ZFS — Advanced Filesystem

If you like this project, consider supporting me on Buy Me a Coffee ☕️

tags:

• storage
• zfs
• filesystem
• snapshots
• raidz
• sysadmin

• linux

  💾 ZFS Cheatsheet

Description

ZFS is a combined filesystem and volume manager with checksumming, copy-on-write, snapshots, cloning, and built-in RAID (mirror, raidz). Used on NAS and backup servers. / ZFS — combined filesystem and volume manager с checksumming, copy-on-write, snapshots, cloning и встроенным RAID (mirror, raidz). Используется на NAS и backup-серверах.

Common use cases / Типовые сценарии:

• NAS / NAS / БД на NAS
• Snapshot-based backups / Snapshot-бэкапы
• Proxmox / FreeBSD storage / Хранилища Proxmox / FreeBSD
• Data integrity protection / Защита целостности данных

Status: Actively maintained (OpenZFS). Alternatives: Btrfs, LVM + ext4/XFS, mdadm. / Статус: активно развивается (OpenZFS); альтернативы — Btrfs, LVM+ext4/XFS, mdadm.

Paths: pools at /dev/disk/by-id/...; datasets under pool root.
Tools: zpool, zfs.

Cross-reference: LUKS, Samba, Ceph, GlusterFS.

Installation

Install OpenZFS / Установка OpenZFS

# Debian/Ubuntu
apt install zfsutils-linux

# RHEL/Rocky/Alma (EPEL or ZFS repo)
dnf install zfs

zfs --version
zpool --version

Configuration

Create pool / Создание пула

# Single-disk pool (testing only!)
zpool create -f tank <DISK>

# Mirror
zpool create -f tank mirror <DISK1> <DISK2>

# raidz2 (2-disk fault tolerance)
zpool create -f tank raidz2 <DISK1> <DISK2> <DISK3> <DISK4>

# raidz1
zpool create -f tank raidz1 <DISK1> <DISK2> <DISK3>

Dataset layout / Раскладка dataset-ов

zfs create tank/data
zfs create tank/data/home
zfs create tank/data/postgres
zfs set mountpoint=/data tank/data
zfs set compression=lz4 tank/data
zfs set atime=off tank/data

Core Management

Pool status / Статус пула

zpool status -v tank
zpool list
zpool get all tank
zpool scrub tank

Dataset operations / Операции с dataset

zfs list
zfs list -o name,used,avail,refer,compressratio
zfs create tank/snapshots
zfs set quota=50G tank/data/postgres
zfs set reservation=10G tank/data/postgres
zfs destroy tank/old
zfs rename tank/old tank/new

Sysadmin Operations

Snapshots / Снимки

zfs snapshot tank/data@$(date +%F)
zfs list -t snapshot
zfs rollback tank/data@<SNAPSHOT>
zfs clone tank/data@<SNAPSHOT> tank/data/clone
zfs destroy tank/data@<SNAPSHOT>

Send/receive / Отправка и приём

# Backup snapshot to remote
zfs send tank/data@<SNAPSHOT> | ssh backup@<HOST> zfs receive backup/data

# Incremental send
zfs send -i tank/data@<SNAP1> tank/data@<SNAP2> | ssh backup@<HOST> zfs receive backup/data

# Resume token
zfs send -t <TOKEN>

Scrub and trim / Scrub и TRIM

zpool scrub tank
zpool status -v tank
zpool trim tank

Logrotate (scrub logs) / Ротация логов

/etc/logrotate.d/zfs

/var/log/zfs/* {
    weekly
    rotate 4
    compress
    delaycompress
    missingok
    notifempty
}

Security

Hardening / Ужесточение

1. Use ashift=12 for 4K-sector disks / Используйте ashift=12 для 4K-дисков
2. Avoid single-disk pools in production / Не используйте single-disk пулы в production
3. Set quotas to prevent fill-up / Задавайте quotas
4. Scrub regularly / Регулярно делайте scrub
5. Encrypt sensitive datasets / Шифруйте чувствительные dataset-ы
6. Test restore before trusting backups / Тестируйте восстановление

zfs create -o encryption=aes-256-gcm -o keyformat=passphrase tank/secure

[!WARNING] Encryption keys are lost if passphrase/key is forgotten — no recovery. / Ключи шифрования не восстановимы.

Backup and Restore

Backup via send/recv / Бэкап через send/recv

zfs snapshot -r tank@nightly
zfs send -R tank@nightly | ssh backup@<HOST> zfs receive backup/tank

Export/import / Экспорт и импорт

zpool export tank
zpool import tank
zpool import -d /dev/disk/by-id tank

Troubleshooting

zpool status -v tank
zfs list -o name,used,avail
zpool scrub tank

Comparison Tables

Storage systems / Файловые системы

Production Runbooks

Runbook: Create production pool / Создание production-пула

1. Identify disks by-id / Определить диски по by-id
2. zpool create -f tank raidz2 <DISKS> -o ashift=12 / Создать pool
3. Create datasets; set compression=lz4, atime=off / Создать datasets
4. Set quotas on user-facing datasets / Задать quotas
5. Export/import test / Проверить export/import
6. Configure scrub cron / Настроить cron на scrub
7. Document pool layout / Задокументировать раскладку

Runbook: Replace failed disk / Замена деградировавшего диска

1. zpool status -v tank — identify failed device / Определить диск
2. Online replace or offline; replace hardware / Заменить диск
3. zpool replace tank <OLD> <NEW> / Заменить в пуле
4. Wait for resilver / Дождаться resilver
5. zpool status — verify healthy / Проверить статус
6. Scrub after resilver / Сделать scrub

Documentation Links

• OpenZFS docs — https://openzfs.github.io/openzfs-docs/
• ZFS wiki — https://openzfs.org/
• man zpool — https://linux.die.net/man/8/zpool
• man zfs — https://linux.die.net/man/8/zfs