NTFS file id inconsistent over time with no balancing on.

[arizzi]

Hello!  I've written a program that does checksums for my files on a stablebit drive pool.  It uses the NTFS id for the file as an identifier. https://learn.microsoft.com/en-us/windows/win32/api/fileapi/ns-fileapi-by_handle_file_information?redirectedfrom=MSDN

I notice across scans that the ntfs id for the file is changed, even if no changes were made to the file (file was not replaced either).  Is this expected behavior?

My workaround is just to scan the drives directly rather than the pool.

[Shane]

Hi arizzi, please see this thread: https://community.covecube.com/index.php?/topic/12577-beware-of-drivepool-corruption-data-leakage-file-deletion-performance-degradation-scenarios-windows-1011/

TLDR, DrivePool's implementation of FileID is persistent across neither reboots nor moves, and increments from zero on the former so collisions are inevitable.

Furthermore, for potential problems with checksums if read striping is not disabled on a duplicated pool see this thread: https://community.covecube.com/index.php?/topic/5414-pool-file-duplication-causing-file-corruption-under-certain-circumstances/

TLDR, to avoid potential checksum/data corruption you should disable Read Striping on any pool that uses duplication.

There is no ETA on a fix.