Frequently Asked Questions

RAID Reconstructor — FAQ

Note In the FAQ list below, the icon "" represents links into the RAID Reconstructor online help file. They open in a new Explorer window.

General Questions

Yes, you find a step-by-step introduction in the RAID Reconstructor help file.

We have also prepared five examples for you to study. They demonstrate some typical usages, as well as best practices for RAID Reconstructor:

  1. RAID-5 with 3 drives on a HP 454:
    This example demonstrates a typical recovery of a broken hardware RAID.
  2. RAID-5 with 3 drives on a HP 454, 1 drive defect:
    Recovery of a RAID-5 after the failure of one drive.
  3. RAID-5 with 4 drives on a Dynamic Disk set:
    Recovery of a Windows Software RAID-5, demonstration on using the XOR Test for finding an unknown start sector.
  4. RAID-0 (Striping) on a HP 454:
    Recovering a striped set (RAID-0)
  5. RAID-5 on Iomega StorCenter IX4-200D:
    Typical recovery of a NAS. Demonstration on finding an unknown start sector with DiskExplorer or the Entropy Test.

You need to move the drives into a working machine as additional drives, or you can use a bootable CD with our software on it to get access to the drives. The drives need to be attached to a non-raid controller for this to work. You can find all the details about creating a bootable WinPE CD here, and about creating a Linux-based CD here.

Any single file RAW image will work. If our software creates it, then a multi-file image can be used. You can not use an Encase image; however, you can output that Encase image to a single file DD image and use that in RAID Reconstructor.

You can create an image in either RAID Reconstructor ("Tools->Create single image file..."), GetDataBack, or DiskExplorer. Each has its own way to create an image so check the help file in each of them. They all produce the same type of image file. Do not use DriveImage XML as that is not a compatible image for data recovery.

A NAS usually is a Linux software RAID. You need to find the start of the RAID partition, usually the last and biggest partition. Run the analysis from there.

Example: NAS Software RAID-5 on Iomega StorCenter IX4-200D

A more convenient way is our special software NAS Data Recovery that does all the work for you.

RAID Reconstructor can reconstruct any file system provided the array is not of a proprietary order, and the start sectors are the same across all the drives in the array.

The XOR test verifies the checksum over a drive set and visually shows you the XOR of the set's drives. Check the help file for detailed instructions.

The entropy test allows you to see the array structure visually. The graph can help you determine the drive order, block size, or if a drive is missing or damaged. Check the help file for detailed instructions.

The drive checksum is the CRC32 checksum over the data of sectors 0 to 5000 of the drive. You can use this checksum to identify a drive after a RaidProbe.

If your RAID was NTFS-formatted, the MFT report shows you MFT entries in the context of individual drives. This can help you identify parameters such as drive order, block size, and rotation.

This report creates a list of file system structures and their locations on the drives. Its purpose is to give you an overview of the content of unknown drives.

The RAID Calculator lets you map sectors between the RAID and its individual drives.

There is also an online version.

Section  1. Define  Questions

Yes. If you have four drives, for example, tell RAID Reconstructor that you have four drive and enter the three good drives, while leaving one of the drive fields empty. The software will rebuild the missing drive from the XOR data of the other three drives. 

Unfortunately, the nature of a RAID-0 does not allow for the loss of a drive in the array. There is not a way to recover this array with software.

A RAID-5 can only be recovered with the loss of one drive. If you have lost two drives, there is not a way to recover this array with software.

A RAID-1 is two or more drives that are exactly the same data duplicated across the RAID-1. You can simply take one of the drives and scan it in GetDataBack to recover the data.

A RAID-10 (1+0) is two RAID-0's put together. You need to determine which drives belong to each side of the RAID-0 and then use RAID Reconstructor to analyze the RAID-0. If you can not determine the correct drives on each side, you should use our RaidProbe Service and let us take care of it for you.

These types of arrays can be recovered using our RaidProbe service. If you already have knowledge about your RAID-6, you can create a VIM file yourself.

There is not a way to add your own parity schemes. You can use our RaidProbe service, and we can determine the parameters and prepare a virtual image (VIM file) for you.

A virtual image file can describe any RAID scheme, but creating one can be tricky.

Section  2. List  Questions

There are two ways. The first is to right-click on the white space to the right of the drive number. You will have a drop-down list to choose from. The second way is to click the three periods at the end of the white place, and the same drop-down list will appear.

The drives are still attached to the RAID controller, which hides the physical drives from Windows. Attach the drives to a non-raid controller, and you should then see them on the list.

Section  3. Open  Questions

Look at the message, and it will tell you what drive the error occurred on.

If it is a RAID-5, remove that drive from the drive list in section 2 and rerun the analysis. If you get more sector read errors, make images of the faulty drives, and then use the images.

If it is a RAID-0, make images of the faulty drives and then use the images instead of the physical drives.

Section  4. Analyze  Questions

If you do not know these settings, do not touch these options. The software will determine and set these parameters automatically during the analysis.

Managed Auto (2020)

The software uses FFT analysis to find repetitive parity patterns on the drives. It attempts to find the number of drives, the drive order, the block size, and possible start sectors before prompting you for the combinatoric sleuth. This is the preferred method since 2020.

Managed Auto (Legacy)

The software uses entropy analysis to find repetitive parity patterns on the drives. It attempts to find the block size and possible start sectors before prompting you for the combinatoric sleuth. This was the preferred method before 2020.


If you have reliable information about the RAID, you can narrow the range of parameters for the combinatoric sleuth.


The software uses the 2020 method to narrow the parameters and automatically proceeds to the combinatoric sleuth.

This means the software did not find the proper parameters for your array. There are many reasons this could happen. If you have excessive file system damage, if your array uses a proprietary order or rotation, if the start sectors on the drives are different, and many more. In a case like this, we recommend using our RaidProbe service and let us determine all the parameters for your array.

This means the software has identified the correct parameters for your array. You should click "Finish" and proceed to section "5. Copy".

No. RAID Reconstructor does not look at the array on a file level, so there is no way to do this. The scope of the software is to determine the parameters of the array only. By getting a "Recommendation" in section 4, you know that you are getting the array's correct parameters. Provided there is no significant file system damage, you should be able to recover the data from the reconstructed array.

Section  5. Copy  Questions

You must purchase the software before you can output the recovery with any of the options in section 5. Once you enter your key, you will have the ability to save your output.

A virtual image is an XML document that tells our other software (DiskExplorer, GetDataBack, and Captain Nemo) how the array is put together and pulls the data directly from the physical media. This prevents you from having to output an image that is the full size of the array or writing the data to a new drive before recovering the data.

It outputs a small file (less than 3 KB in most cases) that will allow you to recover the data using our other software (DiskExplorer, GetDataBack, and Captain Nemo). It spares us the need for a full image or the data being written to a new drive first.

This outputs a raw image file that is the same size as the array. For example, an array of 600 GB will output a 600 GB file. This image is useful if you have a different recovery software you wish to use.

This allows you to write the data back to a new drive directly. This is useful if you expect the data to be completely intact, for example, if you only had a controller failure. You write the data to a new drive, reboot, and have full access to the data. You may even be able to put the drive back in the machine the array was on and boot from it. This selection will destroy any data on the target drive, so be careful when using it.

Section  6. View  Questions

If possible, you should always use Captain Nemo as it is faster and more efficient than using GetDataBack. If you have file system damage and the image does not mount in Captain Nemo, then you should use GetDataBack at that point.

Once the software created an image or a virtual image, these links open the image in the application link you choose. For example, suppose you have a virtual image and you click the Captain Nemo link. In that case, Captain Nemo will launch, and the image will mount itself up and present the data to you (provided there is no file system damage preventing Captain Nemo from mounting the image). If the application you choose is not installed, you will be presented with a link to download it first.

The only time you will get this message is when you have the software installed for the link you clicked but you have not created the output file in section 5. Create the Virtual Image/Image and then click this link.

© 2024 Runtime Software