Skip to main content
StashRaid

Storage Parameters

Usable Capacity

12 TB

Fault Tolerance

4

Array Resilience

Architecture

RAID 5 • Parity

Fault Tolerance

1 Concurrent Failures Allowed

Current Configuration

RAID 5 (Parity)

12 TB Usable

Total RAW: 16 TB

Disk 01DATA
Disk 02DATA
Disk 03DATA
Disk 04PARITY

Total Drives

4 x 4 TB

Efficiency

75% Utilized

Usable: 12 TBWasted to redundancy: 4 TB

Pros & Cons

  • PRO: Good balance of capacity/speed/safety

  • PRO: Cost-effective redundancy

  • CON: Slower writes due to parity calculation

    This setup prioritizes redundancy but has associated performance or capacity costs.

Storage Redundancy Guide

Hot Spare vs Cold Spare

Every RAID array will eventually experience a drive failure. How you prepare for that moment — whether a replacement drive sits idling in the chassis or waits on a shelf — defines your recovery time, your risk window, and your total cost of ownership. This guide breaks down the hot spare vs cold spare decision so you can choose the right strategy for your storage infrastructure.

~8 min readBeginner–Intermediate
SECTION_01: DEFINITION

What Is a Hot Spare?

A drive that sits powered on and idle inside the storage chassis, waiting to be automatically activated when any array drive fails.

Global Hot Spare

A global hot spare is not assigned to any specific array. It can replace a failed drive in any array within the same storage system, provided it meets the size and interface requirements. Most enterprise RAID controllers and NAS operating systems use global hot spares by default.

Best for heterogeneous environments where you have multiple arrays of varying sizes — one large spare can cover every array.

Dedicated Hot Spare

A dedicated hot spare is reserved for a single specific array. It will only activate to replace a failed drive in that particular volume group, even if another array loses a drive.

Useful for segregating spare capacity by workload — a critical database array gets its own spare while a media archive array shares a global pool.

SECTION_02: DEFINITION

What Is a Cold Spare?

A replacement drive stored on a shelf, in a drawer, or in a shipping box — powered off and manually installed only after a failure occurs.

The On-Hand Replacement

A cold spare is the simplest form of disaster preparedness: you buy one (or more) matching drives when you purchase your array drives and keep them offline. When a drive fails, you physically walk to the server room — or dispatch a remote-hands technician — open the chassis, swap the failed drive with the cold spare, and initiate the rebuild manually.

Cold spares accrue zero power-on hours, draw no electricity, and suffer no vibration or thermal wear. They are effectively as fresh as the day you bought them, which makes them ideal for environments where drive replacement happens infrequently.

The trade-off: recovery time is measured in human response time — minutes, hours, or even days — not in milliseconds.

SECTION_03: AT A GLANCE

Hot Spare vs Cold Spare

A side-by-side comparison of the key differences.

DimensionHot SpareCold Spare
Power stateSpinning 24/7, idlePowered off, stored
ActivationAutomatic on drive failureManual install required
Rebuild start timeImmediate (seconds–minutes)Human-dependent (minutes–days)
Wear & tearAccrues power-on hours, no I/OZero wear until deployed
Bay usageOccupies a drive slotNo bay used until installed
CostUpfront drive + bay costDrive cost only (no bay)
Best forCritical production systemsHomelabs, backups, archives
SECTION_04: PRO HOT SPARE

When Hot Spares Save Your Data

The automatic rebuild that starts the moment a drive dies.

Immediate Vulnerability Reduction

The moment a RAID controller detects a failed drive, it begins rebuilding the array onto the hot spare — no human in the loop. In a RAID 5 array, the failure of a single drive puts the entire array at risk: another failure during the rebuild window means total data loss. Every hour the array runs in a degraded state is a gamble.

For large-capacity drives (12 TB, 16 TB, 20 TB+), rebuilds can take 12–48 hours depending on the drive speed, controller throughput, and array utilisation. A hot spare cuts the start time of that rebuild to near-zero, whereas a cold spare adds the delay of noticing the failure, traveling to the server, and installing the replacement before the rebuild can even begin.

Critical for RAID 5 on large drives: The Unrecoverable Bit Error Rate (UBER) of modern drives means there is a real statistical chance that a 12+ TB RAID 5 rebuild hits a read error. Shrinking the degraded window is your best defence. See the RAID Rebuild Time Guide for detailed timing projections.
SECTION_05: CON HOT SPARE

When Hot Spares Waste a Bay

The spare drive spins 24/7, consuming power and accruing runtime — while doing nothing useful.

The Idle Tax

A hot spare is not free capacity. It occupies a drive bay, sits on the SAS/SATA bus, and draws power — typically 3–8 watts — 24 hours a day, 365 days a year. Over three years that is 200–700 kWh of electricity consumed by a drive that never served a single I/O request.

Worse, the drive accumulates power-on hours and experiences the same vibration and thermal cycling as every other drive in the chassis. By the time it is called into service, it may be years older and closer to its own failure MTBF than a fresh cold spare would be.

In a 24-bay chassis, dedicating even one slot to a hot spare reduces usable storage by ~4%. For a homelab or non-critical array, that bay may be better filled with usable capacity.

SECTION_06: COLD SPARE

Cold Spare Trade-Offs

No wear. No power draw. But someone has to be there to swap it.

Advantages

  • •Zero wear — the drive stays in its box until needed.
  • •No power draw — 0 watts idle.
  • •No bay consumed — every slot runs active storage.
  • •Lower upfront cost — no need for an extra controller channel or expander port.
  • •Fresh inventory — firmware revisions and production batches can be selected at replacement time.

Disadvantages

  • •Human response time — minutes to days of vulnerability.
  • •Physical access required — no good for remote sites without a technician.
  • •Risk of unavailable spare — the drive you bought may have been repurposed or the exact model is no longer compatible.
  • •Manual mistakes — wrong drive slot, wrong array, ESD damage during handling.
  • •Firmware drift — the cold spare may ship with different firmware than the array drives, causing compatibility warnings during rebuild.
SECTION_07: SIZING

Spare Sizing Rules

Not every drive can serve as a spare. Size, interface, and firmware all matter.

The Golden Rule: Size Parity

A hot spare must be at least as large as the largest drive in the array it is protecting. For SAS/SATA RAID controllers, if the hot spare is smaller than the failed drive, the rebuild will fail — the controller cannot fit the data onto a physically smaller device.

For global hot spares, the spare must be ≥ the largest drive across all arrays in the system. A 10 TB global spare cannot replace a failed 12 TB drive. For dedicated hot spares, the spare must be ≥ each drive in the array it is assigned to.

Best practice: use a drive of the same model and capacity as the array drives, or one size larger. A slightly larger spare will work — the extra capacity simply goes unused.

Firmware note: Enterprise RAID controllers (Broadcom/LSI, Adaptec) are strict about SATA vs SAS inter-mixing. A SAS hot spare can replace a SAS drive; a SATA hot spare can replace a SATA drive. Some controllers allow mixed interfaces in a single array with reduced performance — check your controller's manual before relying on cross-protocol sparing.
SECTION_08: RAID LEVELS

RAID Level Considerations

Your parity scheme changes how urgently you need that spare.

RAID 5 — Always Use a Hot Spare

RAID 5 can survive exactly one drive failure. If a second drive fails during rebuild, the array is destroyed. With large-capacity drives (12 TB+), rebuilds take 24–48 hours and the UBER risk is non-trivial. A hot spare that kicks in immediately is the single best mitigation for RAID 5's vulnerability. Without one, every minute of degraded operation is a gamble.

RAID 6 — Hot Spare Is Optional

RAID 6 can survive two simultaneous drive failures. The chance of a third failure during the rebuild window is extremely low for most deployments. While a hot spare is still beneficial for uptime — the array returns to full redundancy faster — it is not the existential necessity it is for RAID 5. Many RAID 6 deployments run with cold spares only.

SECTION_09: PLATFORMS

How Hot Sparing Works Across Platforms

Not all storage systems implement hot spares the same way.

TrueNAS (FreeNAS)

TrueNAS uses a spare pool concept. You designate one or more drives as spares in the web UI, and the system automatically swaps a failed drive's slot in the pool with the spare. The rebuild runs in the background while the pool stays online. TrueNAS will cycle through spare drives if multiple failures occur.

Spares can be global across all pools or scoped to a specific pool in TrueNAS Scale.

Hardware RAID (LSI/Broadcom, Adaptec)

Hardware RAID controllers manage hot spares at the firmware level. When a drive fails, the controller automatically begins rebuilding to the designated hot spare — no OS involvement required. Most enterprise controllers support both global and dedicated hot spares, configurable via the controller BIOS or management utilities (StorCLI, MegaRAID Storage Manager).

Rebuilds on hardware RAID are typically faster than software RAID because the controller offloads parity calculations from the CPU.

Synology DSM (SHR)

Synology's SHR (Synology Hybrid RAID) supports hot spares by designating a drive as a "hot spare" in Storage Manager. If any drive in the SHR group fails, DSM automatically rebuilds the volume onto the hot spare. SHR's flexibility means the hot spare does not need to match the failed drive's size exactly — it only needs to be ≥ the smallest drive in the group.

DSM also supports "global hot spares" that protect multiple storage pools on the same NAS unit.

unRAID

unRAID does not support traditional hot spares. The unRAID array is built around a single parity-drive model (with optional dual parity). When a data drive fails, you physically replace it and the array is rebuilt from parity on the next array start. There is no automatic spare activation mechanism in unRAID's design.

The closest equivalent is to keep a precleared drive on hand as a cold spare. When a failure occurs, assign the new drive to the failed slot and start the parity-sync manually. See the RAID Rebuild Time Guide for unRAID-specific rebuild estimates.

SECTION_10: BEST PRACTICES

Best Practices

A pragmatic approach to spare strategy.

1. Always have at least one cold spare on hand

Whether you use hot spares or not, maintain a cold spare for every drive model in your array. Drive models go out of production quickly. If you need a replacement and the exact model is discontinued, you may face firmware incompatibilities or shifting performance characteristics. Buy the spare at the same time you buy the array drives.

2. Hot spares for critical production only

Reserve hot spares for arrays where downtime costs exceed the cost of the spare drive and the occupied bay. For a database server or primary file server, a hot spare is cheap insurance. For a media archive or backup target, a cold spare is usually sufficient.

3. Rotate spares periodically

If a hot spare has been sitting idle for a year, consider swapping it with an active drive to distribute wear across the fleet. Some enterprise arrays support "smart sparing" that rotates spares automatically.

4. Monitor spare health

A hot spare that silently fails is worse than no spare at all. Make sure your monitoring system (SNMP, iDRAC, iLO, IPMI) reports the health status of spare drives just as it does for active drives. A dead spare gives you a false sense of security.

5. Consider the "2:1" rule

For production environments, a common rule of thumb: one hot spare per chassis (or per 24 drives) + one cold spare per unique drive model on-site. This ensures automatic recovery covers the most common single-failure scenario while a cold spare backs up the hot spare itself.

SECTION_11: FAQ

Frequently Asked Questions

Common questions about hot spare and cold spare strategies.

Should I use a hot spare or cold spare?

It depends on your tolerance for downtime. If your array being degraded for a few hours or even a day is acceptable, a cold spare is cost-effective and simpler. If you need the array to return to full redundancy as fast as mechanically possible — or your server is in a remote location — a hot spare is the right choice. Many enterprises use both: a hot spare for automatic recovery and a cold spare as a backup for the hot spare.

Does a hot spare reduce the rebuild window?

A hot spare reduces the time to <em>begin</em> the rebuild to near-zero — no human intervention needed. It does not reduce the rebuild <em>duration</em> itself, which is determined by drive speed, controller throughput, array utilisation, and RAID level. However, starting the rebuild immediately rather than hours or days later means the total vulnerability window is much shorter. See the RAID Rebuild Time Guide for detailed timing estimates per RAID level and drive size.

Can a hot spare smaller than my other drives work?

No — not safely. A hot spare must be at least as large as the largest drive in the array it protects. If it is smaller, the rebuild will fail because the controller cannot fit the data. Using a larger drive is fine — the extra capacity simply goes unused. Some advanced filesystems (ZFS, Btrfs) can handle variable drive sizes, but the spare still needs enough capacity to hold all the data from the failed drive.

Does unRAID support hot spares?

No, unRAID does not support traditional hot spares. The unRAID architecture is based on a parity-protected JBOD model where each data drive is individually mounted. When a drive fails, you must manually stop the array, replace the drive, and rebuild from parity. The closest you can get is to keep a precleared drive ready and script the replacement steps, but there is no automatic spare activation mechanism. Cold spare + prompt manual replacement is the standard unRAID practice.

How many hot spares do I need?

As a baseline: one hot spare per chassis or per 24 drives. In a 12-bay server, one hot spare is typical. If you run multiple arrays with different drive sizes, you may need multiple hot spares or one large global spare that covers the largest array. For very large storage systems (48+ drives), consider two hot spares to cover the risk of a second failure during rebuild. For RAID 5 arrays, a hot spare per array is strongly recommended.

Can I convert a hot spare to active storage later?

Yes, but it requires care. You can remove the hot spare assignment and add the drive to an array as active capacity — but only if no rebuild is currently consuming it. Most controllers (LSI/Broadcom, Adaptec) and NAS OSes (TrueNAS, Synology DSM) allow this via their management interface. Keep in mind that using a hot spare for active storage eliminates your spare coverage until you assign a new one.

Related Guides

Choosing between a hot spare and cold spare is just one part of storage resilience planning. The duration of your rebuild window directly impacts how urgent the spare decision is — especially as drive capacities grow.

RAID Rebuild Time Guide