Upgrading Windows 10 Pro to 20H2 corrupts data partition

Piotr1955 thank you!

Turns out these TCG errors have also appeared in the System logs of the previous Windows version (C:\Windows.old\Windows\System32\winevt\Logs), after every power cycle/reboot. So they might be related to this incident, but not the direct cause/indication since data partition D was healthy and accessible before the upgrade.

After further investigation of installation and System logs, I think I found the exact moment of the incident (logs attached in a zip):

• 11:14:04 - first upgrade reboot.
• 11:14:05 - Applications and Services Logs -> Microsoft -> Windows -> NTFS -> WHC shows "NTFS global corruption action state is now 0."
• 11:14:08 - System log shows "Volume 😧 (\Device\HarddiskVolume5) is healthy. No action is needed."
• 11:14:21 - setupact.log shows the current partition table, with D as NTFS, Bitlocker disabled.
• 11:14:46 - upgrade is shrinking the OS partition C and creating a new recovery partition.
• 11:14:46 - Applications and Services Logs -> Microsoft -> Windows -> Partition -> Diagnostic shows the new partition table with 6 partitions.
• 11:14:49 - Applications and Services Logs -> Microsoft -> Windows -> NTFS -> WHC shows "NTFS global corruption action state is now 1."
• 11:14:49 - System log shows "Volume 😧 (\Device\HarddiskVolume5) requires an Online Scan. An Online Scan will automatically run as part of the next scheduled maintenance task. Alternatively you may run "CHKDSK /SCAN" locally via the command line, or run "REPAIR-VOLUME <drive:> -SCAN" locally or remotely via PowerShell."
• 11:14:49 - Application log shows "Chkdsk was executed in verify mode on a volume snapshot. Checking file system on \Device\HarddiskVolume5"
• 11:14:53 - System log shows "A corruption was discovered in the file system structure on volume D:. The Master File Table (MFT) contains a corrupted file record. The file reference number is 0x1000000000024. The name of the file is "<unable to determine file name>"."
• 11:15:02 - Applications and Services Logs -> Microsoft -> Windows -> NTFS -> WHC shows "NTFS global corruption action state is now 3."

After this, the original recovery partition (#4) and data partition (#5) are both RAW (data looks totally encrypted, on both partitions - first and last sector are identical indicating NTFS, drive has Opal active, but not locked), the new recovery partition pushed them in the partition table (making them #5 and #6).

My guess is that this change in partitions has caused Windows to take some action involving the drive's TCG Opal system, maybe taking ownership/activating it, or when initializing the new partition table - drop ownership/deactivating it and stop providing necessary decryption credentials.

To conclude, partition D is healthy and accessible, and 2 seconds after the partitions change - it turns RAW and the NTFS service is erroring and trying to recover/chkdsk.

Please advise, I must understand what happened in this upgrade bug, and possibly recover the data. Note I have not yet rolled back the upgrade since I don't want to cause more damage and reach a totally unrecoverable situation. If the issue is finally investigated and cause is found, then there might be an indication rolling back will help.

Thank you!