Hi all,
I’m using Kopia on Windows and have a question about selectively restoring files after disk damage.
I had a drive D: located on a fast NVMe SSD. Certain subfolders on D: were regularly backed up using Kopia (for example, I have a snapshot of D:\Data). Unfortunately, the NVMe drive has now failed.
Before it died completely, I was able to copy all still-readable data from D:, including D:\Data, to another drive. However, during the process, there were many read errors, and chkdsk reported that several clusters were replaced. This means that some files are now silently corrupted (still present, but with bad data).
Now, instead of manually checking 1000+ files, or restoring the entire D:\Data folder (which is about 10TB, stored on a slow remote backend), I’m hoping that Kopia can help.
Since Kopia calculates and stores checksums for every file, is there a way to:
- Compare the existing files (copied from the damaged disk) with a snapshot in Kopia, based on checksums, and
- Get a list of files that differ (e.g. corrupted or missing), and
- Restore only those files?
The goal is to validate that only those files are truly damaged, and then restore just them (likely ~1TB total which would be much better than pulling all 10TB unnecessarily).
Is this workflow currently possible with Kopia? Or are there any recommended tools/scripts that could help with comparing the current filesystem against a snapshot based on file hashes (data can also be accessed from Linux if that helps)?
Thanks in advance for any insights!