0x00000017 (ERROR_CRC) — Bad Data on Disk or Transfer

1. Bad SATA or USB Cable (Most Common Cause)

What's actually happening here is the data coming off the drive doesn't match the checksum the drive controller computed when it wrote the data. In practice, on modern hardware, this almost never happens unless the electrical signal got corrupted between the drive and the motherboard. The number one culprit is a cheap, old, or damaged SATA cable or USB cable.

I've seen this on everything from a 2019 Dell XPS with a loose SATA connector to a 2022 custom build where the cable was pinched against the case. On external USB drives, it's often the cable that came in the box — those things are built to a price, not for reliability.

The fix: Replace the cable. Not reseat it, not bend it back into shape — buy a new one. For SATA, use a cable shorter than 18 inches (450mm). Longer cables are more prone to signal degradation at higher speeds. For USB, use a cable rated for USB 3.0 or higher, even if the drive is older. A USB 2.0 cable on a USB 3.0 port can cause intermittent CRC errors.

# Quick check: Force a retry on the same cable
# Open Command Prompt as admin, replace D: with your drive
chkdsk D: /f /r /x

# If chkdsk reports multiple CRC errors on different files,
# it's almost certainly the cable or the drive itself.
# If errors are on one file only, could be software corruption.

After replacing the cable, test with a file copy that failed before. If it works, you're done. If not, move to the next section.

2. Failing Hard Drive or SSD

The second most common cause is the drive itself going bad. On a hard drive, CRC errors often come just before bad sectors start appearing. On an SSD, it can indicate NAND flash cells going bad or the controller starting to fail.

The reason chkdsk helps here is it reads every sector, tries to recover data from spare sectors, and marks bad ones so the OS stops using them. But chkdsk can't fix a drive that's physically dying. If you're getting 0x00000017 errors on the same drive across multiple cables and different ports, the drive is toast.

Check drive health with SMART:

# Run in PowerShell as admin
get-wmiobject -namespace root\wmi -class MSStorageDriver_FailurePredictStatus | Select-Object InstanceName, PredictFailure

# If PredictFailure is True, replace the drive immediately.
# Also check reallocated sector count with:
wmic diskdrive get status

If SMART says the drive is failing, your next step is data recovery. Don't run chkdsk again — it can make things worse on a failing drive by thrashing the heads. Image the drive with ddrescue or HDDSuperClone instead.

For SSDs, check manufacturer tools: Samsung Magician, Crucial Storage Executive, WD Dashboard. A failing SSD will often report a high 'Uncorrectable Error Count' or 'Media Wearout Indicator' near 100%.

3. Memory (RAM) Corruption

This one is rarer but real. If the cable is good and the drive passes SMART tests, the CRC error might be happening because the data got corrupted in RAM before being written to disk. Or during the read, the data passes through RAM and gets corrupted on the way to the application.

I've seen this on a system with a bad overclock on DDR4-3600 that was stable in games but threw CRC errors on large file transfers. The memory controller couldn't keep up with sustained I/O at that speed.

Test your RAM:

# Run Windows Memory Diagnostic
# Press Win+R, type 'mdsched.exe', hit Enter
# Choose 'Restart now and check for problems'

# Or use MemTest86 from a USB boot drive — it's more thorough.
# Let it run for at least one full pass (1-2 hours for 16GB)

If MemTest86 shows any errors, your RAM is unstable. Fix it by: loading XMP/DOCP defaults, lowering speed, or replacing the sticks. A single bit flip in memory can turn a valid CRC into garbage, and the disk controller can't tell the difference.

4. Software Corruption (Rare)

This is what people assume first but is actually the least likely. Windows file system corruption (NTFS metadata issues) can cause CRC errors, but it usually shows up differently — as 'The file or directory is corrupted and unreadable' (0x80070570), not 0x00000017.

If you've ruled out cables, drives, and RAM, then try:

# Check system files
sfc /scannow

# Then check the drive's file system
chkdsk C: /scan   # Safe, no repair, just reports

If SFC finds corrupt system files, run it again after a reboot. If chkdsk finds file system errors without any hardware issues, a format and reinstall might be the cleanest fix. But again — 90% of the time this is a cable or dying drive.

Quick-Reference Summary