I believe the fix could be more simple. When the read error happens, the system read call will return how many bytes were actually read, but because of the error it is less than what is asked for. The program can tell there is an error (because the return is less than the request), but it keeps retrying the same read anyway, even though the outcome is always the same (the read call returns short or no data). It should be a matter of marking the read as bad and moving on. I don't know if it matters or not, but it is happening when I am exploring the file system, as the MFT has read errors. Once I try to open a certain folder, it locks up.

As for using SCSI, I just implemented that in my virtual driver, so that will also work for me But I think the general issue with the system reads should be fixed first, it should be an easy fix.

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

That all said, users of UFS and R-Explorer should not be trying to directly read a failing source drive, rather the clone of said drive.

_________________
Luke
Recovery Force Data Recovery

Yes, most users are not using the programs on a real drive, and that is likely why this bug has gone unnoticed. Also, I just tested it on Windows, and it does not totally lockup, so this this bug is likely only with the Linux version, which again is why it would go unnoticed.

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

What's new in version 8.0:
* Enhanced disk imager function:
-Added settings for read method, timeout, buffer size, direction, protocol (ATA, SCSI), defects skipping, etc.;
-Added optional "safe zone" test after reading errors;
-Saving map of defective blocks in *.sdmf format and logging of copying errors;
-Imaging of used space on the basis of a file system map;
-Imaging of a selected group of files without having to save them out.
* Advanced “Open storage device” procedure:
-Added configurable parameters for data access (method, timeout, etc.);
-Added direct chaining to “read-once” (on-demand imaging);
-Added defects emulation on Read-once and SDLSparse virtual storages;
-Added option to 'stop' the drive (when it works in ATA/SCSI mode).
* Reworked image file loader from fragments:
-Added virtual size and component type indication;
-Added procedure for automatic sorting of component files.
* Extended support of DeepSpar DDI:
-Bitmap-based imaging via DDI;
-Added support of split DDI disk images (metadata+data disks);
-Indication of disks connected through DDI;
-Changed check for DDI availability by using addressed "discovery" packet (it must respond and with the same config).
* Added map of used FS blocks and its conversion to a defects mask (for FAT12/FAT16/FAT32, ExFAT, NTFS, ReFS/ReFS3, HFS+, APFS, Ext2-Ext4, ReiserFS, XFS/XFS5, UFS, UFS/BE, Adaptec UFS, VMFS/VMFS6, VirtualFS).
*Major rework of Hexadecimal Viewer:
-Added address translation to parent for file formats (VHDX/VHD, VMDK/seSparse, VDI, PAR, QCOW/QCOW2, DMG/DMG4), RAID (0, 1, 1E, 3, 5, 6, 6A, Span, BeyondRAID, RDL, MS Storage Spaces), storages created using data mapping (CoreStorage, LVM with thin provisioning, encrypted volumes, etc.), hex view of files;
-Added indication of used FS space in the hexadecimal viewer and moving to the next used/free cluster when indication is enabled;
-Added offset as a percentage to hex inspector;
-File fragments view moved to the right-panel window with added virtual offset and size fields, sorting by offset/virtual offset, "goto virtual offset in file" option;
-Added color indication in HEX of the currently selected file fragment.
* Extended reporting functions:
-Added "error type" column to the displayed report of files and folders;
-Added "Check files for defects" report type to indicate "defective"/"non-imaged" files;
-Added "bad files" report in HTML format;
-Added report of file fragments.
* Added software events log for disk detect events (access, read, etc.), I/O errors (defects), some of VFS events, etc. with a log control (details, save to file, config, clear);
*Optimized file system scan with five options of handling existing file system;
*Modifications to file copying:
-Added optional "summary" line to the copying log (in checkbox mode or when a folder or multiple objcts are being copied);
-Added option to save files with read errors (the error is logged, unreadable blocks are filled with a fixed pattern);
-Added option to move file to $BadFiles folder when file is "copied with read error".
* Fixed a number of bugs and issues.

Wooow, seem to be very promising this update.
Very good job

Unfortunately i don't have Deepspar but i own MRT which all we hate due the way how File Explorer behavies when extracting files.
MRT can't separate properly BAD files from GOOD files becuse it drops the bads one all in single folder (without original folder structure).
This means that many files get renamed, so is near to impossible (very difficult) understanding from which folders are those coming from.

Is UFS able to separate good files from bad files and drop them in a folder (ex called DAMAGED) with original folder structure of source drive?
Will in future UFS support MRT as Deepspar?

MRT staff doesn't want to listen their customer, they will never fix this and other heavy bugs, especially in File Explorer.
It is very annoing that we can't generate a proper list of bad files to send to ours customer, maybe UFS staff can overcome easily to this very annoing mrt shortcoming.
I hope so

_________________
My firmware database:
https://mega.nz/folder/O01DkBRI

I see the bug in the Linux version where it completely locks up on bad sectors has not been fixed

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

Please provide following details:
- software edition (with version);
- OS edition;
- hard drive connection type;
- sequence of actions you did (with all parameters).
Because for drives with bad blocks on metadata area we recommend specific algorithm (we will prepare and publish instructions soon).

Here is the short version without me going back through exact steps because it is late here for me.

- software edition (with version); Latest version of UFS Explorer Pro 8.0 (Linux) downloaded a couple hours before this post.
- OS edition; Ubuntu 18.04.4
- hard drive connection type; SATA, and also as a generic block device through HDDSuperClone.
- sequence of actions you did (with all parameters). While exploring the file system of the NTFS partition on the drive, when expanding a certain folder the program will lock up indefinitely.

The Windows version does not have this issue, it only happens with the Linux version.

There are bad sectors in the MFT on the drive, which is likely causing the issue.

When using the generic block device through HDDSuperClone, I can see that it ends up in an endless loop trying to read a bad sector (or block). It will not stop trying to read the bad sector (block), the program must be terminated.

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

All you need to put pattern to bad-blocks areas with MRT, mount that drive to system , run UFS Pro, acivate feature of bad-block recognition by pattern and start saving data in advanced mode separating good/corrupted files . All folders hierarchy will be kept, bad files you will find in separate folder (problem), same like in Pc3k DE .

_________________
Angel Data Recovery

Dr Kiev ,
Is This Feature For Recovery Explorer Also ?

_________________
Regards
Amarbir S Dhillon , Chandigarh Data Recovery Labs [India]
Logical,Semi Physical And Physical Data Recovery
Website-> http://www.chandigarhdatarecovery.com

Not all, only part of them.
Recovery Explorer products are a cheaper and simpler alternative to UFS Explorer products starting from version 7.0. UFS Explorer is intended for data recovery professionals, while the Recovery Explorer audience is experienced users and system administrators.

Good to know, thank you. I was thinking of the same question Amarbir, so thanks for asking.

_________________
Hard Disk Drive (HDD), Solid State Drive (SSD, SATA, NVMe, etc), USB Flash Drive and RAID Data Recovery Specialist in Massachusetts

Please read article https://www.ufsexplorer.com/blog/work-w ... -drive.php
It describe our recommendation how to use our software.

I have replied to your PM and am offering TeamViewer access to a computer for a technician to view the issue. I am trying to give level 2-3 information about a software bug in the Linux version, but I feel I am only getting level 1 support replies. The Linux version gets stuck in an infinite loop trying to read a bad sector. Hopefully a TeamViewer session will help a technician see the issue.

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

Thanks

_________________
My firmware database:
https://mega.nz/folder/O01DkBRI

So the only reply I have got is to use the advanced SCSI connection method, which does bypass the issue (except with Recovery Explorer because that is not an available option). The issue is being blamed on an internal read issue, which is not the case. So I made a video to demonstrate the issue, since it will never get fixed because no one cares about a bug in the Linux version.

https://youtu.be/Sy5kk84urFc

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

We are working on this problem. We will let you know when bug will be fixed.

Hahaha, the end of the video is gold!

_________________
Joep - http://www.disktuna.com - video & photo repair & recovery service

Well,
When a developer of a tool reports a bug ,a developer on the other side should be speaking to him ,Why is this because some level of tech exchange is bound to happen ,Maximus is a awesome developer and researcher and if a low end tech guys speaks to him i would say its a insult ,So speak to him and try and solve the issue ,Using hddsuperclone and Recovery explorer or UFS Explorer should be a deadly combo for peeps who understand it .

PS : Maximus Keep Up The Good Work Buddie ....

_________________
Regards
Amarbir S Dhillon , Chandigarh Data Recovery Labs [India]
Logical,Semi Physical And Physical Data Recovery
Website-> http://www.chandigarhdatarecovery.com

That is how companies work. Their support is handled in levels, and unless your issue meets the criteria for the next level, it gets a generic response and otherwise ignored. I am not surprised by the difficulty to report a bug, but it definitely is frustrating.

Look at the symlink issue with R-Studio where the users want an option to totally disable symlinks. The only response has basically been “We can’t see a problem so we can’t fix it”. I don’t have any knowledge of the issue myself, but I have been following it and can feel the frustration of the users.

Thinking about it, maybe someone that can consistently reproduce the R-Studio symlink issue should make a video showing the issue. Just saying maybe it could get results.

And FYI, I have absolutely no problem with support with DMDE, but that is likely because it is very small like me, probably only a one person show, so you always talk to the developer/programmer.

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone