Skip to main content

Array configuration

Unraid's storage system combines flexibility with data protection through its array and cache architecture. The array manages your primary storage with optional parity protection, while cache pools accelerate performance.

Below are some important array configuration principles:

Always use your largest drive(s) for parity.

When you add more disks to your array later, remember that you can't use a data disk that's larger than your parity disks. It's a good idea to buy the biggest hard drive for your parity disks at the start. This way, you won't be limited by smaller sizes when expanding later.

If you use two parity disks, they can be different sizes. However, keep in mind that no disk in your array can be larger than the smallest parity disks you have.

Do not use SSDs in the array - save them for pools or unassigned devices.

Unraid does not support TRIM or Discard operations for SSDs in the main array. Over time, this will cause SSD performance to degrade if they are used as array members. For best results, use SSDs in cache pools or as unassigned devices, where these features are supported and long-term performance is maintained. Most modern SSDs, including NVMe, work well in these roles.

Using a cache will improve array write performance.

Instead of writing directly to the main storage, data is first sent to a dedicated disk or a group of disks. This data is then moved to the main storage at scheduled times, usually once a day at 3:40 AM. The great thing is that data saved to the cache still appears through your user shares, so you don't have to change how you access your files.

Creating a cache pool helps keep your cached data safe.

Using only one cache device puts your cached data at risk until it is moved to the main array. To protect your data at all times, use multiple devices configured as a cache pool. This setup provides redundancy for cached data, reducing the chance of data loss due to a cache device failure.

SSD cache devices are great for apps and virtual machines.

Using SSDs helps applications and virtual machines (VMs) run faster because they can access data more quickly. SSDs work well in a cache pool, giving you an excellent mix of speed, efficiency, and data security.

Encryption is turned off by default.

If you want to use encryption on your system, you must reformat the disk with an encrypted file system type—this process erases all existing data on the drive. Before enabling encryption, move your data off the disk, change the file system to an encrypted option, format the disk, and then move your data back. For details, see How to encrypt a drive in Unraid.

Keep in mind that using encryption can make it harder to recover data if something goes wrong, so only use it if you really need it.

Disk Recognition and Port Flexibility

Unraid identifies disks based on their serial numbers and sizes, not the specific SATA ports they're connected to. This means you can switch drives between different SATA ports without affecting their assignments in Unraid. This feature is particularly useful for troubleshooting hardware problems, like finding a faulty port or replacing unreliable power or SATA cables.

caution

Your array will not start if you assign or attach more devices than your license key allows.

Start/Stop the array​

When your system starts up, it usually powers up the array of disks automatically. However, if you've recently changed the disk setup, such as adding a new disk, the array will remain off to allow you to check your configuration.

caution

Keep in mind that you'll need to stop the array first to make any adjustments. Stopping it will fully stop all Docker containers and network shares, shut down or hibernate VMs, and your storage devices will be unmounted, making your data and applications inaccessible until you restart the array.

To start or stop the array:

  1. Click on the Main tab.
  2. Navigate to the Array Operation section.
  3. Click Start or Stop. You may need to check the box that says "Yes, I want to do this" before proceeding.

Array operations​

Unraid provides several maintenance and configuration options for your storage array. Key operations include:

Adding disks​

Setting up new disks​

  1. Go to Main → Array Devices in the interface.
  2. Pick the slot where you want to add the disk and select the disk from the dropdown list.
  3. By default, new array drives will be formatted with XFS. If you want to use ZFS or BTRFS instead, select your preferred file system from the drop-down menu.
Hot-Swap Feature

If you're using modern Unraid-compatible hardware, you can change disks without shutting down your server. This feature, known as hot-swap, has been available in LimeTech servers since the beginning.

Clear vs. Pre-Clear​

Unraid requires disks to be in a cleared state - completely filled with zeros and marked with a special signature - before adding them to a parity-protected array. This process ensures parity integrity and keeps the array online during disk preparation.

Clearance is mandatory when adding a data disk to a parity-protected array. If you are adding a parity disk or working with an array that does not use parity, clearance is not required.

The built-in clear operation writes zeros to the disk in the background, allowing the array to remain available. Once the process is complete, the disk must be formatted before use. This method is efficient, requires no third-party tools, and is ideal for quick expansions.

For a more thorough approach, the pre-clear operation – available through plugins like Unassigned Devices Preclear – performs a pre-read to check for bad sectors, zeros the disk, and then verifies integrity with a post-read. This extra step helps detect early drive failures but takes significantly longer and requires manual plugin installation.

Rule of thumb: Use the built-in clear operation for fast additions to your array, or choose pre-clear if you want extensive testing of disk health before putting a drive into service.

Clear vs. Pre-Clear Comparison

ClearPre-Clear
PurposeEnsures basic compatibilityTests the drive and prepares it for use
SpeedWorks quicklyMuch slower process
Array ImpactRuns in the backgroundNeeds the disk to be removed from use
Best ForIdeal for quick expansionsGreat for checking new drives

Clear quickly prepares a disk for use, making it compatible with your system and allowing you to add storage in just a few hours.

Pre-Clear takes longer but provides a thorough test, ensuring a new disk is reliable before it becomes part of your array.

Critical Note

Avoid formatting a pre-cleared disk before including it in the array, as this removes the clearance signature.

Adding data disks​

Adding data disks is a great way to increase your Unraid array's storage space. By adding more disks, you can keep more files, applications, and backups, while still protecting your data.

Important Note:

When adding a new data disk, it must be the same size or smaller than your current parity disks. If you want to use a larger disk, you'll need to replace the parity disks first with the larger one, then use the old parity disks as a data disk.

How to add a data disk

To add a data disk to a parity-protected array:

  1. Stop the array: Use the Unraid WebGUI to stop the array.

  2. Turn off your server: Power down your server safely.

  3. Install the new hard drive(s): Add the new disk(s) physically.

  4. Power the server back on: Start your server again.

  5. Assign the new disk(s): Use the WebGUI to assign the new disks to available data slots.

  6. Start the array: Start the array again.

    • Unraid will automatically start a Clear operation on the new disk. This fills the disk with zeros and ensures it's ready for use while maintaining data integrity.
    • The clear operation takes place in the background, so you can still access your array. However, the new disk won't be usable until the process is complete and the disk is formatted.
    • If you've pre-cleared the disk, Unraid will skip this step, recognizing that it's already ready to use.

  7. After Clearing Completes: Once the clearing process is finished, the disk will show up as unmountable in the WebGUI. You'll be prompted to format it.

    important

    Always double-check the serial number to make sure you're formatting the right disk. Formatting will erase everything on that disk, and this action cannot be undone.

  8. Confirm the Format: Check the confirmation box and click the Format button. Unraid will give you a warning about the outcomes of this action.

  9. Formatting Process: The formatting will take a few minutes. Once it's completed, the disk will show as mounted and ready for use. It's normal to see a small amount of used space due to file system overhead.

note

You can add multiple disks, but they won't be usable until all of them have been cleared and formatted.

Adding parity disks​

Parity disks are a helpful feature that provide an extra layer of protection for your data. They allow your storage system to recover information if a hard drive fails. While using parity disks is optional, it's highly recommended to keep your data safe.

Requirements
  • Single Parity: The disk must be at least the same size as your largest data drive.
  • Dual Parity: Both parity disks must also be at least the size of your largest data drive, but they can be different sizes from each other.

Adding a parity disk is similar to adding a data disk, but there's one important step to keep in mind:

  1. After you select the parity disk and start your storage array, a process called parity calculation begins.
  2. While this process runs, you can still access your files. However, the system might run a bit slower because it's working to calculate the parity.
In-depth: What is parity and why does it matter? - Click to expand/collapse

Parity in Unraid protects your data by ensuring that you can recover from a drive failure. With the use of a parity disk, any single missing disk can be reconstructed by combining the information from all other data drives along with the parity data.

Unraid relies on this real-time parity for two primary purposes:

  • Rebuilding data when a drive fails
  • Correcting errors when a bad sector is found

This feature allows you to keep your data safe, even as drives will eventually fail over time.

In-depth: How does Unraid parity work? - Click to expand/collapse

Unraid works with a special bit called the parity bit, which is stored for each bit position across all data disks and is managed by a dedicated parity disk. The parity is calculated such that the total number of bits in each position across all disks (including the parity disk) is always an even number, a process known as even parity.

Here's how it breaks down:

  • Unraid uses the XOR (exclusive OR) operation for this calculation.
  • For example, if you have four drives, and their 57th bits are 1, 1, 1, 1, the parity bit for that position will be 0 (because 1 + 1 + 1 + 1 + 0 = even).
  • However, if the 57th bits are 1, 0, 0, 0, the parity for that position will be 1 (to ensure the total remains even).

When adding a new drive, Unraid first clears it by writing zeros to all bits. This allows for quick inclusion into the protected array since writing zeros does not interfere with the current parity calculations.

In-depth: How is data reconstructed using parity? - Click to expand/collapse

Unraid can reconstruct a missing drive or repair a bad sector using the following steps:

  • It reads all remaining drives, including the parity disk.
  • It applies even parity to solve for the missing data.

For example, if drive 2 fails:

  • If the related bits are 1, 1, 1 (in addition to the parity), Unraid calculates: 1 + x + 1 + 1 + 0 = even. Here, x must be 1 to maintain an even total.
  • If the bits show 1, x, 0, 0, 1 = even, then x will equal 0.

This parity feature allows Unraid to "simulate" the missing disk. You can continue using the array as if the drive were still functional. Once you install a new drive in the failed slot, Unraid will recreate all the missing data based on the current state of parity and the other disks.

To ensure ongoing protection and effective recovery, it's important to run regular parity checks.

How big does my parity disk need to be? - Click to expand/collapse

Your parity disk must be equal to or larger than your largest data disk. For instance:

  • If your largest data disk is 10TB, your parity disk should be at least 10TB.
  • You can use various other data disk sizes freely, as long as none exceeds the size of the parity disk.
In-depth: How does parity affect write performance? - Click to expand/collapse

Writing to a parity-protected array involves four disk operations for each write: reading data, reading parity, writing data, and writing parity. This process requires a full rotation from each affected drive, so the overall write speed is limited by the slowest drive involved in the operation.

Adding a faster parity disk won't necessarily improve write speeds unless you're writing to multiple slow data disks simultaneously, allowing the parity disk to handle the load. Typically, the write speed is constrained by the slowest data drive being written to.

In-depth: How does Dual Parity work? - Click to expand/collapse

Dual parity enables recovery from two simultaneous disk failures. In Unraid, the second parity disk doesn't simply mirror the first. Instead:

  • Parity 1: Employs standard XOR (even) parity calculations.
  • Parity 2: Utilizes a more complex algorithm (Galois field, rather than Reed-Solomon as in traditional RAID 6), allowing Unraid to rebuild from any two missing disks at the same time. This feature is significant for larger arrays, where the risk of multiple failures increases. Dual parity significantly boosts resilience without the added overhead of mirrored redundancy.
Remember

When setting up a new data storage system, add your data disks first, making sure they all use a compatible file system. After placing your data disks, you can add a parity disk to protect against drive failures.

Once your parity disk is added, remember that any new disk you want to include in the array must be cleared (zeroed) before it's integrated. This is to ensure that the parity remains valid and continues to protect your data.

Upgrading parity disks​

You can upgrade your parity disk device(s) to a larger one(s) to use larger-sized disks in the array or add an additional parity disk.

caution

If you're planning to take the following steps and only have one parity disk, keep these points in mind:

  • Your data will be unprotected until the parity rebuild is complete. If a data drive fails during this time, you could lose the information on that drive.
  • If you already have a data drive that has failed, proceeding with these steps will prevent you from being able to rebuild that drive. In this case, you should follow the Parity Swap procedure instead.

To remove a parity drive:

  1. Stop the array: Use the WebGUI to stop the array.

    tip

    If your server supports hot-swap, you do not need to power down to change disks. You can safely skip steps 2 and 4.

  2. Power down the server: Turn off your server to safely make hardware changes.

  3. Install the new parity disk: Place the new, larger parity disk into your server.

  4. Power up the server: Turn the server back on.

  5. Assign the new disk: Go to the parity slot in your settings and assign the new disk, replacing the old one.

  6. Start the array: Use the WebGUI to start the array again.

Once the array is back up, Unraid will start building parity on the new disk. You can still access your data during this time, but keep in mind that it might be a bit slower until the process is done. Once the parity build is complete, your data is protected again!

Important Tips

  • Keep the old disk: Don't remove the old parity disk until the new parity build finishes. If a data disk fails during the upgrade, having the old disk might help you recover your data.

  • Dual parity users: If you have two parity disks, upgrade one at a time for added safety.

  • Repurpose the old disk: After the upgrade, you can use the old parity disk as a data disk if it still meets your storage needs.

Replacing disks​

You may need to replace disks in your array for two main reasons:

  1. Capacity upgrade: Your storage is nearly full, and you want to use larger disks.
  2. Failure or retirement: A disk has failed or is no longer supported.

While the process for replacing disks is similar in both cases, be cautious, as there is a risk of data loss during the replacement. Parity devices help protect your data; one parity device can safeguard against a single disk failure, while two can protect against losing data if two disks fail. Always be aware of your protection level during disk replacements.

Upgrading capacity​

When upgrading your data drive to a larger one, here are some points to keep in mind:

  • Unmountable disks: Fix any unmountable disks before upgrading. An unmountable status won't be resolved during the rebuild.

  • Single parity risks: If you have single parity, you're at risk of data loss if another drive fails during the upgrade. Seek advice in forums if this happens.

  • Dual parity protection: With dual parity, you're safe from losing data if one drive fails while upgrading a single drive. You can also upgrade two drives at once, but note that this increases the risk since you won't have protection against another drive failing.

  • Backup the old disk: Keep the original disk intact until the upgrade is confirmed successful. This provides a fallback option if anything goes wrong.

Warning

Replacing drives always involves some risk. If another drive fails during the upgrade, especially with only single parity, you may experience data loss. Always check the health of your drives before beginning the process, and ensure that you keep the old drive intact until the upgrade is complete and your data is confirmed to be safe.

To upgrade an existing data disk:

  1. Run a parity check: First, ensure the integrity of your data by running a parity check. Go to Tools → Parity Check and make sure there are zero errors. If parity isn't valid, rebuilding the disk will corrupt its file system.

  2. Stop the array: Next, you'll want to stop the array. Navigate to Main → Array Operation and select Stop.

  3. Unassign the target disk: Once the array is stopped, find the disk you want to replace. Remove it from its slot in Main → Array Devices.

  4. Start the array: Now, start the array again. Unraid will emulate the missing disk using the existing parity and data disks. You have two modes to choose from:

note

Starting the array in Normal Mode at this point is optional. You can use it to verify that the emulated disk mounts and the data looks correct before proceeding, but it is not required for the upgrade process.

  1. Stop the array again: The array should now show that the unassigned disk has failed. Stop the array once more.

  2. Assign the replacement disk: Insert your new, larger disk into the vacant slot in the array.

  3. Start the array to rebuild: Finally, start the array again. Unraid will now rebuild the contents of the emulated disk onto the new disk. The file system will automatically adjust to take full advantage of the larger disk's capacity.

Replacing failed/disabled disks​

What is a failed/disabled disk?

A failed or disabled disk is one that Unraid has stopped using for writing data, usually because it encountered a write error. Keep in mind these don't always mean the drive itself is broken; sometimes, it might be due to bad connections, power issues, or just a temporary glitch.

Some indications include:

Red "X" indicates write error

This red "X" shows which disk needs replacement.


Failure notification alert

Keep an eye out for these alerts.


Click to see how different parity configurations handle disk failures
Failure ScenarioNo ParitySingle ParityDual Parity
1 Disk FailureData LostRebuildableRebuildable
2 Disk FailuresData LostData LostRebuildable
Tips for Safe Rebuilds
  • For Single Parity: You can only replace one disk at a time. If another disk fails during the rebuild, you could lose data.
  • For Dual Parity: You can replace one or two disks at the same time, but be cautious if more than two disks fail.
important

If you experience a situation where more disks fail than your parity can protect:

  1. Immediately halt all write operations to prevent further data loss.
  2. Seek help by posting your issue in the Unraid forums along with diagnostics.
  3. Do not attempt a rebuild until you've received advice - focus on salvaging your data first.
How to diagnose the issue

To diagnose the issue, start by checking the syslog and SMART reports:

  • If the syslog shows that the drive has reset, there might be a problem with the connection or cables.
  • SMART reports can help assess the drive's health, but the best way to check is by running a SMART extended test. If it completes without errors, the drive is likely fine.
  • If you see CRC errors, it usually points to issues with the cabling. Keep an eye on these errors, as they accumulate over time and don't reset.
tip

To stay informed about drive issues, enable notifications in Unraid. This way, you are alerted right away if something goes wrong, reducing the risk to your data.

What is emulation?

When a drive is disabled, Unraid will use its parity and other working drives to emulate the failed drive. This means your system continues to operate, and you can still access the data stored on the emulated drive:

  • Unraid will stop writing to the physical drive, and any updates will be saved in parity and the emulation instead.
  • Before replacing the failed drive, you can check and recover data from the emulated drive, which keeps the physical drive intact for potential recovery.
  • If you can't access the emulated drive, repairing the file system is essential before any rebuild since fixing the file system is quicker and more effective than a rebuild.
Preparing for drive replacement

When you need to replace a drive, make sure the new drive is at least as large as the failed drive but not larger than your smallest parity drive. If you have to use a larger drive, follow the Parity Swap procedure.

While it's not mandatory, many users choose to pre-clear new drives to test them and help prevent early failure. You can do this with the Preclear plugin, a SMART extended test, or manufacturer tools.

caution

If you try to rebuild using an unmountable emulated drive, the new drive will also become unmountable. Always repair the file system before attempting a rebuild.

If a disk in your Unraid system has failed or is disabled, don't worry! You can replace it and recover your data. Just follow these simple steps. Remember, you need a new disk that is the same size or larger than the one you're replacing, but it can't be larger than your smallest parity disk.

important

Before you start, make sure to check for any disk errors or warnings. Unraid must be able to read all the remaining disks without any issues for the rebuild to work correctly. If another disk is failing, it could lead to data loss.

To replace and rebuild a disk:

  1. Stop the array:

    • Log in to the Unraid WebGUI (the web interface).
    • Find the option to stop the array. This is necessary before you can change any disks.
    tip

    If your server supports hot-swap, you can skip the next step and just stop the array.

  2. Power down your server (only if not using hot-swap):

    • If your server doesn't support hot-swap, you'll need to shut it down completely.

  3. Replace the failed disk:

    • Remove the old, failed disk from your server.
    • Insert the new disk. Just remember, it has to be at least the same size as the old disk but no bigger than your smallest parity disk.

  4. Power up the server (if you powered it down):

    • Turn your server back on if you shut it down.

  5. Assign the new disk:

    • Go back to the Unraid WebGUI.
    • Find the slot for the failed disk and assign your new disk to that slot.

  6. Confirm your action:

    • You'll see a confirmation box. Check the box that says Yes, I want to do this and confirm.

  7. (Optional) Choose Maintenance Mode:

    • You can select Maintenance Mode, which can make the rebuild process faster. However, during this time, you won't be able to access the array at all.
    • If you choose Maintenance Mode, make sure to click Sync to start the rebuild.

  8. Start the rebuild:

    • Click Start to begin the process. Unraid will copy your data from the emulated disk to the new disk.
    • If your new disk is larger, Unraid will manage the extra space for you.
warning

If Unraid prompts you to format the new disk during the rebuild, do not do it. Formatting will wipe all data and make recovery impossible.

What to expect during the rebuild

  • The array will still be available to use during the rebuilding process (unless you're in Maintenance Mode), but it might run slower.
  • Rebuilding can take several hours, depending on your disk sizes and system usage.
  • The new disk will have the same file system as the original.
  • If the old disk was unmountable due to file system issues, the new disk will also not be mountable. Please fix any file system issues before starting the rebuild.
Additional notes
  • The rebuild process won't change the file format of the disk; it simply restores its previous form.
  • Always check the health of every disk before you start a rebuild to avoid any potential issues.

Parity swap​

A parity swap is a special procedure in Unraid used when you need to replace a data disk with a disk that is larger than your current parity disk. The process moves your current parity disk to the data slot, then installs a new, larger disk as the new parity disk. This ensures your array remains protected and allows for larger data drives in the future.

Example scenario

You have an Unraid setup with a 2TB parity drive and a 1TB data drive and want to replace the 1TB drive with a 4TB drive.

First, assign the new 4TB drive as the parity drive, which will replace the 2TB drive. Then move the original 2TB drive to the data slot, and remove the 1TB drive entirely.

After these changes, you'll have a 4TB drive as your new parity, ensuring you can add future data drives up to 4TB. The 2TB drive will now hold your existing data, while the 1TB drive can be repurposed.

This swap keeps your data secure and your array protected, allowing for future upgrades.

Prerequisites
  • Before starting, ensure the data drive you want to replace is disabled. If the drive has failed (shows a red indicator), it is already disabled. If the drive is healthy but you want to replace it, unassign the drive and start the array once without it to force Unraid to mark it as disabled.
  • If your replacement data drive is not larger than your parity drive, use the standard Replacing a Data Drive procedure instead.
  • This procedure is only needed for replacing data drives in an Unraid array with a disk larger than the current parity drive. If you only need to upgrade your parity drive, simply remove the old parity drive, add the new one, and start the array. Parity will rebuild automatically.
Warnings
  • Always verify the health of all drives using SMART reports before starting a %%parity swap|parity-swap%. Attempting this procedure with another failing or unhealthy disk increases the risk of data loss.
  • Preclear the new disk if possible. While not required, preclearing stress-tests the drive and reduces the risk of early failure.
  • Correctly identify all drives before beginning. Note the model numbers and last four characters of each serial number to avoid mistakes during assignment.

To carry out a parity swap:

note

If the drive to be replaced is already disabled (failed), you may not need to perform steps 1–4. If you have already installed the new replacement drive (for example, after preclearing), you can skip steps 5–8.

  1. Stop the array if it is running.
  2. Unassign the old data drive if it is still assigned. If the drive was previously healthy, you may see error notifications for a missing drive - this is expected.
  3. Start the array. If prompted, check the box confirming your action. The data drive should now show as "Not installed."
  4. Stop the array again. At this point, the array treats the drive as failed.
  5. Power down the server.
    tip

    If your system supports hot-swap, you do not need to power down to remove or install disks. Just make sure the array is stopped before making any hardware changes.

  6. (Optional) Remove the old drive. You may wish to keep it installed for testing or reassignment.
  7. Install the new drive. Pre-clearing is strongly recommended, but formatting is not needed.
  8. Power on the server.
  9. Stop the array if it started automatically. If you see a message about retrying to unmount disk shares, disable Docker and/or VM services in Settings, reboot, and try again.
  10. Unassign the parity drive.
  11. Assign the new drive to the parity slot. You may receive error notifications - this is normal.
  12. Assign the old parity drive to the data slot of the drive being replaced. Both the parity and replacement data drives should now display blue status indicators.
  13. Go to Main → Array Operation. You should see a Copy button and a message stating "Copy will copy the parity information to the new parity disk."
  14. Confirm and start the copy process. Check the confirmation box and click Copy. The array will not be available during this operation.
    • The copy process can take many hours, depending on disk size. When finished, the array will be stopped and ready for a data rebuild.
  15. Start the array to begin the data rebuild.
    • Confirm your action if prompted. The array is now started, and data is reconstructed onto the new data drive.
    • You can use the array during the rebuild, but for best performance, limit usage.
    • The rebuild process will also take several hours.
warning

Never format a drive during this process. Formatting will erase all data and update parity, making recovery impossible.

After completion, you will have a larger parity disk and a replaced data disk. Many users run a parity check afterward for extra confidence, though this is optional.

Removing disks​

There may come a time when you want to remove a disk from your Unraid array. Whether you're looking to save on power, retire an old or unreliable drive, or repurpose hardware for a different use, the process is fairly straightforward.

Removing parity disks​

If you find that you no longer need the level of parity protection in your setup, you can remove a parity disk whenever you want.

To remove a parity drives:

  1. Stop the array: Begin by accessing the WebGUI and stopping the array.
  2. Unassign the parity drives: Locate the parity drives slot and set it to "Unassigned."
  3. Start the array: Finally, restart the array to commit the changes and successfully remove the parity drives from the configuration.
warning

Before you make any changes, please keep this in mind: If you already have any failed data drives in your array, removing a parity disk can decrease the number of failures Unraid can handle without risking data loss.

  • With dual parity, you can recover from a single failed drive, but be aware that you won't be able to survive another failure during the rebuild process.
  • With single parity, you lose all redundancy, meaning that any additional drive failures could result in data loss.

Removing data disks​

Removing a data disk is a straightforward process, but it's important to remember that you'll need to perform a parity sync afterwards. Until this sync is complete, there's a risk of data loss if another disk fails.

You can choose from two methods to remove a data disk:

  1. Standard method - Recommended for most users.
  2. Parity-preserve method - Suitable for advanced users who want to maintain parity during the removal.

This method is ideal for users looking for a simple way to remove a data disk.

  1. Stop the array: Begin by stopping the array to safely remove the disk.
  2. (Optional) Document Your disk assignments: For your reference, you may want to take a screenshot of your current disk assignments from the Main tab.
  3. Reset the array configuration: Use the Reset the array configuration procedure. Be sure to choose the option to preserve all current assignments.
  4. Unassign the disk: Identify and unassign the disk you wish to remove.
  5. Start the array: After unassigning the disk, start the array again, but do not check the "Parity is valid" box.

Once you start the array, a parity sync will initiate if you have a parity disk assigned. Keep in mind that the array will be vulnerable to data loss until this sync completes.

tip

Move any important data off the disk before removal. The parity sync process may take several hours, depending on disk size and system activity.

Checking array devices​

Regularly checking your Unraid array is crucial for maintaining data integrity and identifying potential issues before they result in data loss. You can initiate a check using the Check button under Array Operations. Depending on your array's configuration, this button allows you to conduct either a parity check or a read check.

The check button lets you perform parity and read checks

For convenience, you can schedule these checks to run automatically at intervals that suit you by navigating to Settings → Scheduler. It's advisable to perform automated, non-correcting checks on a monthly or quarterly basis to ensure the ongoing health of your data.

Parity & Read checks​

A parity check is essential when your array includes at least one assigned parity device. During this process, all data disks are read simultaneously. The system recalculates the parity and compares it to the information stored on your parity disk(s).

In-depth: Parity sync vs. Parity check - What's the difference? - Click to expand/collapse

A parity sync reads all data disks and creates parity from scratch, saving the results on the parity drives. This process typically happens when you add or replace a parity drives or after significant configuration changes.

A parity check reads all data disks along with the parity drives. It recalculates the parity values and compares them to the stored values. There are two modes for running a parity check:

  • Correcting (CORRECT): If it finds a mismatch, it updates the parity disk and logs the event.
  • Non-correcting (NOCORRECT): Only logs any errors found without updating the parity.

To avoid excessive logging, only the first 100 addresses with errors are reported during a parity check. These mismatches are referred to as sync errors and indicate where the parity does not align with the data, typically due to issues such as sudden power losses or hardware problems. Each sync error is counted in 4KiB blocks - this is the system's I/O unit size (known as the Linux page size).

In-depth: What does "valid" mean for parity and array disks? - Click to expand/collapse

A valid parity disk indicates that after the last successful parity sync, Unraid recognized the parity as "good," meaning it's suitable for data recovery if a disk fails. The "valid" status applies to every disk in the array, not just the parity one. If all disks except one are valid, Unraid can reconstruct the missing or failed disk's data using parity.

Unraid maintains the parity disk's "valid" status even when some sync errors occur following a non-correcting check. This ensures that you can still recover a failed disk. If the parity were marked as invalid, you would risk having two invalid disks after just one additional failure, leading to potential data loss.

While a parity check is in progress, you can still use the array. However, be aware that performance may slow down due to drive contention, as both file operations and the check will compete for resources.

Normally, if the parity check identifies errors, Unraid will update the parity disk to align with the computed data and increment the Sync Errors counter.

tip

If you prefer to run a check without making corrections, simply uncheck Write corrections to parity before starting. This way, errors will be reported but not fixed.

important

A correcting parity check is automatically initiated after an "Unsafe Shutdown" - this occurs when the array is restarted without being properly stopped. The most frequent cause of such an event is unexpected power loss, which can leave write operations incomplete. Using an uninterruptible power supply (UPS) is highly recommended to avoid this scenario. This measure helps maintain the integrity of your data and prevents potential issues.

What are parity errors?

Parity errors happen when the parity information your system calculates doesn't match what's stored on your parity drives. This mismatch can arise from several issues, including:

  • Sudden power loss or unsafe shutdowns
  • Failing drives or disk errors
  • Problems with cables or connections
  • Unclean shutdowns or unexpected system crashes

When parity errors occur, either the parity drives or the data disks could be out of sync. Ideally, a parity check should report zero errors so everything functions correctly.

What to do if you encounter parity errors

If your parity check reveals errors:

  1. Investigate the problem: Start by checking SMART reports, which provide detailed information about your drives' health. Look for any signs of disk or connection problems.

  2. Seek help from our forums: If you cannot resolve the issue on your own, consider reaching out for advice on the Unraid forums. The community can offer valuable insights and suggestions based on their experiences.

  3. Run another check: After you've addressed any hardware issues, running another parity check is a good idea to ensure everything is back in order.

Rebuild Time

Remember that parity checks can take several hours, depending on the size of your disks and the activity level of your system. For the best results, schedule these checks during times of low system usage.

Check history​

Whenever the system performs a parity check or a read check, it keeps a record of what happened. You can view these details easily by clicking the History button found under Array Operations in the interface.

For those who want to dive deeper, all these records are saved in a text file located in the config directory on your Unraid USB flash device.

Spinning disks down or up​

Unraid allows you to control the power states of your hard drives. You can easily spin them up or down, and manage SSDs to be active or in standby. This helps save energy, extend the life of your drives, and reduce noise when the disks are not being used.

Why spin down or up?

  • Spin down: If you have drives that aren't used often, spinning them down can save energy and help them last longer.
  • Spin up: If you know you'll need files soon, spinning up the disks ahead of time can cut down wait times.

How to control spin states

You can control disk spin states through the Main tab in the WebGUI:

  1. Go to the Main tab and locate your array devices
  2. Find the spin control buttons - each disk will show either:
    • Click the 🔘 button to Spin Up if the disk is currently spun down
    • Click the 🟢 button to Spin Down if the disk is currently spinning
  3. Click the appropriate button for the action you want:
    • Click Spin Down 🟢 to power down an idle disk and save energy
    • Click Spin Up 🔘 to power up a disk that's currently spun down

Remember that if a disk is being accessed (like if you're opening a file), it will stay active and ignore any spin-down request.

When a disk is spun down, its temperature won't show in the WebGUI. However, once any application or user accesses it, it will automatically spin up.

tip

Use the spin controls to save power and reduce wear on your drives. Remember that disks that are actively being used will stay on until all tasks are finished.

Reset the array configuration​

Resetting your array configuration is an important step that should be undertaken carefully. This process is usually necessary when removing a disk, starting fresh with a new array layout, or fixing disk assignment issues. Please note that this action can impact data protection and parity, so ensure you only proceed when truly needed.

Common reasons to reset your array include:

  • Removing or replacing disks: If you need to take out or swap out any disks in your array.
  • Starting anew: When you're looking to create a brand new layout for your array.
  • Fixing disk assignment errors: To correct any issues with how disks are currently assigned.
  • Recovering from configuration problems: When facing challenges with your existing setup.

Resetting your disk configuration can be done via the new config page

To reset your array configuration:

  1. Go to the Tools page and click on New Config.
  2. You can keep some existing disk assignments if you only wish to make minor adjustments. This can save time and minimize the chance of errors.
  3. Confirm your choice by checking the box, then click Apply.
  4. Head back to the Main tab. After applying the changes, your configuration will have been reset.
  5. Make any necessary changes to your configuration by assigning or unassigning disks as appropriate.
  6. Launch the array in Normal or Maintenance Mode to finalize your updates.
important
  • Data preservation: Unraid will attempt to recognize previously used drives and preserve data where possible.
  • Impact on parity: Removing a data drive will always invalidate parity unless that drive was zeroed before removal.
  • Disk reordering: Changing the order of disks won't affect parity1, but it can invalidate parity2.
caution

When you see the Start button, there is a checkbox labeled Parity is Valid. Only check this box if you are certain it is correct or if an experienced Unraid user has advised you to do so during recovery. Incorrectly checking this option can lead to data loss.

Do not use New Config if your goal is to rebuild a disk. Performing a New Config clears the array history required for a rebuild, and Unraid will not offer to rebuild the disk afterward. Follow the disk rebuild procedure instead.

Undoing a reset​

If you find that you need to reverse a reset:

  1. Access your flash device over the network (SMB).
  2. Locate and open the config folder.
  3. Rename the file super.old to super.dat.
  4. Reboot your server, and your prior array configuration should be restored.

Status reports​

Unraid provides status reports that help you keep track of the health of your storage array. These reports are a quick way to check if any of your disks are disabled or having issues with reading or writing data.

  • Current status: Status reports show the current condition of your array. It's important to note that this information resets after you restart your system, so that it won't keep a history of past issues.

  • No historical data: If you want to see what has happened before a reboot, you'll need to look elsewhere, as these reports don't save past states.

important

Remember that the status reports don't include SMART data. SMART reports give you a more detailed view of individual disk health. So, even if your status report shows everything is fine, checking the SMART reports regularly is still a good idea to catch any potential problems early.

Array write modes​

Unraid provides various write modes for managing array operations, each with its own pros and cons regarding speed, power consumption, and drive wear. Knowing how these modes work, along with the role of a cache drive or pool, can help you fine-tune your server to best suit your needs.

Write modes at a glance

Write ModeSpeed (Typical)Power usageWhen drives spin upData protectionBest use case
Read/Modify/Write20–40 MB/sLowOnly parity and target driveYesMost workloads, energy savings, small writes
Turbo Write (Reconstruct)40–120 MB/sHighAll drivesYesLarge file transfers, array rebuilds, parity checks
Cache Write (SSD/NVMe)50–110 MB/s (SSD), 250–900 MB/s (NVMe)*VariesCache drives onlyNo (until moved)Apps, VMs, frequent writes, maximizing speed

Read/Modify/Write

Click to expand/collapse

How it works

This mode reads the existing data and the parity, calculates the new parity, and then writes the updated data. Only the parity drive and the target data drive spin up, which results in lower power usage and less wear on the drives. However, it can be slower due to the additional read/write cycles.

When to use

  • Anytime, especially if you want energy savings and idle drives to spin down.
  • Great for small or infrequent writes.

Turbo Write (Reconstruct write)

Click to expand/collapse

Turbo write, also known as reconstruct write, is a feature designed to boost the writing speed of your Unraid array. It works by reading all data drives and updating the parity simultaneously. This process eliminates the delays caused by waiting for the platters to rotate, as seen in the default write mode. However, it's important to note that all array drives need to be spinning and functioning properly for this to work effectively.

How it works

  • When you write new data, Unraid reads from all the other data drives and recalculates the parity at the same time. Both the new data and the updated parity get written together.
  • All drives in the array must be operational and actively spinning.
  • This method significantly enhances write speeds compared to the default writing mode.

When to use Turbo Write

  • Utilize turbo write when transferring large, sequential files to the array.
  • It's effective during array rebuilds or parity checks, as all drives will already be spinning.
  • This mode is very useful when minimizing write time is a priority and you can confirm all drives are healthy.

When to avoid Turbo Write

  • Avoid using this mode if you want your drives to spin down during idle times to conserve energy.
  • Turbo write isn't ideal for small or infrequent write operations since it causes all drives to spin up for every write, increasing power usage and wear on the drives.
  • If you suspect that any drive is failing or is unreliable, it's best to steer clear of turbo write since it relies on every drive being operational.
info

Turbo write is best suited for bulk operations and scenarios requiring high throughput. However, it might not be the best choice for everyday tasks if you're focused on energy savings and minimizing drive wear.

Cache Write

Click to expand/collapse

How it works

Data is written first to a fast SSD or NVMe cache and can then be moved to the main array later by a process called the Mover. The speeds vary:

  • SSD: 50–110 MB/s
  • NVMe: 250–900 MB/s (which can utilize 10GbE networks)

Once data is moved to the array, it is protected by parity.

When to use Cache Write

  • For shares with frequent write operations like applications, virtual machines, or downloads.
  • To enhance performance and reduce any perceived write latency.

Performance Expectations

  • Without a cache drive: Average 20–30 MB/s, with peaks up to 40 MB/s.
  • With SSD cache: 50–110 MB/s.
  • With NVMe cache: 250–900 MB/s depending on network or drive constraints.
tip

Consider using a cache pool (multiple devices) for added redundancy and data protection before the Mover runs.

Automated Solutions
  • The Auto mode (a future feature) will engage turbo write only when all drives are already spinning.
  • Community plugins (search for "Turbo Write" in the Apps tab) may offer enhanced automation or scheduling options.

To change Write Mode:

  1. Navigate to Settings → Disk Settings.
  2. Locate Tunable (md_write_method).
  3. Choose your preferred mode:
    • Read/Modify/Write (default)
    • Reconstruct Write (Turbo Write)
    • Auto (future feature)
  4. Click Apply to confirm your choice.
Quick recap
  • Use Turbo Write when you need speed, but be aware of increased power consumption and drive spin-up.
  • Utilize Cache Write for optimal performance, particularly with SSD or NVMe drives.
  • For most users, the default write mode offers the best balance unless you specifically need higher speeds.

Read modes​

When using Unraid, the speed at which you can read files is mainly determined by the individual drive that holds each file. Unlike traditional RAID systems, which combine multiple drives to improve performance, Unraid stores each file on a single disk. This means read speeds won't be boosted by the combined speeds of multiple drives.

Performance expectations

  • Typical single HDD: 70–250 MB/s (depends on drive model, age, and data location)
  • Typical SATA SSD: 400–550 MB/s
  • NVMe SSD (in a pool): 250–7,000 MB/s (PCIe generation and network/PCIe limitations apply; e.g., 10GbE network caps at ~1,100 MB/s)
Special cases
  • If a disk is disabled and its data is being reconstructed, Unraid will use the remaining drives along with parity information to recreate the data. During this process, the read speed may slow down to 30–60 MB/s or even lower, depending on the slowest drive in your system.
  • Any ongoing operations in the array, such as a parity check or rebuilding a drive, can also affect read performance. This is due to increased movement of the drive heads and overall contention for resources.

Cache pools​

Cache pools in Unraid provide significant advantages, particularly for write-heavy tasks, virtual machines (VMs), and Docker containers. These pools operate separately from the main array and can be set up with multiple drives using either the BTRFS or ZFS file system, supporting various RAID configurations for speed and data protection.

Cache pools vs. the main array

FeatureCache pool (BTRFS)Cache pool (ZFS)Main array (Unraid)
Read speedSSD: 400–550 MB/s, NVMe: 250–7,000 MB/s*SSD: 400–550 MB/s, NVMe: 250–7,000 MB/s*HDD: 70–250 MB/s (per disk)
Write speedSSD: 400–550 MB/s, NVMe: 250–7,000 MB/s*SSD: 400–550 MB/s, NVMe: 250–7,000 MB/s*20–120 MB/s (parity mode dependent)
Data protectionRAID 1/RAID 10; RAID 5/RAID 6 (experimental, not for critical data)RAID 1/RAID 10; RAIDZ1/RAIDZ2/RAIDZ3 (stable, production-ready)Parity-based, file system agnostic
ExpansionMix drive sizes; add/remove devices dynamicallyLimited add/remove device support; cannot remove from RAIDZ; single-device add to expand single-vdev RAIDZ in Unraid 7.2Add drives, but no striping or performance scaling
Recovery complexityHigher risk of data loss; BTRFS tools requiredHigher risk of data loss; ZFS tools requiredEasier parity-based rebuilds
Best forApps, VMs, frequent writesApps, VMs, frequent writes, enterprise workloadsBulk storage, media libraries

*Actual NVMe speeds depend on PCIe generation, cooling, and network bandwidth (e.g., 10GbE caps at ~1,100 MB/s).

Pros of cache pools

  • Higher performance: NVMe pools can saturate 10GbE/40GbE networks (1,100–3,500 MB/s).
  • Flexible RAID: Both BTRFS and ZFS support RAID 1/RAID 10 for redundancy without matching drive sizes.
  • Low latency: Ideal for databases, VMs, and Docker containers.
  • ZFS advantages: ZFS provides enterprise-grade features like data integrity checking, compression, and snapshots.

Cons of cache pools

  • No parity protection: Data is unprotected until moved to the array.
  • Recovery risks: BTRFS RAID 5/RAID 6 is unstable; single-drive pools lack redundancy.
  • ZFS considerations: ZFS requires more RAM and has stricter hardware requirements than BTRFS.

For more detailed information about cache pools, including how to set them up, manage them, and advanced features, check the Cache pools page.

Troubleshooting​

Troubleshooting array start failures​

If your array won't start, follow these steps to identify and fix common problems. Look for error messages under Main → Array Operation.

Message:
Too many wrong and/or missing disks!

Single parity error

With one parity drive, you can only have one missing disk. With two parity drives, two disks can be missing and you can still start the array, and so on. Parity helps until you can replace the missing disk.

What to Do:
Replace the missing disk. For dual-parity configurations, replace the missing disks one at a time. If you can't recover the data (or if more than 2 disks fail in a dual-parity setup), go to Tools → New Config to perform the New Config procedure.

Disk failure during a rebuild​

If a second disk fails while you're rebuilding another one, what you can do will depend on your parity setup.

Single parity disk

If one disk fails during the rebuild of another, the rebuild will stop because the data can't be accurately restored. Unfortunately, you won't be able to recover your data in this situation.

Dual parity disk

If you have two parity drives, you have more options:

  • You can wait for the first rebuild to finish and then deal with the second failed disk.
  • Or, you can stop the current rebuild, replace the second failed disk, and then start the array to rebuild both disks at the same time.

If the first rebuild is almost done, it's usually better to let it finish. If it just started, it might be faster to rebuild both together.

warning

Rebuilding disks puts a lot of stress on all drives, which increases the chance of new failures. Always check drive health using SMART reports before starting a rebuild.

Rebuild Time

Be prepared for the rebuild process to take several hours. The time can vary based on disk size and how busy your system is. Larger disks and busy systems may take longer.