| #180: X-Ways Forensics,
X-Ways Investigator, WinHex 21.7 released
Feb 19, 2026 |
This mailing is to announce the
availability of version 21.7, with official release date Feb 17,
2026.
License owners please go to
https://www.x-ways.net/winhex/license.html
as always for the latest download instructions including the latest log-in
credentials (!), details about their licenses, and upgrade or renewal
offers. Please do not ask us for the download password. Your organization
has access to it already if eligible, as described.
Service releases are announced in the
Announcement section of the
forum,
and you can subscribe to instant e-mail notifications of postings in that
section if you have a forum profile. You can create such a profile
here
(if you have our log-in credentials). If you wish or need to stick with an
older version for a while, please switch to the latest service release of
that version.
Upcoming Training Events
| Dates |
Location |
Target Region |
Course |
Delivered by |
Mar 2-6 |
Online |
America, Europe |
X-Ways Forensics 2 |
X-Ways |
Mar 3-6 |
Salt Lake City, UT |
USA |
X-Ways Forensics 1 |
H-11 |
Mar 23-27 |
Online |
America, Europe |
X-Ways Forensics 1 |
X-Ways |
Apr 20-24 |
Online |
Europe, Asia |
X-Ways Forensics 2 |
X-Ways |
Apr 20-23 |
Davie, FL |
USA |
X-Ways Forensics 1 |
H-11 |
Apr 27-May 1 |
Online |
Europe, Asia |
X-Ways Forensics 1 |
X-Ways |
May 4-7 |
Guelph, ON |
Canada |
X-Ways Forensics 1 |
F111th |
May 11-15 |
Online |
America, Europe |
X-Ways Forensics 1 |
X-Ways |
May 12-15 |
Scottsdale, AZ |
USA |
X-Ways Forensics 1 |
H-11 |
Jun 1-5 |
Online |
Europe, Asia |
X-Ways Forensics 1 |
X-Ways |
Jun 8-12 |
Online |
America, Europe |
X-Ways Forensics 2 |
X-Ways |
Sep 14-18 |
Online |
Europe, Asia |
File Systems Revealed |
X-Ways |
Please sign up for our training notifications
here
if you would like to be kept posted on future training dates.
What's new in X‑Ways Forensics 21.7?
(where applicable, changes
also affect X‑Ways Investigator, WinHex, and X‑Ways Imager)
File Type Support
-
The play duration of certain video files that cannot
be determined and added to the Metadata column during the metadata extraction
step can now be extracted when capturing sporadic still images.
-
If you select multiple video files whose play
durations are known in the Metadata column, the total play duration of
all these videos combined is computed and shown below the directory
browser. This enables you and others (e.g. lawyers) to better understand
the amount of video data, for example to assess how complete the
coverage of surveillance videos is or to judge the amount of illegal
videos found, in a more meaningful way than measuring it in megabytes,
gigabytes or terabytes, especially for a computer layman.
-
Updated support for PNG, TIFF and WEBP files in
the internal graphics display library.
-
More pictures can now be identified as belonging to
the “No device” class, which are known to not have been generated by
optical input devices like cameras or scanners, but purely by software.
-
The propensity score in the summary table was
superseded with the introduction of the confidence about the device
class.
-
Self-extracting archives in the form of Windows PE
.exe files (if they are identified as type “sfx”) are now treated as
general-purpose archives and are thus explored along with ordinary
archives like Zip, RAR, and 7z, revealing their various sections, and
certificates if signed. The PE section that contains data that
can be interpreted as an embedded Zip or RAR archive is then usually
identified and processed as such.
-
Revised processing of .evtx event log files. Fixed
some parsing errors. More complete coverage of data types and output of
the Name attribute.
-
"Uncover embedded data..." now outputs all timestamps
found within BPLists as a separate type of event.
File System Support
-
Support for WofCompressed files in NTFS with resident storage.
-
Support for namespace extended attributes in Ext4 file systems.
-
More robust processing of certain corrupt directory
cluster chains in FAT file systems.
-
For more convenience, when starting off filling a
skeleton image by taking a new snapshot of an already open
volume/partition, a few sectors from the start of that volume/partition
are now included as well to enable the recipient to identify the most
common file systems. Note that you absolutely do not have to take a
volume snapshot and thus transfer all essential file system data
structures into the skeleton image. That could easily include a hundred
thousand names of files and directories names, which may or may not be
necessary or appropriate for your purpose. If you just need the contents
and some metadata of certain files in an NTFS file system for example,
you can specifically include the FILE records and contents of those
files, without the entire $MFT, and thanks to the inclusion of sector 0
(the boot sector) X-Ways Forensics will know what the file system and
the cluster size were, and can find the FILE records with a particular
thorough file system data structure search in the skeleton image
(quickly, thanks to the sparse nature of the image) and will therefore
know the storage locations and names and timestamps etc. of those files
in the volume.
-
A small number of sectors are no longer included in
skeleton images indirectly if they are only read for internal purposes
(e.g. to identify and highlight slack space area).
-
When creating a skeleton image, the contents of small
files that are stored within the $MFT system file can now be
automatically excluded from the acquisition when X-Ways Forensics reads
$MFT to take a volume snapshot. This may seem like a natural choice
since ordinary (larger) files are by default not included in the target
image either unless you specifically include them. However, this
involves redacting data within certain sectors and as such alters the
hash value of the affected sector range in the target image compared to
the source volume. As a compromise, if hashing is active, a second hash
value for the redacted data is included in the .log file, and that
second hash value is the one that is re-computed when you have X-Ways
Forensics verify the integrity of a skeleton image created with this new
option. Resident main file contents and resident alternative data
streams that share the same FILE record as storage space are excluded or
included together.
-
Adding selected files to a skeleton image will now
usually copy those files without slack space, i.e. trigger sector I/O
only for the logical file size.
-
After taking a volume snapshot of the subject volume
that is being acquired as a skeleton image, which includes the essential
file system data structures required to locate all file contents, the
user is now offered to revert to idle mode so that any subsequent random
read operations do not trigger acquisitions any more and the user can
freely click around and navigate in the directory browser and will only
specifically add file contents to the skeleton image using the dedicated
command in the directory browser context menu.
BitLocker Support
-
Informs the user if a fitting startup key for a
BitLocker volume is found in a .BEK file in the case directory and names
that file and where it was found.
-
On BitLocker volumes that it can decrypt, X-Ways
Forensics now tries to automatically detect unencrypted areas. Such
areas can be present if only in-use drive space was encrypted and
rewritten when the BitLocker volume was created, for example for
performance reasons or because the security implications of this were
not understood. If this situation is detected, X-Ways Forensics will
recommend running your analyses also on the undecrypted volume,
bypassing BitLocker decryption. For example a physical keyword search in
the undecrypted sectors in addition to a logical search in the files
found in the decrypted volume could be advisable.
-
There is a new command in the context menu of an
evidence object that is a BitLocker volume that X-Ways Forensics knows
how to decrypt. That command allows to open such a volume without
decrypting the data in any of its sectors, to see what data are
actually, literally stored in them. In that state you could run physical
searches or carve data automatically or manually. Not available in
X-Ways Investigator.
-
The file header signature search can now additionally
and automatically perform a second run on the data directly as stored in
a partition that is protected with BitLocker, bypassing the decryption
algorithm. Either only if the presence of unencrypted areas was detected
by X-Ways Forensics in the BitLocker volume (potentially just seconds
before during the first, regular run of the file header signature
search!) or, if fully checked, on any BitLocker volume that is processed
in its decrypted form.
-
X-Ways Forensics will specifically remember which
files were carved (automatically or manually) while BitLocker decryption
was bypassed so that those files in future can be read correctly even
when BitLocker decryption is otherwise active. The Description column
will identify such files. When working with the decrypted BitLocker
volume, switching between Volume/Partition and File mode for such files
will show the obvious difference between the data that are either passed
through the decryption algorithm in the former modes (falsely, because
it was never encrypted in the first place) or not in File mode.
Performance and Stability
-
Greatly accelerated loading of very large
Passwords.txt files.
-
The password collection in Passwords.txt can now be
tried to open BitLocker volumes using multiple threads for much better
performance.
-
Internal graphics display library thoroughly revised.
-
Does not waste time with certain unnecessary file
system I/O or opening compressed files when including selected files in
a hash set and the hash values can simply be taken from the volume
snapshot.
User Interface
-
The option to assign labels to a parent file now has
a tooltip that defines exactly what to expect: The next (closest) parent
object that is not a directory will be targeted. This option skips
parent directories and keeps looking until a file is found. If no file
is found upwards in the hierarchy, no label will be set.
-
A new related option was introduced, which targets
the so-called ultimate file. That is the parent object highest in the
hierarchy that is a file, i.e. the most aggregate file that indirectly
contains the data. Parent directories (in file or e-mail archives) can
be skipped over optionally. If not, then the last parent file
encountered before a directory will be considered the ultimate file. If
no file is found upwards in the hierarchy, the label will be set to the
selected item itself, if it is a file.
-
Another new option allows to simply assign label to
all the parent object files of a selected file, in a sequence that may
or may not be interrupted by directories. You could then decide later
for example based on file type which of those you actually need (e.g.
e-mails).
-
A new option allows to assign a label to the direct
parent object of a selected file, no matter whether it's a file or
directory.
-
Slightly revised look of the dialog window in which
labels are managed.
-
If a file is destined to appear in the case report
because it was assigned to a label that is includable in the report as a
report table, that file is now marked with a special icon in its name
cell, where also a yellow post-it icon appears if the file was commented
on. The icon for the report is displayed in a fainter color if the label
is not currently selected for output in the report options.
-
Omitting excluded child objects when printing is now
optional.
-
Some icons in the user interface were revised, for
the simultaneous search, copying extracted text, skeleton imaging and
running external programs.
Search Hit Lists and Event Lists
-
The search hit filter now allows to more precisely
define where in the context of a search hit an additional keyword is
required, either to the left or to the right of the search hit or both.
Also, an additional keyword can be required in the search hit itself.
That can be useful if the data in the search hit is variable for example
because it is based not on a fixed keywords, but on a regular expression
(e.g. to match e-mail addresses in general), or because the user has
shifted the offset of the search hit to the left or to the right to
cover related data that needs to be exported etc.
-
For both search hits and events there are now two
distinct menu commands to add items to the report and remove them. (For
search hits there was previously only a single menu command that toggled
that state.)
-
Selected events from all selected evidence objects
can now be included in the case report, near the end, in the order that
was last defined in an event list, e.g. sorted by timestamps for a
chronological timeline view. (Not in X-Ways Investigator.)
-
The description of individual events can now be
changed or set retroactively by the user, using the context menu. (Event
descriptions are currently limited to 255 bytes in UTF-8.).
Miscellaneous
-
Progress notifications can now optionally by output
into subdirectories that are named after the machine on which the X-Ways
Forensics session is running that produces these notifications.
-
Surrogate ASCII patterns for unreadable sectors on
storage devices with errors, redacted sectors in cleansed images etc.
are now prepended with an UTF-8 signature so that the latest version of
the viewer component will display such patterns when viewing or
previewing files that consist of only such text (interspersed with
binary zeroes), assuming that they are text files.
-
X-Tension API: The
XT_PREPARE_TARGETFILESWITHUNKNOWNDATA flag now forces XT_ProcessItem()
and XT_ProcessItemEx() calls for files with unsupported encryption or
compression.
-
Files in certain corrupt/incomplete archives can now
be opened with 0 bytes instead of not at all. That also means that the
X-Tension API function XT_ProcessItemEx() can now receive calls for such
files with (useless) handles.
-
The viewer component was last updated with patches on
our server for download on Nov 2, 2025.
-
The NSRL RDS hash sets, in a format for import into
X-Ways Forensics, have been updated to release 2025.12.1, and are
available for download from the resource directory in both MD5 and SHA-1
versions.
-
The program help and the user manual were updated.
-
Many minor improvements.
Changes of service releases of 21.6:
-
SR-1: Ability to display a rare JPEG variant.
-
SR-1: Fixed inability of the original v21.6 release
to open the same case with the same user account in cooperative mode
more than once (the second time as one's alter ego).
-
SR-1: Using only AND combinations of detections of
the picture content analysis for the categorization as notable did not
work because those combinations were lost. That was fixed.
-
SR-2: Avoided an unnecessary error message about the
creation of a temporary file at start-up in certain situations.
-
SR-2: The data density/compression statistics window
is now more likely in the visible range of a monitor with a low screen
resolution.
-
SR-2: Fixed an exception error that occurred when
computing ed2k along with any other hash value at the same time. (also
in v21.5 SR-10)
-
SR-2: Fixed decrementation of the remaining execution
count of insured dongles after automatic restarts. (also in v21.5 SR-10)
-
SR-2: Fixed device type dependent application of OCR
in certain situations. (also in v21.5 SR-10)
-
SR-3: Simple checksums that are computed on a
multi-byte accumulator, but byte-wise, are now presented in reverse hex
ASCII byte order again like in v21.4 and earlier.
-
SR-3: Fixed an exception error that could occur in
v21.6 when creating a new evidence file container.
-
SR-3: Works with more Tesseract versions.
-
SR-3: Navigating back to a parent file by
double-clicking the .. entry can no longer cause unintended viewing of
the file.
-
SR-3: Support for Windows 11 24H2 Prefetch files.
-
SR-3: Fixed an error in the Undo command in v21.6.
-
SR-3: The character adjustment feature did not work
for indexing in v21.6. That was fixed.
-
SR-4: Fixed decompression of certain WofCompressed
files in NTFS with non-resident storage.
-
SR-4: Support for longer paths and filenames in the
progress notification function.
-
SR-4: Fixed an error in the non-alternative method of
TAR archive extraction in v21.4 and later, which occurred with certain
TAR archives that contain nested archives.
-
SR-4: Fixed an error that caused certain e-mails to
be extracted from within MBOX archives with a size of 4 GB.
-
SR-4: Prevented potential separation of the [XT]
prefix and an actual message in the Messages window sent from an
X-Tension that could occur with multiple threads.
-
SR-5: Fixed a potential instability in mass picture
processing.
-
SR-6: SHA-512 was not usable as a hash for disk
imaging. That was fixed.
-
SR-6: Slightly more accurate representation of the
existence status of deleted files and directories in exFAT whose
respective first cluster is unknown.
-
SR-6: Fixed preview of some rare $I recycle bin files
with v8.5.7 of the viewer component.
-
SR-6: Fixed BitLocker-to-go FAT16 file system
detection.
-
SR-6: X-Tension API: The flags XT_PREPARE_DONTOMIT
and XT_PREPARE_TARGETFILESWITHUNKNOWNDATA combined now override the user
interface setting to omit files whose first cluster of original data is
known not to be available.
Become a certified user of X‑Ways Forensics
Become an
X-PERT (X‑Ways Professional in Evidence Recovery Techniques)
Prove your proficiency
in computer forensics in general and X‑Ways Forensics in particular with our
certification program. After passing the challenging exam, you will be part
of an exclusive circle and enjoy various benefits such as special
recognition, training discounts, updated training material. For further
details, please check
here.
Thank you for your attention! We hope to see you soon
somewhere at https://www.x-ways.net or
on our
Facebook page. You may also follow us on
Twitter/X. Please forward this newsletter to anyone who you think
will be interested. If you wish to subscribe with another e-mail address,
please do so
here.
Kind regards
Stefan Fleischmann
X‑Ways Software Technology AG
Carl-Diem-Str. 32 32257 Bünde Germany |