume snapshot -r /mnt/data /mnt/data-snap1

# Send snapshot to another location
sudo btrfs send /mnt/data-snap1 | sudo btrfs receive /backup/

# Incremental send (only changes)
sudo btrfs subvolume snapshot -r /mnt/data /mnt/data-snap2
sudo btrfs send -p /mnt/data-snap1 /mnt/data-snap2 | \
     sudo btrfs receive /backup/

# Send over network
sudo btrfs send /mnt/snapshot | ssh remote "sudo btrfs receive /mnt/"

Quota Management

Setting Up Quotas

# Enable quotas
sudo btrfs quota enable /mnt

# Create quota group
sudo btrfs qgroup create 1/0 /mnt

# Set limit (10GB)
sudo btrfs qgroup limit 10G 1/0 /mnt

# Check quota usage
sudo btrfs qgroup show /mnt

# Disable quotas
sudo btrfs quota disable /mnt

Performance Optimization

For SSDs

# Mount options for SSD
mount -o ssd,discard=async,compress=zstd,noatime /dev/sdb1 /mnt

# Disable CoW for databases (VM images, databases)
chattr +C /mnt/databases/

# Note: +C only works on empty files/directories

For HDDs

# Mount options for HDD
mount -o compress=zstd,autodefrag,noatime /dev/sdb1 /mnt

# Enable space cache
mount -o space_cache=v2 /dev/sdb1 /mnt

# Optimize for large files
mount -o compress=zstd,max_inline=0 /dev/sdb1 /mnt

Database Optimization

# Disable CoW for database files
mkdir /mnt/postgres
chattr +C /mnt/postgres

# Mount subvolume with nodatasum
mount -o subvol=databases,nodatasum,nodatacow /dev/sdb1 /mnt/db

Maintenance and Monitoring

Filesystem Usage

# Show filesystem usage
sudo btrfs filesystem show /mnt

# Detailed usage
sudo btrfs filesystem usage /mnt

# Device statistics
sudo btrfs device stats /mnt

# Check free space
sudo btrfs filesystem df /mnt

Balance Operations

# Balance filesystem (redistribute data)
sudo btrfs balance start /mnt

# Balance with filters
sudo btrfs balance start -dusage=50 /mnt  # Balance chunks <50% full

# Monitor balance progress
sudo btrfs balance status /mnt

# Cancel balance
sudo btrfs balance cancel /mnt

Defragmentation

# Defragment file
sudo btrfs filesystem defragment /path/to/file

# Defragment directory recursively
sudo btrfs filesystem defragment -r /mnt/data

# Defragment with compression
sudo btrfs filesystem defragment -czstd -r /mnt/data

# Note: Defragmentation breaks CoW links (increases space usage)

Troubleshooting

Common Issues

1. No Space Left (but df shows space)

# Check metadata usage
sudo btrfs filesystem usage /mnt

# Balance metadata
sudo btrfs balance start -m /mnt

# If metadata is full
sudo btrfs balance start -dusage=0 /mnt

2. Read-only Filesystem

# Check for errors
sudo btrfs check --readonly /dev/sdb1

# Try to clear errors
sudo btrfs rescue zero-log /dev/sdb1

# Mount with recovery options
mount -o recovery,ro /dev/sdb1 /mnt

3. Corrupted Filesystem

# Check filesystem (unmounted)
sudo umount /mnt
sudo btrfs check /dev/sdb1

# Repair (use with caution!)
sudo btrfs check --repair /dev/sdb1

# Recovery mount options
mount -o recovery,usebackuproot /dev/sdb1 /mnt

Recovery Tools

# Restore from backup superblock
sudo btrfs rescue super-recover /dev/sdb1

# Find and recover files
sudo btrfs restore /dev/sdb1 /recovery/

# Show backup roots
sudo btrfs rescue chunk-recover /dev/sdb1

Btrfs vs Other Filesystems

Feature Comparison

Feature	Btrfs	ext4	XFS	ZFS
CoW	✓	✗	✗	✓
Snapshots	✓	✗	✗	✓
Compression	✓	✗	✗	✓
Checksums	✓	✗	✗	✓
Built-in RAID	✓	✗	✗	✓
Deduplication	✓	✗	✗	✓
Subvolumes	✓	✗	✗	✓
Linux mainline	✓	✓	✓	✗

Performance Characteristics

# Performance Characteristics:
#
# Sequential Write:
# XFS     ████████████████████ 100%
# ext4    ███████████████████  95%
# Btrfs   ████████████████     80%
# ZFS     ███████████████      75%
#
# Random I/O:
# ext4    ████████████████████ 100%
# XFS     ███████████████████  95%
# Btrfs   ███████████████      75%
# ZFS     ██████████████       70%
#
# Snapshot Performance:
# ZFS     ████████████████████ 100%
# Btrfs   ███████████████████  95%
# ext4    N/A
# XFS     N/A

Best Practices

1. Snapshot Strategy

# Automated snapshots with systemd timer
cat > /etc/systemd/system/btrfs-snapshot.service << EOF
[Unit]
Description=Btrfs snapshot

[Service]
Type=oneshot
ExecStart=/usr/local/bin/btrfs-snapshot.sh
EOF

cat > /etc/systemd/system/btrfs-snapshot.timer << EOF
[Unit]
Description=Daily Btrfs snapshot

[Timer]
OnCalendar=daily
Persistent=true

[Install]
WantedBy=timers.target
EOF

2. Backup Strategy

# Regular send/receive backups
#!/bin/bash
DATE=$(date +%Y%m%d)
btrfs subvolume snapshot -r /mnt/data /mnt/snapshots/data-$DATE
btrfs send /mnt/snapshots/data-$DATE | \
    btrfs receive /backup/snapshots/

# Clean old snapshots
btrfs subvolume delete /mnt/snapshots/data-$(date -d '30 days ago' +%Y%m%d)

3. Monitoring Script

#!/bin/bash
# Monitor Btrfs health

# Check device errors
btrfs device stats /mnt | grep -v ' 0$'

# Check space usage
usage=$(btrfs filesystem df /mnt | grep "Data.*used" | grep -o '[0-9]*\.[0-9]*')
if (( $(echo "$usage > 90" | bc -l) )); then
    echo "Warning: Filesystem usage above 90%"
fi

# Check scrub status
btrfs scrub status /mnt

When to Use Btrfs

✅ Perfect for:

Desktop/laptop systems: Snapshots for easy recovery
Development environments: Quick rollback capability
Container hosts: Efficient storage with deduplication
Home NAS: Built-in RAID and data integrity
Backup servers: Send/receive for efficient replication

❌ Avoid for:

Databases in production: CoW overhead impacts performance
RAID 5/6 requirements: Not stable yet
Maximum stability needs: Use ext4 or XFS
Very large filesystems: ZFS handles scale better

Limitations and Considerations

Current Limitations

RAID 5/6: Still considered unstable
Deduplication: Only offline, high memory usage
Quotas: Performance impact, occasional bugs
Large filesystems: Some operations slow with many snapshots
fsck: Limited repair capabilities compared to ext4

Hardware Considerations

RAM: More demanding than ext4 (1GB minimum recommended)
CPU: Compression and checksumming add overhead
Storage: CoW can increase space usage with many snapshots

Future Development

Upcoming Features

RAID 5/6 stability: Active development
Online deduplication: Planned improvements
Reflink for files: More efficient file copies
Encryption: Native encryption support
Better fsck: Enhanced repair capabilities

Migration to Btrfs

From ext4

# In-place conversion (backup first!)
sudo umount /dev/sdb1
sudo btrfs-convert /dev/sdb1

# Mount as Btrfs
sudo mount /dev/sdb1 /mnt

# If successful, delete ext4 backup
sudo btrfs subvolume delete /mnt/ext2_saved

# If issues, rollback
sudo btrfs-convert -r /dev/sdb1

Fresh Migration

# Create new Btrfs filesystem
sudo mkfs.btrfs /dev/sdc1

# Mount both
sudo mount /dev/sdb1 /mnt/old  # ext4
sudo mount /dev/sdc1 /mnt/new  # btrfs

# Copy with attributes preserved
sudo rsync -avxHAX /mnt/old/ /mnt/new/

# Or use btrfs send/receive for Btrfs to Btrfs

Conclusion

Btrfs represents a significant advancement in Linux filesystem technology, bringing features like snapshots, compression, and checksums that were previously only available in proprietary or non-mainline solutions. While it may not match ext4's maturity or XFS's raw performance, its feature set makes it ideal for many use cases, particularly where data integrity and flexibility are priorities.

The CoW design enables powerful features but comes with trade-offs. Understanding these trade-offs helps you decide when Btrfs is the right choice. For desktop users, developers, and home servers, Btrfs offers compelling advantages. For high-performance databases or maximum stability requirements, traditional filesystems may still be preferable.

As Btrfs continues to mature, it's becoming an increasingly attractive option for Linux systems, offering a glimpse of the future of Linux storage management.

← Back to Filesystems Overview | Compare with ZFS →

Table of Contents

Copy-on-Write (CoW) in Action

Original File State

Storage Blocks

Metadata Trees

Why Copy-on-Write is Brilliant