A larger RAID rackmount system.
Running the Check Disk (CHKDSK) utility on a degraded RAID volume can cause catastrophic, irreversible data loss. If you’ve lost data from a RAID array, contact a professional data recovery provider as soon as possible — do not run CHKDSK or other drive maintenance utilities.
To be clear, CHKDSK can be a useful tool for fixing file system errors on a single, healthy hard drive; it is not data recovery software, however. It’s also a relatively simple program that doesn’t account for RAID structure.
Instead of running software utilities, your first priority should be to safely back up any accessible critical data (if possible) and then create sector-by-sector clones of the remaining healthy drives. Read our guide to cloning hard drives safely.
In this article, we’ll explain more about why CHKDSK is the wrong tool for recovering RAID data. If you need assistance with a failing or degraded RAID volume, we’re here to help. Call 1-800-237-4200 to speak with an expert or submit a ticket online to schedule a risk-free evaluation.
The Basics: CHKDSK and RAID Terminology
When we discuss RAID systems on this site, we often assume that the reader understands the basics of RAID structure. That’s not always a safe assumption — many people run simple RAID 0, 1, and 5 arrays at home, and they may not understand how redundant systems function.
To that end, let’s define a few terms (feel free to skip to the next section if you have a solid understanding of the basics):
- CHKDSK: This is a Windows utility that scans a volume’s file system (like NTFS or ReFS) for inconsistencies in the metadata — the data that organizes your files. When it finds errors, it attempts to “fix” them by removing corrupt file records, marking unstable sectors as “bad,” and trying to make the file system map consistent. Sometimes, it also deletes data that is unreadable.
- RAID (Redundant Array of Independent Disks): RAID combines multiple physical hard drives into a single logical volume for performance or redundancy. In redundant arrays like RAID 5 or RAID 6, data is striped across drives along with parity information. This parity data is a form of mathematical checksum that allows the array to reconstruct the data from a failed drive on the fly.
- Degraded State: A RAID array enters a degraded state when one or more of its member drives fail, but not enough to cause total failure of the array. For example, a RAID 5 can tolerate one failed drive. The array is still “online,” but it’s running with no redundancy and is actively using parity to calculate the missing drive’s data for every read request.
A degraded RAID must be rebuilt, and it’s a good idea to back up any important data before starting a rebuild. It’s also very possible that additional drives will fail during the rebuild, since it’s a fairly intensive process.
For that reason, if you have any data on a degraded RAID that is not backed up, we recommend consulting with a data recovery professional before taking additional steps. Of course, many rebuilds are completed without issue — but if you start the process by running a filesystem utility intended for use with single drives, you’re setting yourself up for failure.
What Happens When You Run CHKDSK on Degraded RAID
When a RAID is in a degraded state, the process of calculating data from parity is stressful for the remaining drives. If one of these working drives develops a minor read error or has a slow response time, the RAID controller can fail to reconstruct a block of data correctly. When this happens, it passes stale, corrupt, or incomplete data up to the operating system.
The operating system doesn’t know about the RAID’s internal problems; it only sees a corrupted file or directory structure. This is often what prompts a user or system administrator to run CHKDSK, hoping to fix the file system error.
Here’s where things get dangerous: CHKDSK scans the volume and sees what it interprets as massive file system corruption (which is really just the stale data being fed by the struggling array). Following its programming, CHKDSK tries to “fix” the problem by:
- Reading the incorrect data from the degraded volume.
- Making decisions based on that faulty information (for instance, deciding that a critical file system structure like the Master File Table is completely corrupt).
- Writing those “fixes” back to the array.
When CHKDSK writes these changes, it permanently overwrites correct data with garbage or deletes pointers to thousands of files. We frequently see cases in our labs where a recoverable degraded RAID 5 case was greatly complicated because someone ran CHKDSK. The utility effectively fixes the file system by destroying the data within it.
Action Plan for Degraded RAID Systems
If your RAID controller is reporting a failed drive and the array is in a degraded state, do not run any file system repair utilities. Do not run CHKDSK, fsck, or any third-party disk repair software. Instead, follow these steps:
- If Possible, Get a Backup. If you can still access critical files, back them up immediately to a separate storage device. Prioritize your most important data first. Do not attempt to back up the entire volume, as high I/O activity could cause another drive to fail.
- Identify the Failed Drive. Use your RAID controller’s management software to clearly identify which drive has failed. Be absolutely certain of the failed drive’s identity and position in the array. Label all drives with their position in the array.
- Create Drive Clones. Before you attempt a rebuild or introduce a new drive, create a full, sector-by-sector clone of every remaining online drive in the array. A hardware-based disk imager is ideal for this. With clones, recovery attempts can be performed without risking the original media. Once again, if you don’t have access to those tools — call us at 1-800-237-4200. We’d rather help our clients avoid data loss situations than recover from them.
- Attempt a Rebuild. Only after you have secured complete clones of the healthy drives should you replace the failed disk with a brand new one and initiate the rebuild process through your RAID controller.
Professional RAID Data Recovery Services
At Datarecovery.com, our engineers work with failed and degraded RAID arrays every day. We use purpose-built hardware to create safe clones of member drives before performing any recovery procedures, ensuring we never put your original media at risk. We also support our non-destructive process with our no data, no charge guarantee: If we can’t recover the data you need, there’s no charge for the attempt.
If your RAID is degraded or has failed, don’t risk running software utilities. Contact our RAID recovery specialists at 1-800-237-4200 or submit a case online for a free, no-risk evaluation.
