RAID1 mirroring - simple and reliable · Linux

RAID1 is the simplest RAID level that provides redundancy. Every block is written to all disks; reads come from any disk. RAID1 is the standard choice for boot disks and small, critical data.

How RAID1 works

Data-loss riskcreate RAID1— mdadm --create builds the array. The default metadata is 1.2 (at the start of the device); for a boot device, use --metadata=1.0 (at the end). Use --level=1 for RAID1. The array is immediately available but is degraded until the initial sync completes.

$ mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/sdb /dev/sdc

mdadm: Note: this array has metadata at the start and may not be suitable as a boot device. If you plan to store '/boot' on this device, please ensure that your version of mdadm supports metadata at the start (e.g., v1.2 or later).
mdadm: Defaulting to version 1.2 metadata
mdadm: array /dev/md0 started.

When to use RAID1

Use RAID1 for:

Boot disks - the OS image is mirrored, so any disk can fail and the host boots from the surviving one.
Small, critical data - log volumes, audit trails, configuration files.
Workloads with mostly reads - boot drives, read-heavy application data.

Do not use RAID1 for:

Large data volumes (capacity is half the raw size).
Write-heavy workloads (no performance benefit over a single disk).
The same goal is better served by RAID10 (which gives both striping and mirroring).

Detecting a failed disk

Read-only / Safemdadm detail degraded— State: clean, degraded, recovering means one device has failed and the array is rebuilding from the surviving one. The Rebuild Status shows progress. Working Devices is 2 (one active, one spare being rebuilt). Active Devices is 1 - only one mirror leg is healthy.

$ mdadm --detail /dev/md0 | grep -E 'State|Active|Working'

State : clean, degraded, recovering
Rebuild Status : 12% complete
Active Devices : 1
Working Devices : 2

When a disk fails, the array becomes degraded but stays online. The kernel serves reads from the surviving leg and reconstructs writes to the new disk (or a spare) when it appears. Performance drops because every read becomes a single-disk read.

Replacing a failed disk

Data-loss riskreplace failed disk— --fail marks the member faulty so md stops using it. --remove then detaches it, which is only legal once the device is non-active. --add attaches the replacement and the kernel starts the rebuild immediately. Confirm the serial of /dev/sdb with lsblk before the --fail, and re-read lsblk after the swap because /dev/sdd may not be the name you expected. Watch /proc/mdstat for progress.

$ sudo mdadm /dev/md0 --fail /dev/sdb; sudo mdadm /dev/md0 --remove /dev/sdb; sudo mdadm /dev/md0 --add /dev/sdd

mdadm: set /dev/sdb faulty in /dev/md0
mdadm: hot removed /dev/sdb from /dev/md0
mdadm: added /dev/sdd
mdadm: rebuild started

If the disk is failing but has not failed - pending sectors, occasional read errors - prefer mdadm /dev/md0 --replace /dev/sdb --with /dev/sdd. The array rebuilds onto the spare while the suspect disk stays in service, so a read error mid-rebuild is still recoverable. See the degraded-rebuilds lesson for the full procedure.