Thread fault NtfsParseThread

[jlexer]

I just purchased ZAR for a server which the Hardware RAID5 volume was reported as RAW. The volume was NTFS and it is now showing as RAW after a failed disk and the volume was reconstructed at the hardware level. This is a Dell server, with a PREC2 Quad raid card. The volume is a RAID5 of 10 x 73 GB SCSI disks.

I ran ZAR with the simple volume /existing RAID setting, It ran apparently fine for about 12 hours until 'Parsing Filesystem: Reading(87%)', when I received the error:

An error has occurred and it is not possible to continue execution
The program will now terminate. The error message was
Thread fault NtfsParseThread
Do you want to anonymously submit the error report?

I answered yes, at which time ZAR shuts itself down.

Please help

[Alexey V. Gubin]

How large the log file is,
C:\Program Files\ZAR\logfile.txt?
if you ZIP it and it will be less than 10MB zipped, email it to development@z-a-recovery.com, would be a nice addition to the error report. If it is larger, I'll set up an FTP so you can upload it to us.

[Alexey V. Gubin]

I checked the log file and it looks like a typical RAID rebuild failure. So the conclusion is that the RAID was rebuilt wrong way at the hardware level. ZAR then runs out of ability to sort the things out and crashes.

If the hardware rebuild was complete, this most likely means the data is now trashed beyond our capabilities to repair. RAID5 withstands loss of one disk; however if the controller messes up the rebuild, or if the faulty disk is misidentified, for 10x array about 10% of data is overwritten by the new parity and thus lost. Worse yet, this data is distributed evenly across the disk.

You can try this with some other data recovery software, but I think most files will be recovered unusable. From the log, it is looks pretty much like a lost case.

Another option is an in-lab recovery, but you have to tell them specifically that it is suspected that a rebuild went wrong.