[bug?] Disk corruption and system crash when mirroring

[gembler]

Hello,

I have a reproducable failure when mirroring one USB HD to another USB HD. Some relevant details:

- The OS is Windows 11 Pro 24H2.
- FFS version is 14.4 64 bit 7/26/25 Donation Edition.
- The mirror operation had been working for over a year prior to the failure.
- Both USB HD drives are the same type and connected to a Dell Thunderbolt dock USB 3.0 ports.
- The source HD file system is NTFS and the sink HD file system is exFAT.
- The triggering condition appears to be copying a large file, an Outlook OST file at slightly over 1 GB size, or a large folder, a Firefox Profiles folder at about 743 MB size.
- Prior to the mirror operation, the sink HD had 2.7 TB available space.
- The mirror operation fails and displays an error dialog box that the sink HD is out of space.
- Sometimes clicking the skip or cancel buttons in the dialog box trigger a system crash with error code 0x34.
- After system restart, the sink HD contains a new folder with name either FOUND1.000 or RecycleBin~c1be.ffs_tmp.
- The FOUND1.000 folder contains over a hundred "orphaned" fragment files.
- The RecycleBin~c1be.ffs_tmp folder contains a mirror of a Firefox Profiles folder heirarchy with some files.
- File Explorer->Properties for the sink HD shows the HD is full.
- Running the Properties->Tools->Error checking repairs the HD and restores the HD free space.
- The sink HD passed two full chkdsk runs twice with no errors found.
- The failure happens when database file tracking is either enabled or disabled.
- Deleted and overwritten files are retained in the Recyle bin.
- FFS runs with Admin rights and with Fail-safe file copy and Copy locked files enabled.
- FFS log files show normal messages with one warning message: "Some files will be synchronized as part of multiple folder pairs.".

I searched the forum but didn't find a failure that quite matched this one.

Let me know if there is anything else you need to debug this.

Thank you,

Gary

[xCSxXenon]

Sounds like a failing disk. Use GSmartControl to check the health attributes of the drives

[gembler]

I ran GSmartControl Extended Self-Test and after 9 hours it found no errors. However, the Error Log shows a few "Error: ABRT 8 sectors at LBA = (Address values)" errors so perhaps the drive is failing.

Is there any way for FFS to detect and prevent the system crashes? This would help make FFS more robust with regards to the state of the HDs and OS.

[xCSxXenon]

Likely a fail indicator.
And no, this error is something that the OS should catch and handle, not software. It is very likely occurring during a system call, which isn't accessible by the user space by design.

[gembler]

Understood. Here is a suggestion: Add a tip about this to the Troubleshooting section of the FAQ.

[xCSxXenon]

Since it is unrelated to FFS, it wouldn't make sense to add it to the FAQ. It's more general computer knowledge