From Unraid | Docs
Jump to: navigation, search

If you are running any release of unRAID v4 or v5, then you want the older documentation, please go to FAQ for unRAID v4 and v5.

Important Note and Warning: this FAQ is being updated for unRAID v6, but some information is still out-of-date and only applies to v4 or v5!

See also

Other unRAID related FAQ's

  • FAQ for unRAID v6 - a smaller version on the forum, with some of the more common questions
  • Docker FAQ
  • We're hopeful there will be more FAQ's for unRAID-related addons. To FAQ maintainers - please see this thread.

The unRAID FAQ is an attempt to answer many of the common questions that current and prospective users have about the unRAID system. We hope to see it grow in coverage and detail, beyond the current limited set of questions and answers, and the common use of links to answers rather than a full answer. The process of adding questions and answers involves first adding a question with a link to a relevant forum thread, then hoping for someone to take the time to add a full writeup. This FAQ is a work in progress and will evolve as unRAID changes. It is nearly entirely written by users of unRAID. It is not an official document written by Lime-technology LLC.

With apologies, this page is under construction, and the included info is still rather meager and dated, but with time and your help, it will improve.


General Questions about unRAID

What is unRAID?

What is unRAID?
Demo of unRAID - on YouTube
Another overview
Technology and Uses - a more detailed introduction

What are the advantages of unRAID over similar products?

No striping, so safer for multiple drive failures; parity protection, so resistant to drive failures; flexible, can mix and match drive brands, sizes, and types, easily add additional drives, etc...
Each drive is an independent file system. An unRAID array can be thought of as a parity protected JBOD. If the array fails, the individual drives are still accessible, unlike traditional RAID arrays. Spin down can be controlled per drive, so a drive with rarely accessed files may stay off (spun down) for months, saving power costs, and possibly increasing its life. With standard striped RAID arrays, the entire array must be spun up or down, so generally stays spun up.
Some old threads (from single parity days) about advantages/disadvantages of unRAID and similar products:
More old threads about the advantages of unRAID in the unlikely event of multiple drive failures:

What are the disadvantages of unRAID compared to similar products?

No striping. So, although it performs well, better than many NAS solutions, it's generally slower than a RAID 0, RAID 5, RAID 6, or RAID 10, etc. If performance is very important for an array you may be building, then one of those 4 RAID types will be a better choice than unRAID.

Is unRAID for me?

We unRAID users think so, of course. But read for yourself a few old forum threads:

Minimum System Requirements?

  • See Hardware requirements
  • See Hardware recommendations
  • Linux compatible motherboard and components (almost all are if they use Intel or AMD processors)
  • 64 bit CPU (almost all modern Intel and AMD processors are)
  • 1GB of RAM (recommend 2GB or more depending on usage)
  • Compatible USB flash drive with unique ID (most are but a few aren't)
  • Can operate without video, monitor, keyboard, or mouse
  • See Hardware Compatibility for extensive hardware and compatibility information, however motherboard section is obsolete!

How is unRAID licensed?

The official information is found here.
The license is controlled by a key file, which is tied to a unique GUID (Globally Unique IDentifier) associated with your unRAID flash drive. The unRAID software is identical with each license, only the features and drives are limited with the Plus and Trial versions.
Apart from the limit on the maximum number of drives, the array is the same under all licenses. No conversion of data or software is necessary.

What is the difference between the unRAID licenses?

The official information is found here.

Can I split a two pack license with a friend/stranger?

Two pack licenses are no longer sold, as of May 16, 2015. Prior to that, they were available at a discount to allow users to create a backup unRAID, a second USB flash drive in case the first one failed. That is no longer necessary, because replacement keys are quickly available online, from within the software.

How much usable space will I get from an unRAID array?

Parity information resides only on the parity drive, the entire drive, so only the size of the parity drives are unavailable for use.

How fast is unRAID?

This is a very hard question to answer, as it depends mostly on your hardware choices, your networking choices, and your configuration choices. unRAID is based on a lean but up-to-date Slackware distribution of Linux, ideal for file servers, so it should provide the best possible speed that your hardware can do. It does not take advantage of hardware RAID cards, and it does not use RAID striping, and therefore cannot boast the performance gains associated with those features. Please see the earlier questions comparing unRAID with other types of RAID.
Users and reviewers have generally found a NAS based on unRAID to be significantly faster than many competitive NAS boxes, and to be very similar to the faster NAS implementations that are not based on high performance RAID cards or RAID types. (this needs citations!)
Because the emphasis is on data security through parity protection, writes to parity-protected drives can take much longer than reads. When not in turbo-write mode (all drives spun up), reads are passed through as one I/O, but writes require four I/O's, because both data drive and the parity drive must be read first, then both drives written to, which requires a full drive platter rotation. For very large files (a prime use of unRAID servers), drive buffering is useless.
There are many factors that affect speed numbers, and many opinions about them! Please see the unRAID forums for discussions about unRAID performance (such as here (old)). The old v4 and v5 Improving unRAID Performance page may have tips and ideas for maximizing performance.

How does parity work?

In general, a parity process is designed to detect a single bit change across a given set of bits, by setting the value of an added bit such that a summation across that set of bits is forced to a known value. The added bit is known as a parity bit.
In unRAID, the parity bits are stored on the primary parity drive independent of the data drives. This parity bit works across the set of bits in the same relative bit position on each drive. So the 57th bit of the parity drive is the parity bit for the 57th bit of all of the data drives. A parity check of the 57th bit position is therefore a summation of all of the 57th bits of every drive including the parity drive, and checking to see if the summation is an EVEN number. Why an EVEN number? unRAID uses what is called 'even parity', which simply means that the summation process (using XOR internally) across that set of bits must return a value that is an EVEN number. If the value returned is not EVEN, then the parity bit is toggled so that the parity check WILL return an EVEN number.
When a drive is added to a parity protected unRAID array, it is first cleared, by writing zeroes to all bits of all sectors of that drive. That way, the parity drive does not have to be changed, even though you are adding all of the bits of the new drive to the parity info, because the value of the bit at every position on the added drive is zero, and of course adding zero to an EVEN number equals an EVEN number.
For more explanations and discussion (some very old), see:

If a drive fails, will I have full access to my unRAID array?

Yes, BUT ...
If a drive fails or is missing, you will still have full use of the entire array, including the contents of the failed or missing drive, just as with any similar RAID system. You can copy files from all drives including the missing one, AND you can even copy files TO the missing drive, although performance will be considerably slower. The unRAID system is designed to reconstruct all of your data and the file systems, IF the parity info is being fully maintained with all of your data drives.
HOWEVER, any time you are running without access to the failed or missing drive, any time you are running without all green status balls, you are UNPROTECTED against further drive failure. If a second drive fails, you will probably lose the contents of BOTH failed or missing drives! It is strongly recommended to resolve the problem with the first drive as soon as possible.
As a cautionary tale, please see "Stopped. Invalid configuration. Too many wrong and/or missing disks!" .

What does reconstruct mean?

Reconstruct is a term that refers to rebuilding the data contents of a disk using the parity information in a RAID system. By reading the parity disk plus all of the other data disks, the system can regenerate the data of the target disk.

How hard is unRAID to use if I don't know Linux?

Not hard at all. Although the unRAID server software is based on a slimmed-down Linux, it is managed almost entirely from your web browser, typically on a Windows or Mac computer. Some users are happy with that, but many want to take advantage of the many tweaks and addons for unRAID, and these usually require a little hands-on work. But they are completely optional, and are generally accompanied with lots of help and instructions, and there is a very helpful user community. Many users find this makes for a good introduction to Linux, at their own pace. See also this thread, especially this post, for more comments.

Does unRAID need Internet access?

The unRAID server software generally does not require Internet access. Of course, you will need Internet access from another desktop to download the software and software updates, and to read the unRAID forums and this Wiki!
However, there are many benefits to providing Internet access to your unRAID server. The unRAID software and your plugins and Docker containers can all be updated from within the software. Usability and manageability are improved with email and other notifications. unRAID supports NTP, the Internet time service, so if you enable the NTP service, your server will keep accurate time. In addition, expert unRAID users have created many addons for unRAID, such as plugins, Dockers, and VM's, that can provide numerous application services such as torrent support, etc. All of these benefits are completely optional. See also this.

How to get Help

Where can I go to find help?

  • First, see Need help? Read me first!
  • The other main starting point for help is the Troubleshooting page. You will find basic troubleshooting help, and get specific help with boot and flash drive and hard drive issues, and tips on where to get more help.
  • Search the Table of Contents above, for questions that seem similar to yours.
  • Read the UnRAID Manual 6.
  • Search the unRAID forums for similar problems.
  • The UnRAID Wiki contains the full unRAID Manual, and a wealth of User Contributed material.
  • Lastly, register on the unRAID forums, and post about your problem. See the Need help? Read me first! page for help with that.

Basic Start Up Questions

How do I get started?

Glad you asked! Check out the Getting Started page.
Also check out the Getting Started with unRAID page. It's an older guide but provides an organized list of many sources of information, as well as some step by step guides, such as the Configuration Tutorial page and the Building an unRAID Server page (old).

How do you 'manage' an unRAID server?

While you can do many things from the unRAID server console, using Linux commands (guided by other users or the Wiki!), almost everything necessary to manage and control your unRAID server and drive array can be done from the unRAID Web Management Utility. The Web Management Utility is a set of standard web pages that you access from any web browser on any of your networked machines. From the various Utility pages, you can see the status of the array and each of your drives, make changes to the configuration of drives and shares and users and many other settings, stop and start the array, and run parity checks and other server operations.

How do I access the Web Management Utility?

The default name for an unRAID server is tower, so the URL you would browse to, in your favorite web browser, is simply:
--- on Mac stations, you may have to use ---
Of course, if you have changed the name of your unRAID server, then you would substitute your server name for tower.
Note that the slashes preceding the server name are forward slashes, not backslashes. If you try to browse to \\tower, you will either get an error returned, or you will get a web page of files and folders, similar to Windows Explorer.

Why does //tower not work in my browser?

If the IP address works in your browser but not //tower, then you are most likely using a Mac, so see this post, read the rest of the forum thread too. See also this post. --- (This question is repeated below in the FAQ#Apple Mac Questions section.)

Can unRAID run headless?

Yes it can, *if* your motherboard can. There are a few motherboards that apparently will not boot without a video card attached. Even if you have to have a video card, you can still run without keyboard and monitor and mouse. Some users keep a cheap video card in an anti-static bag, placed in the bottom of the computer case, ready for special use. Some troubleshooting requires seeing the console, especially if networking is not working.
If you try to run headless, and it appears to freeze prior to booting, try pressing the <F1> key. If that works and it boots, then you will need to re-attach the monitor and change a BIOS setting, something about 'Halt on errors'. Change it to halt for nothing.

What size should the parity drive be?

Parity drives must be equal to or larger than your largest data drive. See this thread for more info.

Any tips on selecting a drive and port for the parity drive?

When writing to the array the overall speed is often limited by the rotational speed of the disks involved. This is because when writing to a disk the existing contents of each sector being written must first be read, then parity calculations made, and finally the same sector written. To write a sector that was previously read, the platter on the disk must rotate at least once before the disk head can get to the sector to write it. For this reason, if possible, use a disk with a higher rotational speed for the parity disk, as it is involved in every "write" to the array.
All that being said, there are many unRAID servers with older IDE or "green" drives, that spin at slower speeds, so don't worry if your parity drive spins at 5400 RPM. It will be slightly slower to write to, because it spins slower, but will still be perfectly usable. See also "How does a drive's rotational speed (5400 RPM, 7200 RPM, etc.) affect parity check speed?".
See also Improving unRAID Performance#Move Parity Drive Off PCI Bus.
See also Parity.

Is there a guide to moving all of my existing data to my new unRAID server?

There is no one guide, but there are a number of discussions. There are various methods available, with trade offs between speed and safety.

unRAID Operational Questions

Why is so much RAM being used?

Linux manages RAM differently than Windows. Free RAM is considered wasted, so some or all of it is used for caching, for better performance. RAM used for caching will appear to be allocated, but it is immediately and freely made available to any other program requesting it. For a great explanation, please see Linux ate my RAM!
When using the free -m command to see how much memory is free, don't look at the free amount in the Mem: line, look at the free amount in the +/- buffers/cache: line.

Is it possible to migrate the parity disk from one drive to another to increase the maximum drive size in an unRAID array?

Yes, unassign the old Parity drive, then assign the new one and start the array. It should then rebuild the Parity drive.
It's often recommended to run a parity check first, to make sure there are no drive problems on ANY of the array drives. But if the Parity build fails, you can always restore the previous Parity drive. Make sure the box is checked that Parity is already valid.

Can I replace the parity drive with a SMALLER drive?

Yes, but ONLY if the smaller drive is equal to or larger than the largest data drive. Try unassigning the Parity drive, then starting the array and then stopping it (to make unRAID forget the last Parity drive), then assign the new one and start the array. It should then rebuild the Parity drive.
If for any reason that doesn't work for you, then you will need to make a copy of your array drive assignments, then run New Config (on the Tools page), and reassign your data drives. Then assign the new Parity drive, start the array, and let it build Parity.

How do I upgrade the unRAID software?

Clicking Check for Updates on the Plugins page is the usual and preferred way to upgrade. Plugins don't usually require a reboot, but upgrading the unRAID software itself WILL require a reboot. It will clearly instruct you.
In general, every unRAID release announcement contains the instructions for upgrading the unRAID software.
To manually upgrade, in the usual case, only the 2 files bzroot and bzimage HAVE to be copied from the unRAID distro. As of unRAID 6.2, you need to also copy bzgui, if you are using it. Once they are copied to the flash drive, reboot.

What happens if two data disks fail at the same time?

No data lost at all if you have dual parity drives installed! You will need to rebuild both drives of course.
Unfortunately if you have only one parity drive, you could very well lose data on both disks. But unlike other RAID systems, you will not lose the data on disks which did not fail. And if one of the disks that failed is the parity drive, then you only lost the data on one disk! See also FAQ#How_do_I_recover_data_from_an_unRAID_disk.3F.
However, here's another big advantage of unRAID, the fact that each data drive has an independent file system. When other RAID systems talk about drive failure, they mean that a failure on one drive has caused the loss of data on that drive AND on all of the other drives. On an unRAID system, a failure on a drive means that *possibly* data has been lost on the *one* drive! But it's possible that with certain file system repair tools, there will be NO LOSS at all! The only time you would lose ALL of the data on a drive is when the drive completely fails, either won't spin up at all or won't respond to commands at all. That's a rare situation. Normally, all or most of the data is recovered.

How do I replace a hard disk?

Please see Replacing a Data Drive

How do I replace multiple data drives with a single larger drive?

Please see Replacing Multiple Data Drives with a Single Larger Drive

How would I replace a smaller disk and add new disks at the same time?

It is ALWAYS best to do operations like that one at a time, not at the same time. You should do one of those actions first, then the other. While such an operation might be possible, experience has shown that there's too great a chance to become confused and lose data.
Most likely, you will replace the smaller drive with one of the new drives, and let it rebuild, then afterward add the other drives.

How do I remove a hard disk that I do not plan on replacing?

The basic method is to go to the Tools page and click New Config, then assign all the drives you want to keep. When you next start the array, it will rebuild parity. Obviously, it is vital that you know what your current drive assignments are, especially the Parity drive, so make notes or screen captures beforehand.
For more methods, including a method that maintains parity protection, see the Shrink array page.

How do I remove multiple drives?

The basic method is to go to the Tools page and click New Config, then assign all the drives you want to keep. When you next start the array, it will rebuild parity. Obviously, it is vital that you know what your current drive assignments are, especially the Parity drive, so make notes or screen captures beforehand.
For more methods, including a method that maintains parity protection, see the Shrink array page.

Can I reorder my drives within my array?

Yes you can UNLESS you have dual parity drives. If you have 2 parity drives, you will have to unassign the second parity drive, then rearrange them as desired and restart the array, then stop the array and reassign the second parity drive and rebuild parity on it again.
Warning: Do NOT try to add or remove any drives during this, as that would invalidate parity. And do NOT try this procedure if one or more drives has failed or been removed.
Assuming you only have one parity drive assigned, reassign your drives as desired and start the array. You may have to put a check in the box Parity is already valid.

What if there is something really wrong, for example, multiple drives are missing?

If the system detects a configuration where it cannot start, such as more drives missing or wrong than the number of parity drives, then you will not be able to start the array. You must either correct the situation or reset the array configuration data (with Tools->New config).

How do I move my existing unRAID array to a new system?

This step isn't strictly necessary, but is a wise precaution, just in case something goes wrong! Before you shut down your current server, make sure you have notes of which disks (model and serial number) are assigned to which disk number (logical slot) in your array. Or do a screen capture of the Main page, press Alt-PrntScrn, paste in MS Paint, IrfanView, or similar, and print it out.
Make sure you stop the array and shut the old system down cleanly.
After you assemble your new system and add your drives and flash drive, boot it up and test. It should all look the same!
An unRAID system consists of just your drives and the flash drive. The unRAID software is on the flash drive, along with all of your configuration and drive assignments. So when you boot your unRAID server, the motherboard and other hardware components are always a brand new system to it. It starts by analyzing your hardware and tailoring a brand new Linux system to it, then using your configuration to configure itself, then locates your drives by their serial numbers. So moving to new hardware is nothing special to it!

Can I have multiple unRAID servers on the same network?

Certainly, just make sure they each have a unique server name and a unique IP address. For more detail (but an old thread), see Install two unRAID machines on same network.

How can I access my unRAID server remotely?

It's not updated for v6, but see http://lime-technology.com/forum/index.php?topic=2112.0

How is the key file used?

The key file is tied to a specific USB flash drive by its unique GUID, and is embedded with which license you purchased. It can be renamed but must end in .key, and must reside in the config folder of the flash drive it is tied to. If you have purchased multiple licenses, you can keep all of your key files in all of the config folders, and unRAID will open each and find the one that corresponds to the drive it just booted.

What is the super.dat file?

See this for a description, and this for the technical detail. [old threads]

What if I accidentally assign a data drive to the parity slot?

If you haven't clicked the Start button, then nothing has happened, and you can unassign it and assign it correctly to a data drive slot. If you did Start the array, then a parity rebuild has probably begun and is overwriting your data! Stop the array as quickly as possible! Then carefully read and follow the instructions in this forum thread.
It may sometimes be possible to use the TestDisk live CD, to recover the data partition on your data drive.

Why is my disk being marked as Read-Only

A disk will be mounted as read-only if the operating system detects corruption in the file system that resides on it. It is to prevent further corruption of the disk, making repair more difficult or impossible.
If a disk is mounted as read-only, you should be able to fix the damage by following the procedure described in the Check Disk Filesystems section.

Why am I getting repeated parity errors?

Parity errors should NOT recur, and if they do, something is wrong. If a correcting parity check finds a parity error, it corrects it immediately, so that although it reports a parity error, and the count of parity errors is incremented, there should not be any more actual parity errors in the system.
If you run another parity check soon after the first, and it reports more parity errors, then serious troubleshooting is required immediately! There are 3 kinds of tests that are recommended.
  • Memtest - Run the Memtest that appears on the unRAID boot screen. Most memory errors appear within a few minutes, others within a half hour of so, others may take many hours to show up. Plan on running it all night. The array will be completely down and inaccessible during this test. No memory errors are acceptable. If it takes 24 hours and produces only 1 error, you still need to replace one or more memory sticks. If you can't completely trust your memory, then you can't trust your system.
  • SMART reports - obtain SMART reports for ALL of the drives. If a drive is going bad, then bad sectors may be increasing, and each one may cause parity errors, because the parity check has to use the sector data returned to it, and a bad sector cannot return the correct data. From the SMART reports, it is usually obvious which drive is causing the trouble. See the Troubleshooting#Hard drive failures section on smartctl commands for getting SMART reports, and running SMART tests.
  • reiserfsck - run reiserfsck tests on all of the unRAID data drives. Do NOT run it on the parity drive. If the Reiser file system on a particular drive has been corrupted, it is remotely possible that it may be causing inconsistent results with that drive, for unknown reasons. See Check Disk Filesystems for instructions.
See here and here for examples of this kind of problem, and there are a few others too. Not all cases of this problem have been resolved. If the 3 tests above do not resolve the problem, check for cabling issues, and heat issues, and lastly consider replacing the motherboard. There may be another cause, but it is unknown at this time.
Also, a limited number of UnRAID users have reported problems with controller cards based on the SIL3132 chipset (whether 'real' or 'counterfeit' chips is unknown) - causing repeated parity errors. Specifically, high IO volume (as during a parity check/rebuild) on both 'sides' of the 2-port card. See here and here. Just to be clear -- there are thousands of SIL3132 based cards in successful use. And certainly not "all" (or even most) SIL3132 chipset based cards have this problem. But if you are having repeating parity errors, AND are using SIL3132 cards -- these two threads may be of interest to you. The issue experienced can create SILENT corruption of parity, and/or if PARITY is used to rebuild a data drive, can result in a corrupted new data drive.

How can I get unRAID to accept my perfectly good parity drive?

How can I avoid rebuilding my parity or data drive when I know it is already good?

In unRAID v4 and v5, it was known as the Trust My Array procedure. Marking the Parity drive as already valid is much simpler to do in v6. Please see the following HowTo, but use the v6 instructions.

How can I get my unRAID server to shut down at the same time every day?

Make sure that the Powerdown plugin is installed, then add an at command to your go script, similar to the following, which stops the array and powers down the server at 11PM:
echo "powerdown" | at 23:00

How can I get my unRAID server to wake up on demand?

How can I do a read-only parity check?

Make sure that the check box for "Write corrections to parity disk" is NOT checked, before clicking the Check button

What is the safe way to rearrange disk numbers, assignments, slots, etc?

With unRAID, you can safely change disk assignments, renumber drives, and move drives to different slots and connectors and controllers. Do NOT try to add or remove any drives at the same time, as those actions would invalidate parity, and should be done separately. Do NOT try this procedure if one or more drives has failed or been removed.
Procedure for single parity (assumes you only have one parity drive assigned):
  1. Stop the array (if started)
  2. From the Tools Menu, click on New Config
  3. Rearrange, reconnect, and reassign your drives as desired
  4. Start the array. You may have to put a check in the box "Parity is already valid".
IMPORTANT WARNING!!! If you have setup dual parity, then you CANNOT change disk assignments or renumber the drives, without invalidating the second parity drive. You can safely change the drive physical connections, to different controllers even, but if you have 2 parity drives, you cannot change drive assignments - unless you do the following:
Procedure for dual parity (assumes you have two parity drives assigned):
  1. Stop the array (if started)
  2. From The Tools Menu, click on New Config
  3. Unassign the second parity drive
  4. Rearrange, reconnect, reassign them as desired (but don't assign the second parity drive)
  5. Restart the array
  6. Stop the array
  7. Reassign the second parity drive
  8. Start the array and let it rebuild parity again

How do I keep accurate time on my server?

See "Using NTP for accurate server time". There is also some helpful information on setting the time zone and dealing with daylight savings time.
See also NTP.

Why do I get extremely slow write speeds and/or network dropouts when drives are >99% full?

This is usually only an issue with drives formatted with ReiserFS, and somewhere over 90% full. For this reason, many users have converted to XFS, which doesn't have the problem.
Old link, see This.

ACPI Questions? Will server Power button gracefully shut down the server?

[ Old thread, still useful ] See ACPI Questions? Will server Power button gracefully shut down the server?.

unRAID Console Questions

What is a console?

A console is the text screen you see on the attached monitor of any computer that boots into an operating system but does not boot into a GUI desktop. It's the screen you see when you boot your unRAID server. With special software, a console can also be emulated on a remote computer. After entering a user name and password, you will typically see a command line prompt, somewhat similar to the old MS-DOS command prompt. unRAID is a stripped down Linux distro, based on Slackware (currently Slackware 14.2), with added features specific to unRAID. The console therefore is a standard Linux console (running under the bash shell, sort of like a console in Windows is a DOS shell). A console allows you to type commands, optionally with additional parameters, and get appropriate responses back, all within a scrolling text screen.
To get the same console from another machine, we use an SSH or Telnet client such as PuTTY For more information on PuTTY, SSH, and Telnet, see Terminal Access. The windowed box these clients provide is almost identical to the boot console, very useful for remote control of your unRAID box. PuTTY is better than the Windows Telnet utility because it allows mouse usage and more compound keystrokes and the function keys. Both are free. PuTTY provides both SSH and Telnet, but SSH is recommended now over Telnet. Telnet is not secure.
(Depending on the context, the console often refers to the screen/keyboard of the physical unRAID server machine. For example, pressing [Ctrl][Alt][Del] at the console will cause the server to shutdown if the powerdown package is installed. This won't work from a PuTTY terminal prompt.)

How is a Linux console different from a Windows console?

There are many functional similarities, but they are as different as the underlying operating systems. Even those things that seem very similar will have subtle but important differences. To help you understand some of these, here are a few of the more obvious similarities and differences: (note: DOS and Windows will be referred to as just Windows below) : (note #2: this entire article needs to be finished, reviewed, and corrected by more expert Linux users)
  • Executable files
    • In both, you type individual commands with optional parameters, or batch or script files containing multiple commands, at a console prompt, and typically receive text responses back, in a scrolling text window
    • In Windows, executable files are denoted as such by the file extension, such as .com, .exe, .bat, etc
    • In Linux, executable files generally do not have special file extensions, but must be flagged as executable, typically with a command like: chmod +x filetoexecute
  • Slashes
    • Windows uses the backslash to separate parts of a path
    • Linux uses the forward slash
  • Letter Case
    • In Windows, case does not matter. You can execute any program name in upper or lower or mixed case. Commands and most command parameters including file names are case-insensitive.
    • In Linux case matters! Program names and file and path names and most command parameters MUST be typed exactly as they actually are. For example, file searches will not find files of the other case (eg. ls -l /boot/s* will not show /boot/Sample).
  • Path - the system path - they are used similarly, but there are 2 differences
    • Path syntax
      • Windows uses a semicolon to separate the directories listed in the path (eg. C:\Windows\system32;C:\Windows )
      • Linux uses a colon (eg. /usr/sbin:/sbin:./:/bin )
    • Path searching
      • Windows searches for a command in the current directory first, then down the list of directories in the path
      • Linux does not search the current directory by default, unless ./ has been added to the path, the . represents the current directory. The standard unRAID release does specify it in the path, so it behaves like Windows.
  • Drive letters / mount points
    • Windows refers to installed drives by letter (eg. C:, D:), and automatically 'mounts' each drive and its file system
    • Linux does not automatically mount the drives and file systems. They are mounted by mount commands, manually or as configured in the startup, or by scripted control. The drives are not assigned letters, but are associated with a 'mount point', a folder within an existing file system tree. (In fact, one drive could be mounted in the folder tree of another drive.) For example, your flash drive is mounted as /boot, a Movies share is mounted as /mnt/user/Movies, and Disk 1 is mounted as /mnt/disk1
  • Root folder
    • In Windows, the root folder of any drive is the top level folder, with a path like C:\ . Since the unRAID flash drive would be accessed as a network drive, the files in the flash root folder would have a path something like \\Tower\flash\*.*
    • In Linux, the root folder of a drive is accessed through the 'mount point', with the path of the mount point. Disk 3 is mounted at /mnt/disk3, so the path of the root folder of Disk 3 is /mnt/disk3. The files in the flash root folder would be /boot/* when viewed within a Linux console. (Note: do not add a /boot folder to the flash! The /boot folder already IS the root folder of the flash.)
  • Root user
    • Windows has an all powerful user, with the user name of Administrator
    • Linux has the all powerful user known as root
  • File attributes (such as system, hidden, executable)
    • Windows uses mainly the System, Hidden, ReadOnly, and Archive attribute flags
    • Linux uses Readable, Writable, and eXecutable in 3 ways, 1. for the file owner, 2. others within the same group as the owner, and 3. for other users not in the same group as the owner. (eg. rwxr--r-- is readable, writable, and executable for the owner, but read only for others in the same group as the owner, and read only for all others) These attributes are also used for directory permissions, but the permissions have slightly different meanings. Execute permission on a directory indicates you have permission to use that directory as part of a path to a file. Read permission indicates you may get a listing of contents of the directory, and Write permission indicates you may create or delete files in that directory. Note: If you have permission to write to a directory, you may delete any file in the directory, even if the file itself is not writable. So, if you cannot delete a file, enable "write" permission for the folder the file is in.
    • Windows uses the console command attrib to change them
    • Linux uses the chmod command to make changes to them
    • Because you can view and use files on an unRAID server through both Windows and Linux, there has to be a mapping of these very different attributes. These mappings, especially of the Windows System and Hidden attributes, has often lead to considerable confusion. Making a file or files executable in Linux will usually turn on the System and Hidden flags, which may cause the files to appear to disappear from Windows. Please see this post for more information. This post has a nice diagram of the attribute mappings.
  • End-of-lines - see the question below: "Why do my scripts have problems with end-of-lines?"
  • more to come
    • commands (dir/ls, rd/rmdir, md/mkdir, mem/top?, cd../cd .., etc)
      • see the DOS links below
    • dir (case, format, aliases, etc)
    • find / ff / dir/s / which
    • doskey / aliases / .bashrc
    • key usage (arrows, function keys, etc)
      • Most special keys cannot be used within a Linux console
      • Shift-PgUp and Shift-PgDn can be used within a Linux console to page up and down through previous screens
  • More information

What commands can I use at the unRAID console?

Only Linux commands and scripts can be used, and only those that have been included in the unRAID release, or that you have installed. Many users install the NerdPack plugin, which provides access to a number of additional commands.
For help with commands you may wish to use, please see the following pages:

How can I change the number of lines displayed in the boot console?

See http://lime-technology.com/forum/index.php?topic=1017.0.

Why do my scripts have problems with end-of-lines?

Linux and Unix use a single Linefeed (0a) as an end-of-line. DOS and Windows use a Carriage Return and Linefeed pair (0d0a) for each end-of-line. Mac's use a single Carriage Return (0d) for an end-of-line. In general, you must use the appropriate end-of-line character(s) when attempting to execute a script or batch or command file, with each of the operating systems mentioned.
Typically with unRAID, users create or copy scripts on a Windows machine, and need to strip the Carriage Returns from the file before it will work correctly on the unRAID server. Here is a pair of commands that can be used to properly convert a file for unRAID's use. It uses the monthly_parity_check.sh script as an example.
fromdos  <monthly_parity_check.sh  >tempfile
mv  tempfile  monthly_parity_check.sh
The Linux tool fromdos strips the carriage returns and writes a tempfile, then mv writes the tempfile back to the original.
The Windows Notepad does not deal correctly with Linux-style line endings. You may want to use NotePad2 instead. See also the UnRAID Topical Index, Editors topic.

Hard Drives

What is recommended preventative maintenance for the hard drives?

See Troubleshooting#Hard_drive_failures - "How to prevent failures".
See also Best of the Forums#Maintenance Recommendations (links are old)
See also http://lime-technology.com/forum/index.php?topic=2135 (link is old)

How fast should my parity check be?

It all depends on the type of system you have, IDE, Mixed, SATA and the hardware your using. Take a look at the User_Benchmarks to see what others are reporting. You can add yours as well.
See also "How does a drive's rotational speed (5400 RPM, 7200 RPM, etc.) affect parity check speed?" (link is old)

How do hard drives become disabled?

When a write operation fails on a disk in a protected array, the system will disable the disk. A disabled disk will no longer be used in any way by the system. The disk still appears as a share, and you may still read and write it. This is because unRAID is working behind the scenes to simulate the failed disk. (Since the array is already using its redundancy features to simulate the failed disk, if another disk failure occurs, you will cause data loss on BOTH disks, so it is important to fix the problem quickly.)

What does the Red Ball mean?

What does the Red X mean?

A red X icon to the left of a drive on the unRAID Main page is an indication it has been taken out of service because a "write" operation has failed. (It may also be red if you have unassigned it to simulate a failed drive.) See Troubleshooting#Hard_drive_failures.
Note: a drive that was assigned as part of the array and then simply unassigned is still part of the unRAID array. It is treated exactly the same as a disk that had failed. In fact, if a second drive were to fail you could lose data.
If you want to remove a drive from the array permanently, and do not wish to immediately replace it, see How do I remove a hard disk that I do not plan on replacing?
Do not be misled by the fact that you can still read and write to the drive with a red X indicator. You are, in fact, writing to the parity drive as if the failed drive was working. When reading, you are reading all of the remaining drives and re-constructing the data on the failed drive. If a drive has a red X on the unRAID Main page, it has been taken out of service. You will need to take corrective action, as a second concurrent disk failure will almost certainly result in lost data.
To fix a drive with a red X, see What do I do if I get a red X next to a hard disk?

How do I recover from a hard disk failure?

See the Hard drive failures section

What is wrong with my Seagate 1.5TB drive?

You probably need a firmware upgrade, see http://lime-technology.com/forum/index.php?topic=2773.msg22889#msg22889.

Warning: this FAQ is being updated for unRAID v6. Information below this line may be out-of-date.

How do I re-enable a failed disk?

Once a disk is disabled, its contents may be invalid, because there may have been uncompleted writes. unRAID will ignore it, so all further read requests to that disk will be serviced by reading Parity and all the other Data disks in order to reconstruct the requested data on-the-fly. All further write requests will be similarly simulated.

It is very important to determine WHY the disk was disabled, because very often it is not the fault of the drive. More often than not, we find that there is an interface issue to the drive, such as a bad SATA cable (very common) or bad or loose SATA cable connectors, or power issues such as bad or loose power cable or its connectors, bad power cable splitter, too many drives on the same 12v power rail, or bad power supply. Some times, it is just a disk controller that went offline, making it impossible to access the drive. Obviously, you do not want to return/RMA a drive or throw it out, if there is nothing wrong with it, just a problem with its cables or power.

The first step then is to examine both the syslog to see what kind of errors are being reported, and the SMART report for the drive, to check its status and errors. Please see the Troubleshooting page for more information and instructions. For advanced users, see the The Analysis of Drive Issues page for examples of drive and interface error messages.

If you determine that the drive is bad, then it needs to be replaced, see "How do I replace a hard disk?".

If the drive was fine, and you either found and corrected an interface issue, or could not find anything wrong, then the disabled drive still needs to be re-enabled. There are 2 ways to do it, and the choice depends on whether you have made changes to this drive after it was disabled. Remember. It was taken out of service and disabled because a "write" to it failed. It is guaranteed to not have the correct current contents.

If you do not care about losing anything recently written to the drive, or can live with a file that was partially written with the subsequent "write" failed, or file-system corruption because the file structures were not written properly, then read the Trust My Array procedure, and see if your array qualifies for it. If so, then the Trust My Array procedure is perfect for quickly returning your array to all green. Let the ensuing parity check complete. Use of the Trust Procedure will most likely result in a corrupted file-system in need of repair, or missing/corrupt files that were not written being lost. You have been warned. It is almost never the correct action to take. If used, you should then check the filesystem for corruption and then get parity in sync with a parity check.

If however you have made changes to the drive that you do not want to lose, or do not want to risk file-system corruption, then you will need to rebuild the drive in-place, with the following procedure (very similar to replacing a failed drive).

  1. Stop the array, go to the Devices page and un-assign the disk
  2. Start the array.
  3. Stop the array, go to the Devices page and re-assign the disk
  4. Go back to the Main page - the system should detect a "new" drive for the one disabled, click the Start button to start the rebuild (parity-reconstruct) of the disk (you may have to click the little check box under the Start button that says "I'm sure I want to do this") (post with screen shots)
  5. Drive will now be rebuilt, takes a while; the array can be used at the same time, but we recommend waiting until the rebuild is complete. The procedure of un-assigning the drive, starting the array without it, and then re-assigning it allows a drive to be used as its own replacement.

How do I recover data from an unRAID disk?

There are a number of situations under which you might want to read data from an unRAID ReiserFS formatted disk. Note that unRAID disks are formatted with the ReiserFS v3.6 file system.

If you are trying to recover files and folders from a drive with a Reiser file system that appears to be corrupted, please read each of the following articles and threads:

If the Reiser file system is fine, but you want to access its files outside of your unRAID array, here are 6 ways to do it:

  1. Install the drive in your unRAID server or to an external eSATA port, mount the drive, and copy the files. Instructions are in this How To, but ignore the NTFS and FAT instructions and use -t reiserfs instead of -t ntfs-3g or -t vfat as the file system type. Do NOT add it to the array!
  2. Install the drive in an external USB enclosure, attach to your unRAID server, mount the drive, and copy the files. Instructions are in this How To, but ignore the NTFS instructions and use -t reiserfs instead of -t ntfs-3g as the file system type.
  3. Set up a test unRAID system, connect the physical disk to the test server, boot it with the unRAID Basic version, and add your disk to the array in one of the disk slots. BE CAREFUL NOT TO ASSIGN IT TO THE PARITY SLOT!!! You can then start the test array and unRAID will share the disk, allowing you to access the data.
  4. If you are using User Shares AND the folder structure on this drive matches your User Shares AND you want to import its data into your User Shares, then unassign your current Cache disk, and install and assign this drive temporarily as your Cache disk. You can now start the mover script to have your files automatically moved off this drive and into your User Shares. This is obviously a very special case, but a shortcut if it happens to match what you are trying to do. [1]
  5. Use a Linux live CD to boot any computer with this drive installed, and mount it with the reiserfs file system.
  6. Connect your disk to your Windows computer, and install a ReiserFS driver (see this link for directions and links), or see this for YAReg and rfstool. See also this and this.

How can I undelete files from an unRAID disk?

There are no known undelete tools for the Reiser file system, which the unRAID system uses for each data drive. However, there are techniques that can be used to recover files.

See here for how to perform this on an unRAID array: http://lime-technology.com/forum/index.php?topic=5087.msg47070#msg47070

See this article and comments for some general help: http://antrix.net/journal/techtalk/reiserfs_data_recovery_howto.comments.

Why is a disk showing as Unformatted?

Whatever you do, do NOT format the drives if you expected them to have data, or data loss could occur!
  • Update: As of unRAID v4.5-beta7, disks should no longer show up as "unformatted" if they are "busy".

unRAID will show a disk as "unformatted" whenever it is unable to mount the ReiserFS file system that it expects to exist on it. A disk can fail to be mounted (and as a result, show up as "unformatted") when ANY of the following occur:

  • A disk is still in the process of being mounted. This may occur when first starting the array if it was not cleanly stopped as transactions in its journal are re-written to the disk. This process can take 5 or more minutes on a system with lots of file activity just prior to the non-clean shutdown of the array. (A power failure will often cause this need to replay transactions) Be patient and just refresh the browser every few minutes. Do not format the drive! If after 10 minutes the drive is still showing as unformatted, post a copy of your syslog in a new thread on the unRAID forum requesting guidance. The odds are good the file system is corrupted and needs repair.)
  • The disk has a corrupt reiserfs file system and is unable to be mounted.
  • The disk is defective (although most times, if defective, it will show up as missing)
  • The disk has a different kind of file system other than "reiserfs" (A disk with an NTFS file system, or FAT file system, of ext2 file system will show as unformatted)
  • A different disk in the array was "busy" and was unable to be unmounted when the "Stop" button is pressed. When this occurs, the array will not stop when you press the "Stop" button. Those disk(s) showing as unformatted were able to be unmounted, but now that they are unmounted, they do not appear to have a file system on the mount point, therefore they show as unformatted. The busy disk(s) will still show their contents. This should only apply to unRAID versions prior to v4.5-beta7.
    • (If you Telnet to unRAID and "cd" to a folder on a drive, it is "busy." Or if you, or a script you add, is reading or writing a file at the time the unmount of that disk is attempted, it is "busy" and will not be unmounted. Either can result in other drives being falsely reported as "unformatted" since unRAID can cannot properly stop the entire array.)
    • In this last situation of a "busy" disk causing others to show as unformatted, simply stop the process keeping the disk busy and press "Stop" once more. The array will then stop as expected.
  • The disk is really unformatted. (It may even be partitioned, but no file system was ever created on it)

Why is a temp not showing for a drive?

This is almost always an indication that SMART is not enabled for a drive. As of unRAID v4.3 beta 6, SMART is enabled automatically before the temp is requested. For prior versions or if the temp is still missing, see http://lime-technology.com/forum/index.php?topic=2120.0.

For an older method (but only for IDE drives), see http://lime-technology.com/forum/index.php?topic=479. This method involves adding cat commands to the go script. Ignore the output of the cat command. See also http://lime-technology.com/forum/index.php?topic=94.msg2134#msg2134.

Also make sure that SMART is enabled in the BIOS menus (hard drive settings), *if* there is a setting for it.

Is my drive too hot?

For some thoughts on drive temps -

Any guidance on drive cooling, fans, air flow?

Do 'green' drives work with unRAID?

Yes they do, and work well, usually creating savings in power costs. Note that although they can be used for both parity and data drives, for best write performance, select a faster drive for the parity drive.

A good thread about Western Digital 'Green' drives: http://lime-technology.com/forum/index.php?topic=1540.0

How can I find out more information about a hard drive?

The following commands require the Device ID, which you can get from the Web Management page, Devices tab. Locate your drive, then look for the Device ID in parentheses. It is always 3 lowercase letters, beginning with either hd or sd, eg. sda, sdk, hdc, hdg. For simplicity, sdx will be used below, and you will substitute the appropriate Device ID for your drive. (Extra spaces are added for clarity only, not needed.)

To view the identity and configuration information for a drive (at the console or telnet prompt)

hdparm  -I  /dev/sdx

To determine the read speed of a hard drive, the following command can be used. The very last number in MB/sec is the one you want, ignore the rest. Although one run will give you a decent result, for better accuracy, take the average of at least 5 runs. See also Check Harddrive Speed.

hdparm  -tT  /dev/sdx

To obtain the SMART info for a drive, including some identity and configuration information, and physical statistics and error history. For more information about SMART and smartctl, see here and here and here and here and here.

smartctl  -a  -d  ata  /dev/sdx

As of unRAID v4.3final, smartctl is included with unRAID. Prior to this, it needed to be copied to the flash drive, see this for links to obtaining it. If you had to copy it to your flash drive, then the command to run smartctl would be

/boot/smartctl  -a  -d  ata  /dev/sdx

Why don't all my drives spin up when I click the Spin Up button?

See http://lime-technology.com/forum/index.php?topic=2164.0

What is the best way to transfer files within my unRAID Server?

Refer to Transferring Files Within the unRAID Server for info on Midnight Commander and other techniques to move files around within your unRAID server.

If my hard drive is reporting the wrong size, how do I restore the full factory capacity?

For an excellent discussion of why the wrong capacity may be reported, and a tool to restore the full hard drive capacity, see http://blog.atola.com/restoring-factory-hard-drive-capacity/.

Why is my hard drive clicking or making popping noises?

Probably because it is failing! Immediately copy all data off of it, and be ready for complete failure of the drive. Once your data is safely off, then run SMART and/or other tests on it (see Troubleshooting#Hard_drive_failures), and obtain SMART reports for it. You can often use the SMART report to facilitate the RMA of the drive.

There are reports of cabling being a possible cause of clicking. See here and here.

Some drive models may have a higher reported incidence of clicking problems, see here for one example.

See http://datacent.com/hard_drive_sounds.php for typical sounds of failing hard drives.

Why did my hard disk Device ID change?

The Linux kernel may change the Device ID's at any time, so do not depend on them staying the same for a particular drive. See http://lime-technology.com/forum/index.php?topic=3200. If the array does not Start automatically, check the Devices tab and correct any drive assignments, then use the Trust My Array procedure.

How can I securely wipe a drive?

If you are through with a drive, and are planning on disposing of it by selling or giving it away, then you should securely clean all of your data off the drive first. The best way currently is to use the Preclear Disk script, with the -n option. More discussion of that, plus a method for wiping with random data, begins here. For some general discussion of wiping methods, their safety and security, see this thread.

Does it hurt the drive by spinning it up and down too much?

See http://lime-technology.com/forum/index.php?topic=3137.msg28209#msg28209

Why does my SATA disk show up as a PATA disk?

Some BIOS have an option to make SATA disks appear to be a native IDE disks. If enabled, a disk may be assigned /dev/hdX as its device instead of /dev/sdX. See this thread

What is an Advanced Format Disk or 4K HDD?

See Advanced Format Drives - includes a FAQ for these drives

How do I rebuild a disk's partition table?

It should be very rare that a user would ever need to rebuild their partition table, but should it ever be corrupted, see HERE for a Joe L. developed script that will rebuild it.

Does unRAID support 3TB and larger drives?

Yes, in the later 5.0beta series. (It was initially added in 5.0beta6.) Note that your disk controller must also support drives larger than 2.2TB. (many do, some do not)

How To Troubleshoot Recurring Parity Errors

This procedure is a test to find what is causing parity changes if a parity sync is continuously finding issues and traditional tools (Memcheck, SMART, reiserfsck) can't find the disk(s) that is having an issue. This procedure will test the disks and the SATA controllers to ensure each disk can be properly read by the system without errors. The basic idea is to read the problem section of each drive and create a md5 checksum for the data read. The script shown below will perform this read and checksum calculation 5 times for each drive and all 5 checksums should be equal if the drive and SATA controller are healthy.

First, ID which block are causing the issue. After a parity sync, this can be found in the syslog:

Something like:
Jan 4 22:15:26 Tower kernel: md: parity incorrect: 279944 (Errors)
Jan 4 22:15:28 Tower kernel: md: parity incorrect: 279958 (Errors)
Jan 4 22:15:31 Tower kernel: md: parity incorrect: 280013 (Errors)

Look at the errors and pick a number of blocks where the errors occur as the section of the drive to test. In the example above, the errors start at block 279944, so the skip number which is used to pick the starting point would be 200000. This is to start a little before the first error. The example has 3 errors which occur within 69 blocks of each other so the count could be say 100000. These skip and count values would specify the section of drive to read goes from block 200000 to 300000. If the parity incorrect numbers are further apart then a larger count would be used. A typical count could be 1000000 or even 10000000. As a rule, try to pick a group of parity incorrect errors that constantly occur. You don't need to cover every error that occurs so pick a bad area of the drive. Note that 2000000 blocks tests about 1gig of the hard drive.

Now that the area to be tested is known, build scripts that will read from these blocks for all drives. The respective scripts will reference each drive as /dev/sdx, where x is a different letter for each drive. these device identifiers can be found on the unMENU or unRAID pages and will appear as sda, sdb, sdc etc. You may find that sdc corresponds to "disk6" or /dev/md6. Remember to use a Linux friendly editor to create these scripts.

So, the script for disk6 which would begin at block 200000 and read 10000000 blocks or 5gig of the drive surface would look like this:


for i in {1..5}
    echo "Begin sdc for the $i time."
    dd if=/dev/sdc skip=200000 count=10000000 | md5sum -b >> sdc.log

Build a script for each drive in the server. You can name them md5_sda.sh, md5_sdb.sh, md5_sdc.sh, md5_sdd.sh etc

Run the line below to make each script executable.

root@Tower:/var/log/hashes# chmod 555 ./md5*.sh

Now, run the scripts simultaneously to put the maximum stress on the drive I/O system. So, separate the scripts on the console by '&'

root@Tower:/var/log/hashes# ./md5_sda.sh & ./md5_sdb.sh & ./md5_sdc.sh & ./md5_sdc.sh & etc

This will run 5 times for each drive and append to the corresponding error log (sdx_sim.log)

While it is running, the output will look something like this:

Begin sda for the 2 time.
10000000+0 records in
10000000+0 records out
5120000000 bytes (5.1 GB) copied, 216.273 s, 23.7 MB/s
Begin sdb for the 2 time.
10000000+0 records in
10000000+0 records out
5120000000 bytes (5.1 GB) copied, 216.397 s, 23.7 MB/s
Begin sdc for the 2 time.

Once these have completed, you can cat the results and see if they are the same.

Here is an example of a bad result. Notice that each md5 result is different.

root@Tower:/var/log/hashs# cat sdc.log
775e9351bb8b6a1ffe11fde1396a8d41 *-
f4cd927ce33785e20f4bd97a755088c6 *-
d1980745346128bc19de66ff4d93cd10 *-
93c331b810d677403d8cbeaf70fa8ec8 *-
74d614bc5cff5cb43e286a3092b5f5e7 *-
14e40296a7ee9a51c99bcec8201a242b *-

Here is an example of a good result where each md5 result is the same.

root@Tower:/var/log/hashs# cat sdd.log
8c6e629c618f3c3078cc2fe4a7976da5 *-
8c6e629c618f3c3078cc2fe4a7976da5 *-
8c6e629c618f3c3078cc2fe4a7976da5 *-
8c6e629c618f3c3078cc2fe4a7976da5 *-
8c6e629c618f3c3078cc2fe4a7976da5 *-
8c6e629c618f3c3078cc2fe4a7976da5 *-

The drives don't need to be unmounted for this test. However, make sure no writes are done to the array while the tests are running. Otherwise, the writes could throw off the results.

Here are links to a few example cases using this technique on the forums;

Finding a bad drive
Bad SiL 3132 controller
Parity errors galore on new unRAID box
Possible causes for repeating parity sync errors?
Intermittent / inconsistent parity errors (Contains an updated script with more advanced checking, courtesy nick5429)


How can I determine if my motherboard will work with unRAID?

  • One way is to check the Hardware Compatibility lists to see if your specific motherboard has already been proven to work.
  • Another method is to see if any motherboards listed in Hardware Compatibility are using the same chipsets for disk controllers and network connectivity as your potential motherboard.
  • A third method, and sometimes easiest if you already own the motherboard, is to prepare a flash drive with unRAID and use it to boot up your potential motherboard. If it boots and you see a "Login" prompt on the system console, it is a good clue that you will be able to boot from your flash drive. If you can browse to the web-management page, it is a very good indication that everything will work on your LAN. If on "Devices" page you can see your disks as potential choices in the drop-down lists, the motherboard is compatible.

What should I look for in a motherboard?

Most important is to find a board that is already known to be compatible. Make sure that any board you are interested in, is either listed in the Hardware Compatibility motherboard lists, or is very similar to one listed, that is, has the same basic chipsets as a board on the list. The Hardware Compatibility page begins with an excellent discussion of this subject, especially the Motherboard introductory thoughts.

Next, see these motherboard guidelines.

Flash Drives

What USB flash drive should I use?

For best results, select a flash drive from the Hardware_Compatibility#USB_Flash_Drives list.

The unRAID Server must be installed on a 128MB or larger USB Flash device. Almost any good quality USB flash drive can be used, but if you will be purchasing an unRAID license, you will need one with a GUID that does not end in a string of zeroes (see GUID question below). Although a 128MB drive will work, a larger size is recommended if you will be enhancing your unRAID server with other addons and enhancements, or storing large logs. 1GB or larger drives are commonly used.

I can't boot from my new unRAID flash. What can I try next?

How do I find my USB flash GUID?

How do I configure the BIOS settings to allow booting from the USB flash drive?

Why does my flash drive disappear as a boot option when I add a hard drive?

For one possibility, see this. For a problem when adding a 16th drive, see this.

If my flash drive dies or is lost, will I lose my license?

See the "Unofficial" USB Policy, here, or here.

Does the speed of the flash drive make any difference?

The speed of the flash drive makes little difference. It is read when booted, but has very little read/write activity otherwise. If you use a USB 1.0 drive, it will take much longer to boot, but once up and running, you will not notice any difference.

Can I remove the flash drive while unRAID is running?

You should not remove the flash drive, while booted with unRAID. See http://lime-technology.com/forum/index.php?topic=2390.msg18477#msg18477.

How do I update files on my flash drive?

See http://lime-technology.com/forum/index.php?topic=2390.0.

How do I upgrade my unRAID flash drive to a larger size drive?

If you do not have an unRAID license (Plus or Pro), then it is simple. Just copy the contents, including config folder, of the current flash drive to the new and larger flash drive, and reboot with the new drive.

If you have an unRAID license, then it is tied to the GUID of that flash drive, and you will need to first apply by Email (support@lime-technology.com) to Lime Technology for a replacement of the key file. Include a complete explanation, and the GUID of the new flash drive. Once you receive it, you will move the contents of the old flash to the new drive, as above, but you will also replace the key file.

Notice: this is NOT an official policy! Lime Technology has been generous in replacing keys in the past, but will have final say as to whether and how a replacement is available, or not, and whether their generosity will continue in the future. Lime Technology has a right to charge a fee for replacement keys.

See also the related flash drive questions and answers above.

What to do about "reset high speed USB device message"?

If you get this message there can be a couple things causing it. It will appear in your syslog and look something like this: Tower kernel: usb 2-1: reset high speed USB device using ehci_hcd and address 2 The first step to take is trying every USB port on your motherboard to see if only certain ones give there reset message. If that does not work, you can try a new flash drive, but be warned that if you have a registered version of unRAID the license will no longer be recognized. If you don't want to try a new flash drive you can log in via telnet and issue the following: echo 64 >/sys/block/sdX/device/max_sectors Make sure to replace the X above with the correct letter your flash drive has. Doing this will not entirely fix the problem (at least it did not for me) but it should decrease the frequency of the above error.

Is there a list of the files on the unRAID flash drive?

For a list and description of the system and user files and folders that may be found on the unRAID flash drive, see Files on the unRAID flash drive.


Why is my add in network card not working?

See http://lime-technology.com/forum/index.php?topic=2109.0

Why is my network dropping out?

Unfortunately, there are many causes of brief or extended network loss. The following discussion touches on methods of discovery, that hopefully will help you find a solution: http://lime-technology.com/forum/index.php?topic=1820. (more to be added)

How do I Stop/Start/Restart the networking on my unRAID server?

See http://lime-technology.com/forum/index.php?topic=2111.msg15508#msg15508

How do I find out what is the speed of my network connection?

ethtool  eth0

The ethtool utility is included in some but not all unRAID distributions. See here for more information about ethtool, and a download link.

What Linux driver is being used by my network card or chipset?

ethtool  -i  eth0

- or -


- or you can find it listed within your syslog, in lines that include eth0. The eth0 line that also includes with driver has the name of the driver and its version.

How can I find out more information about my network connection?

Here are 5 commands that will provide more info, about the driver, about the card and configured parameters and speed, and about its connection statistics. (If ethtool is missing, see the link above. The extra spaces are not needed, added for clarity.) For more detail on these and other network-related commands, see Console Commands for Networking.

ethtool  -i  eth0
ethtool  eth0
ethtool  -S  eth0

Can I connect my unRAID server directly to my PC?

See this post, plus the subsequent posts

Any advice on network cabling?

The unRAID users' favorite place to buy good quality cables and cable accessories seems to be MonoPrice, great prices and quality.

Since gigabit networking is very highly recommended (even if you don't have it yet, you probably will in the future), we recommend Cat5e or Cat6 network cables. Cat5e is the absolute minimum, but Cat6 is better, especially for the future as network speeds continue to climb, and it only costs a little more than Cat5e.

The Structured Wiring - How To site has some very good tutorials and pictures and diagrams of network wiring. It covers from designing your network to tools that are useful to the nitty gritty of network cabling, with lots of good advice. Unfortunately, it is a little out-of-date, with no mention of gigabit, Cat6, or Wireless-N, but it still covers the fundamentals very well.

See also Running cat6 cables.

Why did //tower stop working after changing the router or switch?

If you can no longer access the unRAID Web Management page by using //tower, after adding or changing a router or switch in your network, and none of the IP addresses have changed, and you can still access it by the IP address, then please see this post.

How do I configure NFS mounts?

Starting with unRAID 4.5beta1 and/or 4.4.1, unRAID includes the ability to serve files using the NFS protocol.

This release includes NFS export ability for disk and user shares. To use this feature, go to the Shares page. For the disk shares, you will see a new field called "Disk shares (NFS)". For the user shares you will see the field "Export (NFS)". In these fields you should enter the NFS client specification string. If the export string is left blank, then the share will not be exported via NFS.

For example, to export the share to any system, read/write, enter:


To export the share to read-only, and read/write, enter:,

Notice no white space in there. Refer to the linux documentation for 'exports', e.g., http://linux.die.net/man/5/exports

These 'default' options are set:


Note for Mac users: OS X users will need to use the following export options, otherwise the shares will not mount: [2] (You may need to restart your server.)


In addition a 'fsid' value is set for each user share.

If there are no shares specified to be exported via NFS, then all the various NFS daemons will not be started.

NFS shares have the path '/mnt/user/usersharename', for example the command to mount 'smb://unraid/photos' might look like:

mount unraid:/mnt/user/photos ~/mntpoint

Please consider NFS support to be experimental. For example, there has been no performance tuning done whatsoever (as of 4.5beta1/4.4.2). However, see this thread for improved settings and better performance.

Other Hardware

How much memory does unRAID require?

Prior to v4.3, unRAID required a minimum of 512MB, to a maximum of 1GB. It was recommended that 2 matched sticks of 512MB be installed, to take advantage of dual channel mode.

As of v4.3, a maximum of 4GB is supported, and higher amounts will probably be supported in the future. See the forum threads mentioned in the next question as to why more memory can be helpful.

What about memory?

Apart from the amount of memory (discussed in the previous question), unRAID has no special requirements. But many motherboards do, and compatibility has to be checked carefully, and memory settings need to be matched, such as voltage, speed, etc. There are a number of forum threads that discuss memory. The Best of the Forums#Hardware Suggestions is a great starting point, but read the entire threads, particularly these 3: [3], [4], [5]. See also this thread for more memory tips.

How do I figure out which hardware component is bad?

  • For a helpful tip on isolating a hardware issue when you don't know where the fault is, see the second half of this post
  • Kernel Panic - a little generalized troubleshooting, sometimes bad memory or memory settings, bad PSU, etc; kernel panics are often associated with bad memory
  • See also Troubleshooting#General hardware issues

Why is PCI Express better than PCI?

See this and this.

What are the common bus speeds?

Common Buses and their Max Bandwidth

PCI	132 MB/s
PCI 2.1 64-bit 500 MB/S
AGP 8X	2,100 MB/s
PCI-X 66Mhz      532 MB/s
PCI-X 133Mhz     1064 MB/s
PCI Express 1x	250 [500]* MB/s
PCI Express 2x	500 [1000]* MB/s
PCI Express 4x	1000 [2000]* MB/s
PCI Express 8x	2000 [4000]* MB/s
PCI Express 16x	4000 [8000]* MB/s
PCI Express 32x	8000 [16000]* MB/s
IDE (ATA100)	100 MB/s
IDE (ATA133)	133 MB/s
SATA	150 MB/s
SATA	 300 MB/s
Gigabit Ethernet	125 MB/s
IEEE1394B [Firewire]  100 MB/s
USB Full-Speed            12 Mbit/s (1.5 MB/s).
USB High-Speed (2.0) 480 Mbit/s (60 MB/s)

Interesting Links

Any advice as to power supplies and power cables?

Cache Drive

What is a Cache drive?

See Cache disk and To Cache drive or not to Cache drive?

See also http://lime-technology.com/forum/index.php?topic=2149.msg15814#msg15814

How do I set up a Cache drive?

See Cache disk

The Cache drive requires unRAID v4.3-beta4 or later, and is only available with a purchased license (Plus or Pro). Prior to unRAID v4.5-beta8, it was only available with the purchase of the unRAID Pro license. See the Release notes here: http://lime-technology.com/forum/index.php?topic=1731.0.

Weighing the benefits and disadvantages of using a cache drive

See To Cache drive or not to Cache drive?

What if I want to watch a movie that is still on the Cache drive?

See Question about how the cache drive works

How much faster is write performance using a Cache drive?

See Performance Difference With/Without Cache Drive (somewhat dated)

See also To Cache drive or not to Cache drive?

What if the Cache drive is not big enough?

See http://lime-technology.com/forum/index.php?topic=2698.msg22177#msg22177

Why are my files not moving off the Cache drive?

See Mover not working, files stuck on cache drive

User Shares

What should I set 'Split Level' to?

Here's a visual example: http://lime-technology.com/forum/index.php?topic=1928.0.

Here's an example of Split Level = 2: http://lime-technology.com/forum/index.php?topic=1837.msg13416#msg13416.

Here's a good example of Split Level = 1: http://lime-technology.com/forum/index.php?topic=3483.msg30193#msg30193.

And for a little different but clearer explanation of what to set the split level to, see http://lime-technology.com/forum/index.php?topic=3483.msg30082#msg30082.

See also the UnRAID Manual, User Shares section.

As of unRAID v4.5-beta3 (but possibly available in any version after v4.4), there is a new way to control the split level functionality. See http://lime-technology.com/forum/index.php?topic=3509.msg30684#msg30684.

As of unRAID v4.5-beta5, the behavior of Split Level = 0 has changed again, see the v4.5-beta5 Release Notes. The behavior has changed several times since Split Levels were first announced, so questions about Split Level = 0 need to include the version of unRAID being used.

Why is only the first drive of my User Share filling up?

If you are using a version prior to the release of unRAID v4.3 Final, make sure that the Split Level setting is more than 0, and not blank. Change it to 1 or 2, and test again. Note that this behavior was fixed/changed with the final release of v4.3, and "Split Level = 0" now means -> "split permissible at all levels", that is, there will not be any effort to keep files together on the same drive.

Why is my User Share out of space when there is plenty left?

This has always been a confusing aspect of User Share management, and involves understanding how the Split Level and Allocation settings interact. Please read the following discussions. (Note: be aware that some of these discussions were written prior to the change in the behavior of split-level=0, mentioned in the question above.)

How does including and excluding drives work?

See the UnRAID Manual, Included disk(s) section, also the Included and Excluded disk(s) section within the User Shares section of the UnRAID Manual.

See http://lime-technology.com/forum/index.php?topic=3350.msg28884#msg28884 for a clarification, Inclusions and Exclusions should be all lowercase

See also http://lime-technology.com/forum/index.php?topic=3079

How can I set up anonymous user access?

How can I control read/write access to a User Share?

For a discussion of methods of limiting read/write access while allowing full read-only access, see http://lime-technology.com/forum/index.php?topic=3189.

How do I remove a User Share?

From the UnRAID Manual, User Shares section, Deleting User Shares paragraph, "To delete a User Share, just clear the Share name field and click Apply. Only entirely empty User Shares may be deleted." You should also delete the top level folders for that User Share from every unRAID disk before the next boot, or the User Share may be re-created.

See here and here

Why is access to my shares from a Linux or Unix box so much slower than from a Windows box?

A partial answer is here: http://lime-technology.com/forum/index.php?topic=1975.0.

Why does Split Level go back to zero?

If changes to the Split Level setting are not 'sticking', just Stop the array and reboot your unRAID server, and changes to Split Level should 'keep'. (possibly unRAID 4.2 only)


How can I get User Level Security working on my Windows network?

See "Help - User Level Security"

How do I secure my unRAID server?

  1. Set a strong root password on the Users tab of the Web Management pages.
  2. Adjust the read/write access status of your unRAID shares on the Shares tab of the Web Management page - see the UnRAID Manual, Shares section. Setting shares to Read Only protects them from changes. Setting shares to hidden does not secure them, but can keep kids out of places they don't belong.
  3. Enable User level security (on the Settings tab) and set up users and passwords for your User Shares - see the UnRAID Manual, Users section.

For a good discussion of security for your unRAID server, see Official and/or community stance on security.

How can I get more advanced control of the Samba configuration, users, and permissions?

Why are changes to security settings not working?

It may not be in the manual, but currently, users have found that you need to reboot after making changes to User Security settings, before they take effect. See "User level security woes"

What characters are not legal in an unRAID password?

Short answer: can't use these characters in a password: | & ; ( ) < > space tab ; $ # * @

Long answer: the System Management Utility uses a bash shell to set the system password (using chpasswd command) and the samba password (using the smbpasswd command). The above characters have special meaning to the shell.

See http://lime-technology.com/forum/index.php?topic=974

How long can the unRAID root password be?

Currently, the maximum length via the web interface is 40, but it is possible to run into issues with password lengths greater than 8 with some forms of access to your server. See this forum thread for a good discussion of length and allowable/recommended characters. (This entry will be updated as more accurate info is made available.)

Microsoft Windows Questions

For questions related to both unRAID and Windows

How do I automatically login to mapped network drives from Vista?

See http://lime-technology.com/forum/index.php?topic=3299

Apple Mac Questions

For questions related to both unRAID and the Mac

Why does //tower not work in my Mac browser?

If the IP address works in your browser, but //tower does not, see this post, read the rest of the forum thread too. See also this post. --- (This question is repeated above, "Why does //tower not work in my browser?")

Is there a guide for Mac users?

See UnRAID/MAC Basics

What do Mac users do with the Mac hidden files?

See For the Mac users out there

Is there a way to change the mover's handling of dotted top-level folders on the Cache disk?

Note: this is particularly useful for Mac users, but others may find uses for this tip too.

Integrating unRAID with Other Apps

See the UnRAID Topical Index for a guide to discussions of other applications in the unRAID forums. Currently, you can find lists of links for AFP, BeyondTV, BluRay, FTP, iTunes, MythTV, NewsLeecher, NFS, NTP, Popcorn Hour, SageTV, SMART, SpinRite, Torrents, VirtualBox, VMware, and XBMC, among others. However the entries may or may not be completed or well researched yet.


See http://lime-technology.com/forum/index.php?topic=2668

See also the UnRAID Topical Index, iTunes topic


See http://lime-technology.com/forum/index.php?topic=1800

See also the UnRAID Topical Index, SageTV topic

Why do I have to keep reinstalling a user application?

This also answers the question as to why an app always thinks this is the first time it has run, in other words, why it forgets the previous settings changes.

unRAID is different than most Linux setups, in that it runs entirely in a RAM disk. So apps that are installed to this in-memory file system usually don't realize they are saving their settings to memory, which is lost on reboot. When installing apps in unRAID, you usually have to either install them to a fixed disk location, such as on the Cache drive or the flash drive, or you need a mechanism to copy off the settings file before shutting down, and restoring it when you reinstall the app on the next boot. You will usually have to ask another user of your particular app, about what files need to be saved and restored, and how they handle that. The go file (in the config folder of your flash drive) is like a DOS or Windows autoexec.bat file, and is used often to copy saved application settings to their appropriate operational location.