Bit Rot and Silent Corruption: How to Check an Archive Stays Intact
Files can change quietly on aging storage. Learn what silent corruption is and how a simple checksum list lets you spot damage before it spreads.
Most people assume that a file saved today will be identical when they open it in ten years. Usually it is. But storage is not perfect, and over long periods, small errors can creep in without any warning. A photo that opens with a strange stripe, or a document that suddenly refuses to open, may be a sign of silent corruption.
What silent corruption means
Silent corruption, sometimes called bit rot, is when the contents of a stored file change without any error message. A few bits flip, a sector degrades or a transfer goes wrong, and the system carries on as if nothing happened. Modern drives use error-correcting codes that fix many problems automatically, so true corruption is uncommon. But it is not impossible, especially with aging media, power problems, faulty cables, failing memory or buggy software.
The real danger is that you do not notice. If you copy a damaged file to your backup, the backup now contains the damage. Over time, every copy can end up affected.
Different media, different risks
No storage type is immune:
- Hard drives can develop bad sectors as the magnetic surface and mechanics age.
- Solid-state drives and USB sticks store data as electrical charge, which can fade if the device sits unpowered for a very long time. How long varies widely by product and temperature, so do not assume a drive in a drawer is safe indefinitely.
- Optical discs can degrade through age, scratches or poor manufacturing.
- Network storage and cloud services usually have protection built in, but you still depend on someone else's systems, and errors can occur in transit.
Some modern file systems store checksums for every block and can detect damage automatically, which is a point in their favor. Many everyday drives do not.
The simple defense: a checksum list
A checksum is a short fingerprint calculated from a file's contents. If even one bit changes, the fingerprint changes completely. By recording the fingerprints when your archive is in good condition, you can later check whether anything has altered.
- Generate the list. On Linux, a command such as
sha256sumcan process a folder of files. On macOS,shasum -a 256does the same. On Windows, the PowerShell commandGet-FileHash -Algorithm SHA256calculates the fingerprint for each file. Many free tools offer a point-and-click version. - Save the output in a plain text file kept next to the archive, and also in a second location.
- Verify periodically. Calculate the fingerprints again and compare them. On Linux,
sha256sum -cautomates the comparison against a saved list. - Investigate differences. If a file does not match, it has changed. Decide whether it was an intended edit or damage.
As a rule, do this for files you do not expect to change, such as finished projects, photographs and scanned documents.
Keep more than one copy
A checksum tells you that a file is damaged. It does not repair it. To fix a bad file, you need another good copy. That is why the 3-2-1 backup rule is so valuable: with several copies on different media, one damaged file can be replaced from another. Comparing checksums across copies also tells you which one is correct.
Build verification into your routine
- Check important archives once or twice a year.
- Verify after copying a large collection to a new drive.
- Replace aging drives before they fail rather than after.
- Practice restoring files, as in our guide to testing backups.
- Pay attention to warning signs such as unusual noises, slow reads or repeated errors.
What not to expect
Checksums cannot tell you whether a file was good in the first place, nor protect against deletion or theft. They detect change. Used with multiple copies and regular checks, they are a cheap and effective way to keep an archive trustworthy.
Frequently asked questions
How common is bit rot? True silent corruption is rare on healthy hardware, but over many files and many years it becomes a real possibility, so checking is sensible.
Which checksum algorithm should I use? SHA-256 is a widely used, reliable choice for integrity checks of this kind.
Do I need to check every file every month? No. A yearly or twice-yearly check is enough for most home archives.
Steps
- Create a checksum list: Generate a SHA-256 checksum for each file in your archive and save the results in a text file stored beside the archive.
- Verify on a schedule: Once or twice a year, recalculate the checksums and compare them with the saved list to find any file that has changed.
- Repair from another copy: If a file fails the check, replace it with a good copy from a second location and investigate the drive that held the bad one.