Adding & Replacing Drives, and Rebuilding Parity

Growing Your Storage - The Overview

When your server starts filling up, you have two simple paths: add another drive to increase total space, or replace a small drive with a bigger one. Both are normal, and Unraid supports them well.

Because each data drive in Unraid holds whole files on its own, adding a drive just adds its space to the pile. You are not rebuilding everything from scratch - you are adding a new shelf to the cupboard.

• Add a new drive - plug in another disk and hand its space to the array.
• Replace a small drive - swap a full small disk for a larger one and rebuild onto it.
• Both over time - most folks do a mix as drives fill and prices drop.

Adding a Brand-New Drive

Adding a fresh, empty drive is the friendliest upgrade. You physically install it, assign it in the web page, and Unraid gets it ready. Nothing you already have is touched.

Clearing a brand-new drive can take a while, because Unraid writes zeros across it so parity stays correct. That is normal and safe - just let it run. Formatting afterward is quick.

• With the array stopped, install the new drive and note its serial number.
• In the web page, assign the new drive to an empty data slot.
• Let Unraid clear the drive - this prepares it and keeps parity valid.
• When prompted, format the drive so it is ready to hold files.
• Start the array; your new space is now part of the pile.

Why You Stop the Array to Change Drives

Unraid asks you to stop the array before you add, remove, or reassign a drive. This is a safety step, not busywork. It lets the server change its configuration cleanly, without files being written mid-change.

Think of it like turning off the mixer before you reach into the bowl. Stopping the array pauses all reading and writing so nothing gets scrambled while the drive lineup changes.

• Safely closes all shares so nothing is being written.
• Lets you assign or unassign drive slots without confusion.
• Protects parity from getting out of sync during the change.

Replacing a Failed Drive

When a data drive fails, Unraid does not lose your files - parity has your back. You slot in a healthy replacement and the server rebuilds the missing drive's contents onto it. This is the moment parity earns its keep.

During the rebuild, the server reads every other drive plus parity to recreate the lost files onto the new disk. Your data is being reconstructed - let it finish without interruption.

• Note which drive failed and write down BOTH its serial and the new drive's serial.
• Stop the array.
• Physically remove the failed drive - matching its serial number to be sure.
• Install the replacement (as large or larger than the drive it replaces).
• Assign the new drive to the failed drive's slot and start the array to rebuild.

Replacing With a Larger Drive

Swapping a small drive for a bigger one is how many people grow their server without adding more bays. The steps are much like a failed-drive replacement, with one size rule to respect.

The new data drive can be larger - that is the point. But remember the golden rule: the parity drive must be at least as large as your largest data drive. If your new data drive would be bigger than parity, you upgrade parity first.

• Data drive fits under parity - just replace the data drive and rebuild.
• Data drive bigger than parity - upgrade the parity drive first, then the data drive.
• Doing both - a parity swap can sometimes handle this in one guided move.

How the Rebuild Works

A rebuild sounds like magic, but the idea is simple arithmetic. Parity holds a running total across all your drives. If one drive goes missing, the server works backward from that total to figure out exactly what was on it.

So when you install a replacement, Unraid reads every remaining data drive and the parity drive together, calculates the missing pieces, and writes them onto the new drive. When it finishes, the new drive holds exactly what the old one did.

• Your files are being recreated, not copied from a backup.
• Every other drive must be healthy for the rebuild to work.
• The array often keeps working during the rebuild, just slower.

How Long a Rebuild Takes

Rebuilds are not quick. Depending on drive size and speed, expect anywhere from several hours to a full day or more. A large modern drive can easily take the better part of a day.

The server has to read across every drive to recreate the missing one, and it goes carefully. That is time well spent - a slow, correct rebuild beats a fast, botched one. Plan to start it when you can leave the server undisturbed.

• Drive size - bigger drives take proportionally longer.
• Drive speed - faster drives finish sooner.
• Using the server meanwhile - heavy use during a rebuild slows it down.

Upgrading the Parity Drive (Parity Swap), Simply

Sometimes you want a bigger parity drive - usually so you can add large data drives later. Unraid has a guided "parity swap" that copies parity onto a new, larger drive and frees up the old one to become a data drive.

You do not need to understand the internals. The web page walks you through it: assign the new larger drive for parity, and Unraid handles copying parity over and reshuffling roles. It is a guided process, but still one to do unhurried.

• You bought a drive bigger than your current parity and want to use it.
• You want to reuse the old parity drive as extra data space.
• You are setting up to add large data drives down the road.

Making Sure You Pull the RIGHT Disk

Inside the case, drives look identical. The only reliable way to tell them apart is the serial number printed on each drive's label. This one habit prevents the worst mistake you can make.

If you cannot see a label easily, it helps to have labeled your bays when you built the server. A quick paper map of which serial sits in which slot is worth its weight in gold on the day a drive fails.

• On the web page, find the drive you mean to remove and read its serial number.
• Write that serial down on paper - do not trust your memory.
• At the case, read the serial on each drive's label until you find the match.
• Only unplug the drive whose label matches your note, exactly.

Checking Serial Numbers Before You Touch Anything

Before you open the case at all, get your serial numbers straight on screen. Knowing exactly which drive is which - before your hands are inside - turns a tense job into a calm one.

Every drive Unraid sees is listed with its serial number. Take a moment to note the serial of the drive you are adding, replacing, or removing. This is the single most useful two minutes in the whole process.

• Keep a simple list: slot, drive model, serial number.
• Update it whenever you add or swap a drive.
• Snap a photo of the screen's drive list before you start.

Keeping Power Steady During a Rebuild (UPS)

A rebuild writes across your drives for hours. The one thing that can spoil it is a power cut in the middle. A UPS - a battery backup the server plugs into - protects you from exactly that.

A UPS keeps the server running through brief outages and flickers, and can tell Unraid to shut down gracefully if the power stays out. During a rebuild, that steady power is not a luxury - it is cheap insurance for your data.

• A rebuild is a long, sensitive operation - an interruption can undo hours of work.
• A sudden power loss mid-write can leave things in a messy state.
• A UPS gives the server time to finish or shut down cleanly.

What a "Disabled" Drive Means

One day you may see a drive marked "disabled," often with a red X. Do not panic. It means Unraid hit a write error on that drive and set it aside for safety - but your files are not lost.

A disabled drive has been dropped from active use, and Unraid now serves its files from the other drives plus parity. The drive itself might be genuinely failing, or it might have been a fluke like a loose cable. Either way, take it calmly.

• Do not immediately yank anything - your data is still available.
• Check the drive's health and its cable connection.
• Decide whether to rebuild onto the same drive or a fresh replacement.

The "Emulated" Drive Concept

When a drive is disabled or missing, Unraid can still show you all its files. It does this by "emulating" the drive - recreating its contents on the fly from the other drives and parity. It is clever, and it keeps you running.

So even with a drive physically gone, your shares look normal and your files open fine. The server is doing the parity math in real time to stand in for the missing drive until you replace it and rebuild.

• You keep access to your files while you sort out a replacement.
• It proves parity is doing its job.
• It buys you time to get the right drive and do the swap unhurried.

The New Config Tool - Handle With Care

New Config tells Unraid to forget its current drive lineup and start fresh with whatever you assign. It is useful for big reorganizations - and it is one of the easiest ways to get in trouble if used carelessly.

The danger is that New Config throws away the existing parity relationship. After using it, parity has to be rebuilt from scratch, and until it is, you have no protection. Worse, a wrong move here can lead to formatting a drive you meant to keep.

• You genuinely need to rearrange the whole drive assignment.
• You have a current backup of anything you cannot lose.
• You are following a known-good procedure, step by step.

Shrinking the Array, Briefly

Most people only ever grow their array, but sometimes you want to remove a drive - maybe it is old, or you are consolidating onto bigger disks. It is possible, just more fiddly than adding one.

The key idea is that you first move all the files off the drive you want to remove, so it is empty, then reconfigure the array without it. Because this touches the drive lineup, it usually involves the careful New Config step - so tread lightly.

• Copy every file off the drive you plan to remove.
• Confirm the drive is truly empty of anything you want.
• Reconfigure the array without it, then let parity rebuild.

Testing a New Drive Before Trusting It

Brand-new drives usually work fine, but not always. The unfortunate few that fail tend to fail early. Giving a new drive a quick workout before you rely on it catches those lemons before they cost you.

A common approach is to run the drive through a preclear or an extended self-test, which reads and writes across the whole drive to shake out any weak spots. If it passes, you can trust it. If it stumbles, you return it while it is still new.

• Early failures show up under a full read-write test.
• Better to find a bad drive now than during a future rebuild.
• A clean test gives you confidence the drive is sound.

Checking SMART Health First

Every modern drive keeps a little health report on itself called SMART. Unraid can show it to you. Glancing at SMART before and after a change tells you whether a drive is healthy or quietly struggling.

You do not need to read every line. A few key numbers - like reallocated sectors or pending sectors - are the ones that matter. If those are climbing, the drive is warning you. Unraid can even alert you when they change.

• Check a drive's SMART report before you trust it with data.
• Turn on SMART alerts so the server warns you of trouble early.
• Treat rising warning numbers as a sign to plan a replacement.

A Step-by-Step Safety Checklist

Whenever you add, replace, or remove a drive, run through this short list. It bundles every safety habit in this guide into one calm routine you can follow every time.

• Back up anything you cannot afford to lose, before you start.
• Make sure the server is on a UPS so power cannot cut mid-job.
• Note the serial numbers of every drive involved, on paper.
• Stop the array before changing the drive lineup.
• Match the serial on the label before unplugging ANY disk.
• Assign the new drive to the correct slot, then start the array.

What's in the 27-page PDF

• Adding a brand-new drive to grow your storage
• Replacing a failed drive, and swapping in a larger one
• How parity rebuilds a drive, and how long it takes
• Making absolutely sure you pull the RIGHT disk
• Keeping power steady during a rebuild with a UPS
• A clear, step-by-step safety checklist you can lean on

Why I wrote this

I write these as printable handouts because a guide you can keep in a drawer beats a web page you have to find again. No jargon, no judgment, no rushing. If you'd rather have someone sit beside you and go through it together, that's a flat $99 first visit with the 30-day follow-up included — serving Portage County from Atwater, Ohio. — Bill