Merge pull request #1008 from volatilityfoundation/release/v2.5.0

Release/v2.5.0
This commit is contained in:
ikelos
2023-09-27 20:45:48 +01:00
committed by GitHub
43 changed files with 2030 additions and 333 deletions
+5 -1
View File
@@ -12,8 +12,12 @@ sphinx:
# Optionally build your docs in additional formats such as PDF and ePub
formats: all
build:
os: ubuntu-22.04
tools:
python: "3.11"
# Optionally set the version of Python and requirements required to build your docs
python:
version: 3.7
install:
- requirements: doc/requirements.txt
+4
View File
@@ -4,6 +4,10 @@ API Changes
When an addition to the existing API is made, the minor version is bumped.
When an API feature or function is removed or changed, the major version is bumped.
2.5.0
=====
Add in support for specifying a type override for object_from_symbol
2.4.0
=====
Add a `get_size()` method to Windows VAD structures and fix several off-by-one issues when calculating VAD sizes.
+5 -1
View File
@@ -1,4 +1,8 @@
# These packages are required for building the documentation.
sphinx>=4.0.0
sphinx>=4.0.0,<7
sphinx_autodoc_typehints>=1.4.0
sphinx-rtd-theme>=0.4.3
yara-python
pycryptodome
pefile
+58 -15
View File
@@ -1,7 +1,7 @@
Volatility 3 Basics
===================
Volatility splits memory analysis down to several components:
Volatility splits memory analysis down to several components. The main ones are:
* Memory layers
* Templates and Objects
@@ -13,22 +13,65 @@ which acts as a container for all the various layers and tables necessary to con
Memory layers
-------------
A memory layer is a body of data that can be accessed by requesting data at a specific address. Memory is seen as
sequential when accessed through sequential addresses, however, there is no obligation for the data to be stored
sequentially, and modern processors tend to store the memory in a paged format. Moreover, there is no need for the data
to be stored in an easily accessible format, it could be encoded or encrypted or more, it could be the combination of
two other sources. These are typically handled by programs that process file formats, or the memory manager of the
processor, but these are all translations (either in the geometric or linguistic sense) of the original data.
A memory layer is a body of data that can be accessed by requesting data at a specific address. At its lowest level
this data is stored on a phyiscal medium (RAM) and very early computers addresses locations in memory directly. However,
as the size of memory increased and it became more difficult to manage memory most architectures moved to a "paged" model
of memory, where the available memory is cut into specific fixed-sized pages. To help further, programs can ask for any address
and the processor will look up their (virtual) address in a map, to find out where the (physical) address that it lives at is,
in the actual memory of the system.
In Volatility 3 this is represented by a directed graph, whose end nodes are
:py:class:`DataLayers <volatility3.framework.interfaces.layers.DataLayerInterface>` and whose internal nodes are
specifically called a :py:class:`TranslationLayer <volatility3.framework.interfaces.layers.TranslationLayerInterface>`.
In this way, a raw memory image in the LiME file format and a page file can be
combined to form a single Intel virtual memory layer. When requesting addresses from the Intel layer, it will use the
Intel memory mapping algorithm, along with the address of the directory table base or page table map, to translate that
Volatility can work with these layers as long as it knows the map (so, for example that virtual address `1` looks up at physical
address `9`). The automagic that runs at the start of every volatility session often locates the kernel's memory map, and creates
a kernel virtual layer, which allows for kernel addresses to be looked up and the correct data returned. There can, however, be
several maps, and in general there is a different map for each process (although a portion of the operating system's memory is
usually mapped to the same location across all processes). The maps may take the same address but point to a different part of
physical memory. It also means that two processes could theoretically share memory, but having an virtual address mapped to the
same physical address as another process. See the worked example below for more information.
To translate an address on a layer, call :py:meth:`layer.mapping(offset, length, ignore_errors) <volatility3.framework.interfaces.layers.TranslationLayerInterface.mapping>` and it will return a list of chunks without overlap, in order,
for the requested range. If a portion cannot be mapped, an exception will be thrown unless `ignore_errors` is true. Each
chunk will contain the original offset of the chunk, the translated offset, the original size and the translated size of
the chunk, as well as the lower layer the chunk lives within.
Worked example
^^^^^^^^^^^^^^
The operating system and two programs may all appear to have access to all of physical memory, but actually the maps they each have
mean they each see something different:
.. code-block::
:caption: Memory mapping example
Operating system map Physical Memory
1 -> 9 1 - Free
2 -> 3 2 - OS.4, Process 1.4, Process 2.4
3 -> 7 3 - OS.2
4 -> 2 4 - Free
5 - Free
Process 1 map 6 - Process 1.2, Process 2.3
1 -> 12 7 - OS.3
2 -> 6 8 - Process1.3
3 -> 8 9 - OS.1
4 -> 2 10 - Process2.1
11 - Free
Process 2 map 12 - Process1.1
1 -> 10 13 - Free
2 -> 15 14 - Free
3 -> 6 15 - Process2.2
4 -> 2 16 - Free
In this example, part of the operating system is visible across all processes (although not all processes can write to the memory, there
is a permissions model for intel addressing which is not discussed further here).)
In Volatility 3 mappings are represented by a directed graph of layers, whose end nodes are
:py:class:`DataLayers <volatility3.framework.interfaces.layers.DataLayerInterface>` and whose internal nodes are :py:class:`TranslationLayers <volatility3.framework.interfaces.layers.TranslationLayerInterface>`.
In this way, a raw memory image in the LiME file format and a page file can be combined to form a single Intel virtual
memory layer. When requesting addresses from the Intel layer, it will use the Intel memory mapping algorithm, along
with the address of the directory table base or page table map, to translate that
address into a physical address, which will then either be directed towards the swap layer or the LiME layer. Should it
be directed towards the LiME layer, the LiME file format algorithm will be translated to determine where within the file
the data is stored and that will be returned.
be directed towards the LiME layer, the LiME file format algorithm will be translate the new address to determine where
within the file the data is stored. When the :py:meth:`layer.read() <volatility3.framework.interfaces.layers.TranslationLayerInterface.read>`
method is called, the translation is done automatically and the correct data gathered and combined.
.. note:: Volatility 2 had a similar concept, called address spaces, but these could only stack linearly one on top of another.
+99 -45
View File
@@ -21,57 +21,72 @@ import sphinx.ext.apidoc
def setup(app):
volatility_directory = os.path.abspath(os.path.join(os.path.dirname(__file__), '..', '..', 'volatility3'))
volatility_directory = os.path.abspath(
os.path.join(os.path.dirname(__file__), "..", "..", "volatility3")
)
source_dir = os.path.abspath(os.path.dirname(__file__))
sphinx.ext.apidoc.main(argv = ['-e', '-M', '-f', '-T', '-o', source_dir, volatility_directory])
sphinx.ext.apidoc.main(
argv=["-e", "-M", "-f", "-T", "-o", source_dir, volatility_directory]
)
# Go through the volatility3.framework.plugins files and change them to volatility3.plugins
for dir, _, files in os.walk(os.path.dirname(__file__)):
for filename in files:
if filename.startswith('volatility3.framework.plugins') and filename != 'volatility3.framework.plugins.rst':
if (
filename.startswith("volatility3.framework.plugins")
and filename != "volatility3.framework.plugins.rst"
):
# Change all volatility3.framework.plugins to volatility3.plugins in the file
# Rename the file
new_filename = filename.replace('volatility3.framework.plugins', 'volatility3.plugins')
new_filename = filename.replace(
"volatility3.framework.plugins", "volatility3.plugins"
)
replace_string = b"Submodules\n----------\n\n.. toctree::\n\n"
submodules = replace_string
# If file already exists, read out the subpackages entries from it add them to the new list
if os.path.exists(os.path.join(dir, new_filename)):
with open(os.path.join(dir, new_filename), 'rb') as newfile:
with open(os.path.join(dir, new_filename), "rb") as newfile:
data = newfile.read()
index = data.find(replace_string)
if index > -1:
submodules = data[index:]
with open(os.path.join(dir, new_filename), 'wb') as newfile:
with open(os.path.join(dir, new_filename), "wb") as newfile:
with open(os.path.join(dir, filename), "rb") as oldfile:
line = oldfile.read()
correct_plugins = line.replace(b'volatility3.framework.plugins', b'volatility3.plugins')
correct_submodules = correct_plugins.replace(replace_string, submodules)
correct_plugins = line.replace(
b"volatility3.framework.plugins", b"volatility3.plugins"
)
correct_submodules = correct_plugins.replace(
replace_string, submodules
)
newfile.write(correct_submodules)
os.remove(os.path.join(dir, filename))
elif filename == 'volatility3.framework.rst':
elif filename == "volatility3.framework.rst":
with open(os.path.join(dir, filename), "rb") as contents:
lines = contents.readlines()
plugins_seen = False
with open(os.path.join(dir, filename), "wb") as contents:
for line in lines:
if b'volatility3.framework.plugins' in line:
if b"volatility3.framework.plugins" in line:
plugins_seen = True
if plugins_seen and line == b'':
contents.write(b' volatility3.plugins')
if plugins_seen and line == b"":
contents.write(b" volatility3.plugins")
contents.write(line)
elif filename == 'volatility3.plugins.rst':
elif filename == "volatility3.plugins.rst":
with open(os.path.join(dir, filename), "rb") as contents:
lines = contents.readlines()
with open(os.path.join(dir, 'volatility3.framework.plugins.rst'), "rb") as contents:
with open(
os.path.join(dir, "volatility3.framework.plugins.rst"), "rb"
) as contents:
real_lines = contents.readlines()
# Process real_lines
for line_index in range(len(real_lines)):
if b'Submodules' in real_lines[line_index]:
if b"Submodules" in real_lines[line_index]:
break
else:
line_index = len(real_lines)
@@ -82,36 +97,52 @@ def setup(app):
for line in lines:
contents.write(line)
for line in submodule_lines:
contents.write(line.replace(b'volatility3.framework.plugins', b'volatility3.plugins'))
contents.write(
line.replace(
b"volatility3.framework.plugins", b"volatility3.plugins"
)
)
# Clear up the framework.plugins page
with open(os.path.join(os.path.dirname(__file__), 'volatility3.framework.plugins.rst'), "rb") as contents:
with open(
os.path.join(os.path.dirname(__file__), "volatility3.framework.plugins.rst"),
"rb",
) as contents:
real_lines = contents.readlines()
with open(os.path.join(os.path.dirname(__file__), 'volatility3.framework.plugins.rst'), "wb") as contents:
with open(
os.path.join(os.path.dirname(__file__), "volatility3.framework.plugins.rst"),
"wb",
) as contents:
for line in real_lines:
if b'volatility3.framework.plugins.' not in line:
if b"volatility3.framework.plugins." not in line:
contents.write(line)
# If extensions (or modules to document with autodoc) are in another directory,
# add these directories to sys.path here. If the directory is relative to the
# documentation root, use os.path.abspath to make it absolute, like shown here.
sys.path.insert(0, os.path.abspath('../..'))
sys.path.insert(0, os.path.abspath("../.."))
from volatility3.framework import constants
# -- General configuration ------------------------------------------------
# If your documentation needs a minimal Sphinx version, state it here.
needs_sphinx = '2.0'
needs_sphinx = "2.0"
# Add any Sphinx extension module names here, as strings. They can be
# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
# ones.
extensions = [
'sphinx.ext.autodoc', 'sphinx.ext.doctest', 'sphinx.ext.napoleon', 'sphinx.ext.intersphinx', 'sphinx.ext.todo',
'sphinx.ext.coverage', 'sphinx.ext.viewcode', 'sphinx.ext.autosectionlabel'
"sphinx.ext.autodoc",
"sphinx.ext.doctest",
"sphinx.ext.napoleon",
"sphinx.ext.intersphinx",
"sphinx.ext.todo",
"sphinx.ext.coverage",
"sphinx.ext.viewcode",
"sphinx.ext.autosectionlabel",
]
autosectionlabel_prefix_document = True
@@ -119,7 +150,7 @@ autosectionlabel_prefix_document = True
try:
import sphinx_autodoc_typehints
extensions.append('sphinx_autodoc_typehints')
extensions.append("sphinx_autodoc_typehints")
except ImportError:
# If the autodoc typehints extension isn't available, carry on regardless
pass
@@ -128,17 +159,17 @@ except ImportError:
# templates_path = ['tools/templates']
# The suffix of source filenames.
source_suffix = '.rst'
source_suffix = ".rst"
# The encoding of source files.
# source_encoding = 'utf-8-sig'
# The master toctree document.
master_doc = 'index'
master_doc = "index"
# General information about the project.
project = 'Volatility 3'
copyright = '2012-2022, Volatility Foundation'
project = "Volatility 3"
copyright = "2012-2022, Volatility Foundation"
# The version info for the project you're documenting, acts as replacement for
# |version| and |release|, also used in various other places throughout the
@@ -147,7 +178,7 @@ copyright = '2012-2022, Volatility Foundation'
# The full version, including alpha/beta/rc tags.
release = constants.PACKAGE_VERSION
# The short X.Y version.
version = ".".join(release.split('.')[0:2])
version = ".".join(release.split(".")[0:2])
# The language for content autogenerated by Sphinx. Refer to documentation
# for a list of supported languages.
@@ -180,7 +211,7 @@ add_module_names = False
# show_authors = False
# The name of the Pygments (syntax highlighting) style to use.
pygments_style = 'sphinx'
pygments_style = "sphinx"
# A list of ignored prefixes for module index sorting.
# modindex_common_prefix = []
@@ -196,8 +227,8 @@ pygments_style = 'sphinx'
# html_theme = 'pydoctheme'
# html_theme_options = {'collapsiblesidebar': True}
# html_theme_path = ['tools']
html_theme = 'sphinx_rtd_theme'
html_theme_options = {'logo_only': True}
html_theme = "sphinx_rtd_theme"
html_theme_options = {"logo_only": True}
# Theme options are theme-specific and customize the look and feel of a theme
# further. For a list of options available for each theme, see the
@@ -216,17 +247,17 @@ html_theme_options = {'logo_only': True}
# The name of an image file (relative to this directory) to place at the top
# of the sidebar.
html_logo = '_static/vol.png'
html_logo = "_static/vol.png"
# The name of an image file (within the static path) to use as favicon of the
# docs. This file should be a Windows icon file (.ico) being 16x16 or 32x32
# pixels large.
html_favicon = '_static/favicon.ico'
html_favicon = "_static/favicon.ico"
# Add any paths that contain custom static files (such as style sheets) here,
# relative to this directory. They are copied after the builtin static files,
# so a file named "default.css" will overwrite the builtin "default.css".
html_static_path = ['_static']
html_static_path = ["_static"]
# Add any extra paths that contain custom files (such as robots.txt or
# .htaccess) here, relative to this directory. These files are copied
@@ -275,17 +306,15 @@ html_static_path = ['_static']
# html_file_suffix = None
# Output file base name for HTML help builder.
htmlhelp_basename = 'Volatilitydoc'
htmlhelp_basename = "Volatilitydoc"
# -- Options for LaTeX output ---------------------------------------------
latex_elements = {
# The paper size ('letterpaper' or 'a4paper').
# 'papersize': 'letterpaper',
# The font size ('10pt', '11pt' or '12pt').
# 'pointsize': '10pt',
# Additional stuff for the LaTeX preamble.
# 'preamble': '',
}
@@ -294,7 +323,13 @@ latex_elements = {
# (source start file, target name, title,
# author, documentclass [howto, manual, or own class]).
latex_documents = [
('index', 'Volatility.tex', 'Volatility 3 Documentation', 'Volatility Foundation', 'manual'),
(
"index",
"Volatility.tex",
"Volatility 3 Documentation",
"Volatility Foundation",
"manual",
),
]
# The name of an image file (relative to this directory) to place at the top of
@@ -321,7 +356,15 @@ latex_documents = [
# One entry per manual page. List of tuples
# (source start file, name, description, authors, manual section).
man_pages = [('vol-cli', 'volatility', 'Volatility 3 Documentation', ['Volatility Foundation'], 1)]
man_pages = [
(
"vol-cli",
"volatility",
"Volatility 3 Documentation",
["Volatility Foundation"],
1,
)
]
# If true, show URL addresses after external links.
# man_show_urls = False
@@ -332,8 +375,15 @@ man_pages = [('vol-cli', 'volatility', 'Volatility 3 Documentation', ['Volatilit
# (source start file, target name, title, author,
# dir menu entry, description, category)
texinfo_documents = [
('index', 'Volatility', 'Volatility 3 Documentation', 'Volatility Foundation', 'Volatility',
'Memory forensics framework.', 'Miscellaneous'),
(
"index",
"Volatility",
"Volatility 3 Documentation",
"Volatility Foundation",
"Volatility",
"Memory forensics framework.",
"Miscellaneous",
),
]
# Documents to append as an appendix to all manuals.
@@ -349,10 +399,14 @@ texinfo_documents = [
# texinfo_no_detailmenu = False
# Example configuration for intersphinx: refer to the Python standard library.
intersphinx_mapping = {'http://docs.python.org/': None}
intersphinx_mapping = {"python": ("http://docs.python.org/", None)}
# -- Autodoc options -------------------------------------------------------
# autodoc_member_order = 'groupwise'
autodoc_default_options = {'members': True, 'inherited-members': True, 'show-inheritance': True}
autoclass_content = 'both'
autodoc_default_options = {
"members": True,
"inherited-members": True,
"show-inheritance": True,
}
autoclass_content = "both"
+153
View File
@@ -0,0 +1,153 @@
macOS Tutorial
==============
This guide will give you a brief overview of how volatility3 works as well as a demonstration of several of the plugins available in the suite.
Acquiring memory
----------------
Volatility3 does not provide the ability to acquire memory. The example below is an open source tool. Other commercial tools are also available.
* `osxpmem <https://github.com/Velocidex/c-aff4/releases/download/3.2/osxpmem_3.2.zip>`_
Procedure to create symbol tables for macOS
--------------------------------------------
To create a symbol table please refer to :ref:`symbol-tables:Mac or Linux symbol tables`.
.. tip:: It may be possible to locate pre-made ISF files from the `download link <https://downloads.volatilityfoundation.org/volatility3/symbols/mac.zip>`_ ,
which is built and maintained by `volatilityfoundation <https://www.volatilityfoundation.org/>`_.
After creating the file or downloading it from the link, place the file under the directory ``volatility3/symbols/``.
Listing plugins
---------------
The following is a sample of the macOS plugins available for volatility3, it is not complete and more plugins may
be added. For a complete reference, please see the volatility 3 :doc:`list of plugins <volatility3.plugins>`.
For plugin requests, please create an issue with a description of the requested plugin.
.. code-block:: shell-session
$ python3 vol.py --help | grep -i mac. | head -n 4
mac.bash.Bash Recovers bash command history from memory.
mac.check_syscall.Check_syscall
mac.check_sysctl.Check_sysctl
mac.check_trap_table.Check_trap_table
.. note:: Here the the command is piped to grep and head in-order to provide the start of the list of macOS plugins.
Using plugins
-------------
The following is the syntax to run the volatility CLI.
.. code-block:: shell-session
$ python3 vol.py -f <path to memory image> <plugin_name> <plugin_option>
Example
-------
banners
~~~~~~~
In this example we will be using a memory dump from the Securinets CTF Quals 2019 Challenge called Contact_me. We will limit the discussion to memory forensics with volatility 3 and not extend it to other parts of the challenge.
Thanks go to `stuxnet <https://github.com/stuxnet999/>`_ for providing this memory dump and `writeup <https://stuxnet999.github.io/securinets-ctf/2019/08/24/SecurinetsQuals2019-Contact-Me.html>`_.
.. code-block:: shell-session
$ python3 vol.py -f contact_me banners.Banners
Volatility 3 Framework 2.4.2
Progress: 100.00 PDB scanning finished
Offset Banner
0x4d2c7d0 Darwin Kernel Version 16.7.0: Thu Jun 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64
0xb42b180 Darwin Kernel Version 16.7.0: Thu Jun 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64
0xcda9100 Darwin Kernel Version 16.7.0: Thu Jun 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64
0x1275e7d0 Darwin Kernel Version 16.7.0: Thu Jun 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64
0x1284fba4 Darwin Kernel Version 16.7.0: Thu Jun 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64
0x34ad0180 Darwin Kernel Version 16.7.0: Thu Jun 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64
The above command helps us to find the memory dump's Darwin kernel version. Now using the above banner we can search for the needed ISF file.
If ISF file cannot be found then, follow the instructions on :ref:`getting-started-mac-tutorial:Procedure to create symbol tables for macOS`. After that, place the ISF file under the ``volatility3/symbols`` directory.
mac.pslist
~~~~~~~~~~
.. code-block:: shell-session
$ python3 vol.py -f contact_me mac.pslist.PsList
Volatility 3 Framework 2.4.2
Progress: 100.00 Stacking attempts finished
PID PPID COMM
0 0 kernel_task
1 0 launchd
35 1 UserEventAgent
38 1 kextd
39 1 fseventsd
37 1 uninstalld
45 1 configd
46 1 powerd
52 1 logd
58 1 warmd
.....
``mac.pslist`` helps us to list the processes which are running, their PIDs and PPIDs.
mac.pstree
~~~~~~~~~~
.. code-block:: shell-session
$ python3 vol.py -f contact_me mac.pstree.PsTree
Volatility 3 Framework 2.4.2
Progress: 100.00 Stacking attempts finished
PID PPID COMM
35 1 UserEventAgent
38 1 kextd
39 1 fseventsd
37 1 uninstalld
204 1 softwareupdated
* 449 204 SoftwareUpdateCo
337 1 system_installd
* 455 337 update_dyld_shar
``mac.pstree`` helps us to display the parent child relationships between processes.
mac.ifconfig
~~~~~~~~~~~~
.. code-block:: shell-session
$ python3 vol.py -f contact_me mac.ifconfig.Ifconfig
Volatility 3 Framework 2.4.2
Progress: 100.00 Stacking attempts finished
Interface IP Address Mac Address Promiscuous
lo0 False
lo0 127.0.0.1 False
lo0 ::1 False
lo0 fe80:1::1 False
gif0 False
stf0 False
en0 00:0C:29:89:8B:F0 00:0C:29:89:8B:F0 False
en0 fe80:4::10fb:c89d:217f:52ae 00:0C:29:89:8B:F0 False
en0 192.168.140.128 00:0C:29:89:8B:F0 False
utun0 False
utun0 fe80:5::2a95:bb15:87e3:977c False
we can use the ``mac.ifconfig`` plugin to get information about the configuration of the network interfaces of the host under investigation.
+1
View File
@@ -25,6 +25,7 @@ There is also some information to get you started quickly:
:caption: Getting Started
getting-started-linux-tutorial
getting-started-mac-tutorial
getting-started-windows-tutorial
+4 -3
View File
@@ -67,9 +67,10 @@ return a dictionary of plugin names and the plugin classes.
Determine what configuration options a plugin requires
------------------------------------------------------
For each plugin class, we can call the classmethod `requirements` on it, which will return a list of objects that
adhere to the :py:class:`~volatility3.framework.interfaces.configuration.RequirementInterface` method. The various
types of Requirement are split roughly in two,
For each plugin class, we can call the classmethod
:py:func:`~volatility3.framework.interfaces.configuration.ConfigurableInterface.get_requirements` on it, which will
return a list of objects that adhere to the :py:class:`~volatility3.framework.interfaces.configuration.RequirementInterface`
method. The various types of Requirement are split roughly in two,
:py:class:`~volatility3.framework.interfaces.configuration.SimpleTypeRequirement` (such as integers, booleans, floats
and strings) and more complex requirements (such as lists, choices, multiple requirements, translation layer
requirements or symbol table requirements). A requirement just specifies a type of data and a name, and must be
+5 -2
View File
@@ -38,10 +38,13 @@ class WarningFindSpec(abc.MetaPathFinder):
"""Mock find_spec method that just checks the name, this must go
first."""
if fullname.startswith("volatility3.framework.plugins."):
warning = "Please do not use the volatility3.framework.plugins namespace directly, only use volatility3.plugins"
warning = f"Import {fullname}: Please do not use the volatility3.framework.plugins namespace directly, only use volatility3.plugins"
# Pyinstaller uses walk_packages/_collect_submodules to import, but needs to read the modules to figure out dependencies
# As such, we only print the warning when directly imported rather than from within walk_packages/_collect_submodules
if inspect.stack()[-2].function in ["walk_packages", "_collect_submodules"]:
if inspect.stack()[-2].function not in [
"walk_packages",
"_collect_submodules",
] and inspect.stack()[-3].function not in ["_collect_submodules"]:
raise Warning(warning)
+7 -14
View File
@@ -353,7 +353,9 @@ class CommandLine:
###
if args.file:
try:
single_location = self.location_from_file(args.file)
single_location = requirements.URIRequirement.location_from_file(
args.file
)
ctx.config["automagic.LayerStacker.single_location"] = single_location
except ValueError as excp:
parser.error(str(excp))
@@ -456,19 +458,10 @@ class CommandLine:
Returns:
The URL for the location of the file
"""
# We want to work in URLs, but we need to accept absolute and relative files (including on windows)
single_location = parse.urlparse(filename, "")
if single_location.scheme == "" or len(single_location.scheme) == 1:
single_location = parse.urlparse(
parse.urljoin("file:", request.pathname2url(os.path.abspath(filename)))
)
if single_location.scheme == "file":
if not os.path.exists(request.url2pathname(single_location.path)):
filename = request.url2pathname(single_location.path)
if not filename:
raise ValueError("File URL looks incorrect (potentially missing /)")
raise ValueError(f"File does not exist: {filename}")
return parse.urlunparse(single_location)
vollog.debug(
f"{__name__}.location_from_file has been deprecated and moved to requirements.URIRequirement.location_from_file"
)
return requirements.URIRequirement.location_from_file(filename)
def process_exceptions(self, excp):
"""Provide useful feedback if an exception occurs during a run of a plugin."""
+1 -1
View File
@@ -480,7 +480,7 @@ class Volshell(interfaces.plugins.PluginInterface):
accessor = resources.ResourceAccessor()
with accessor.open(url=location) as fp:
self.__console.runsource(
io.TextIOWrapper(fp.read(), encoding="utf-8"), symbol="exec"
io.TextIOWrapper(fp, encoding="utf-8").read(), symbol="exec"
)
print("\nCode complete")
+10 -6
View File
@@ -26,6 +26,7 @@ import importlib
import inspect
import logging
import os
import traceback
from typing import Any, Dict, Generator, List, Tuple, Type, TypeVar
from volatility3.framework import constants, interfaces
@@ -183,7 +184,11 @@ def import_file(module: str, path: str, ignore_errors: bool = False) -> List[str
try:
importlib.import_module(module)
except ImportError as e:
vollog.debug(str(e))
vollog.debug(
"".join(
traceback.TracebackException.from_exception(e).format(chain=True)
)
)
vollog.debug(
"Failed to import module {} based on file: {}".format(module, path)
)
@@ -219,8 +224,7 @@ def list_plugins() -> Dict[str, Type[interfaces.plugins.PluginInterface]]:
def clear_cache(complete=False):
glob_pattern = "*.cache"
if not complete:
glob_pattern = "data_" + glob_pattern
for cache_filename in glob.glob(os.path.join(constants.CACHE_PATH, glob_pattern)):
os.unlink(cache_filename)
try:
os.unlink(os.path.join(constants.CACHE_PATH, constants.IDENTIFIERS_FILENAME))
except FileNotFoundError:
vollog.log(constants.LOGLEVEL_VVVV, "Attempting to clear a non-existant cache")
@@ -310,6 +310,14 @@ class SqliteCache(CacheManagerInterface):
new_locations = on_disk_locations.difference(cached_locations)
missing_locations = cached_locations.difference(on_disk_locations)
# Missing entries
if missing_locations:
self._database.cursor().execute(
f"DELETE FROM cache WHERE location IN ({','.join(['?'] * len(missing_locations))})",
[x for x in missing_locations],
)
self._database.commit()
cache_update = set()
files_to_timestamp = on_disk_locations.intersection(cached_locations)
if files_to_timestamp:
@@ -437,15 +445,6 @@ class SqliteCache(CacheManagerInterface):
progress_callback(100, "Reading remote ISF list")
self._database.commit()
# Missing entries
if missing_locations:
self._database.cursor().execute(
f"DELETE FROM cache WHERE location IN ({','.join(['?'] * len(missing_locations))})",
[x for x in missing_locations],
)
self._database.commit()
def get_identifier_dictionary(
self, operating_system: Optional[str] = None, local_only: bool = False
) -> Dict[bytes, str]:
+14 -1
View File
@@ -367,6 +367,7 @@ class WinSwapLayers(interfaces.automagic.AutomagicInterface):
progress_callback: constants.ProgressCallback = None,
) -> None:
"""Finds translation layers that can have swap layers added."""
path_join = interfaces.configuration.path_join
self._translation_requirement = self.find_requirements(
context,
@@ -382,11 +383,13 @@ class WinSwapLayers(interfaces.automagic.AutomagicInterface):
swap_sub_config, swap_req = self.find_swap_requirement(
trans_sub_config, trans_req
)
counter = 0
swap_config = interfaces.configuration.parent_path(swap_sub_config)
if swap_req and swap_req.unsatisfied(context, swap_config):
# See if any of them need constructing
for swap_location in self.config.get("single_swap_locations", []):
# Setup config locations/paths
current_layer_name = swap_req.name + str(counter)
@@ -398,7 +401,17 @@ class WinSwapLayers(interfaces.automagic.AutomagicInterface):
# Fill in the config
if swap_location:
context.config[current_layer_path] = current_layer_name
context.config[layer_loc_path] = swap_location
try:
context.config[
layer_loc_path
] = requirements.URIRequirement.location_from_file(
swap_location
)
except ValueError:
vollog.warning(
f"Volatility swap_location {swap_location} could not be validated - swap layer disabled"
)
continue
context.config[
layer_class_path
] = "volatility3.framework.layers.physical.FileLayer"
@@ -10,7 +10,9 @@ expect to be in the context (such as particular layers or symboltables).
"""
import abc
import logging
import os
from typing import Any, ClassVar, Dict, List, Optional, Tuple, Type
from urllib import parse, request
from volatility3.framework import constants, interfaces
@@ -55,6 +57,30 @@ class URIRequirement(StringRequirement):
# TODO: Maybe a a check that to unsatisfied that the path really is a URL?
@classmethod
def location_from_file(cls, filename: str) -> str:
"""Returns the URL location from a file parameter (which may be a URL)
Args:
filename: The path to the file (either an absolute, relative, or URL path)
Returns:
The URL for the location of the file
"""
# We want to work in URLs, but we need to accept absolute and relative files (including on windows)
single_location = parse.urlparse(filename, "")
if single_location.scheme == "" or len(single_location.scheme) == 1:
single_location = parse.urlparse(
parse.urljoin("file:", request.pathname2url(os.path.abspath(filename)))
)
if single_location.scheme == "file":
if not os.path.exists(request.url2pathname(single_location.path)):
filename = request.url2pathname(single_location.path)
if not filename:
raise ValueError("File URL looks incorrect (potentially missing /)")
raise ValueError(f"File does not exist: {filename}")
return parse.urlunparse(single_location)
class BytesRequirement(interfaces.configuration.SimpleTypeRequirement):
"""A requirement type that contains a byte string."""
+2 -2
View File
@@ -44,8 +44,8 @@ BANG = "!"
# We use the SemVer 2.0.0 versioning scheme
VERSION_MAJOR = 2 # Number of releases of the library with a breaking change
VERSION_MINOR = 4 # Number of changes that only add to the interface
VERSION_PATCH = 1 # Number of changes that do not change the interface
VERSION_MINOR = 5 # Number of changes that only add to the interface
VERSION_PATCH = 0 # Number of changes that do not change the interface
VERSION_SUFFIX = ""
# TODO: At version 2.0.0, remove the symbol_shift feature
@@ -234,3 +234,48 @@ BLUETOOTH_PROTOCOLS = (
"HIDP",
"AVDTP",
)
# Ref: include/uapi/linux/capability.h
CAPABILITIES = (
"chown",
"dac_override",
"dac_read_search",
"fowner",
"fsetid",
"kill",
"setgid",
"setuid",
"setpcap",
"linux_immutable",
"net_bind_service",
"net_broadcast",
"net_admin",
"net_raw",
"ipc_lock",
"ipc_owner",
"sys_module",
"sys_rawio",
"sys_chroot",
"sys_ptrace",
"sys_pacct",
"sys_admin",
"sys_boot",
"sys_nice",
"sys_resource",
"sys_time",
"sys_tty_config",
"mknod",
"lease",
"audit_write",
"audit_control",
"setfcap",
"mac_override",
"mac_admin",
"syslog",
"wake_alarm",
"block_suspend",
"audit_read",
"perfmon",
"bpf",
"checkpoint_restore",
)
+16 -4
View File
@@ -272,6 +272,7 @@ class Module(interfaces.context.ModuleInterface):
symbol_name: str,
native_layer_name: Optional[str] = None,
absolute: bool = False,
object_type: Optional[Union[str, "interfaces.objects.ObjectInterface"]] = None,
**kwargs,
) -> "interfaces.objects.ObjectInterface":
"""Returns an object based on a specific symbol (containing type and
@@ -284,6 +285,7 @@ class Module(interfaces.context.ModuleInterface):
symbol_name: Name of the symbol (within the module) to construct
native_layer_name: Name of the layer in which constructed objects are made (for pointers)
absolute: whether the symbol's address is absolute or relative to the module
object_type: Override for the type from the symobl to use (or if the symbol type is missing)
"""
if constants.BANG not in symbol_name:
symbol_name = self.symbol_table_name + constants.BANG + symbol_name
@@ -299,8 +301,13 @@ class Module(interfaces.context.ModuleInterface):
if not absolute:
offset += self._offset
if symbol_val.type is None:
raise TypeError(f"Symbol {symbol_val.name} has no associated type")
if object_type is None:
if symbol_val.type is None:
raise TypeError(
f"Symbol {symbol_val.name} has no associated type and no object_type specified"
)
else:
object_type = symbol_val.type
# Ensure we don't use a layer_name other than the module's, why would anyone do that?
if "layer_name" in kwargs:
@@ -308,7 +315,7 @@ class Module(interfaces.context.ModuleInterface):
# Since type may be a template, we don't just call our own module method
return self._context.object(
object_type=symbol_val.type,
object_type=object_type,
layer_name=self._layer_name,
offset=offset,
native_layer_name=native_layer_name or self._native_layer_name,
@@ -381,6 +388,7 @@ class ModuleCollection(interfaces.context.ModuleContainer):
def __init__(
self, modules: Optional[List[interfaces.context.ModuleInterface]] = None
) -> None:
self._prefix_count = {}
super().__init__(modules)
def deduplicate(self) -> "ModuleCollection":
@@ -400,9 +408,13 @@ class ModuleCollection(interfaces.context.ModuleContainer):
def free_module_name(self, prefix: str = "module") -> str:
"""Returns an unused module name"""
count = 1
if prefix not in self._prefix_count:
self._prefix_count[prefix] = 1
return prefix
count = self._prefix_count[prefix]
while prefix + str(count) in self:
count += 1
self._prefix_count[prefix] = count
return prefix + str(count)
@property
@@ -253,6 +253,7 @@ class ModuleInterface(interfaces.configuration.ConfigurableInterface):
symbol_name: str,
native_layer_name: Optional[str] = None,
absolute: bool = False,
object_type: Optional[Union[str, "interfaces.objects.ObjectInterface"]] = None,
**kwargs,
) -> "interfaces.objects.ObjectInterface":
"""Returns an object created using the symbol_table_name and layer_name
@@ -262,6 +263,7 @@ class ModuleInterface(interfaces.configuration.ConfigurableInterface):
symbol_name: The name of a symbol (that must be present in the module's symbol table). The symbol's associated type will be used to construct an object at the symbol's offset.
native_layer_name: The native layer for objects that reference a different layer (if not the default provided during module construction)
absolute: A boolean specifying whether the offset is absolute within the layer, or relative to the start of the module
object_type: Override for the type from the symobl to use (or if the symbol type is missing)
Returns:
The constructed object
+9 -2
View File
@@ -19,11 +19,18 @@ vollog = logging.getLogger(__name__)
try:
# TODO: Find library for windows if needed
try:
# Linux/Mac
# Linux
lib_snappy = ctypes.cdll.LoadLibrary("libsnappy.so.1")
except OSError:
lib_snappy = None
try:
if not lib_snappy:
# macOS
lib_snappy = ctypes.cdll.LoadLibrary("libsnappy.1.dylib")
except OSError:
lib_snappy = None
try:
if not lib_snappy:
# Windows 64
@@ -31,7 +38,7 @@ try:
except OSError:
lib_snappy = None
if lib_snappy:
if not lib_snappy:
# Windows 32
lib_snappy = ctypes.cdll.LoadLibrary("snappy32")
+9
View File
@@ -111,6 +111,11 @@ class Intel(linear.LinearlyMappedLayer):
"""Returns whether a particular page is valid based on its entry."""
return bool(entry & 1)
@staticmethod
def _page_is_dirty(entry: int) -> bool:
"""Returns whether a particular page is dirty based on its entry."""
return bool(entry & (1 << 6))
def canonicalize(self, addr: int) -> int:
"""Canonicalizes an address by performing an appropiate sign extension on the higher addresses"""
if self._bits_per_register <= self._maxvirtaddr:
@@ -259,6 +264,10 @@ class Intel(linear.LinearlyMappedLayer):
except exceptions.InvalidAddressException:
return False
def is_dirty(self, offset: int) -> bool:
"""Returns whether the page at offset is marked dirty"""
return self._page_is_dirty(self._translate_entry(offset)[0])
def mapping(
self, offset: int, length: int, ignore_errors: bool = False
) -> Iterable[Tuple[int, int, int, int, str]]:
+180
View File
@@ -0,0 +1,180 @@
import logging
import struct
from typing import Optional
from volatility3.framework import constants, interfaces, exceptions
from volatility3.framework.layers import elf
from volatility3.framework.symbols import intermed
vollog = logging.getLogger(__name__)
class XenCoreDumpLayer(elf.Elf64Layer):
"""A layer that supports the Xen Dump-Core format as documented at: https://xenbits.xen.org/docs/4.6-testing/misc/dump-core-format.txt"""
_header_struct = struct.Struct("<IBBB")
MAGIC = 0x464C457F # "\x7fELF"
ELF_CLASS = 2
def __init__(
self, context: interfaces.context.ContextInterface, config_path: str, name: str
) -> None:
# Create a custom SymbolSpace
self._elf_table_name = intermed.IntermediateSymbolTable.create(
context, config_path, "linux", "elf"
)
self._xen_table_name = intermed.IntermediateSymbolTable.create(
context, config_path, "linux", "xen"
)
self._segment_headers = {}
super().__init__(context, config_path, name)
def _extract_result_array(
self, varname: str, segment_index: int
) -> interfaces.objects.ObjectInterface:
hdr = self._segment_headers[segment_index]
result = self.context.object(
self._xen_table_name + constants.BANG + varname,
layer_name=self._base_layer,
offset=hdr.sh_offset,
size=hdr.sh_size,
)
result.entries.count = hdr.sh_size // result.entries.vol.subtype.size
return result
def _load_segments(self) -> None:
"""Load the segments from based on the PT_LOAD segments of the Elf64 format"""
ehdr = self.context.object(
self._elf_table_name + constants.BANG + "Elf64_Ehdr",
layer_name=self._base_layer,
offset=0,
)
segments = []
self._segment_headers = []
for sindex in range(ehdr.e_shnum):
shdr = self.context.object(
self._elf_table_name + constants.BANG + "Elf64_Shdr",
layer_name=self._base_layer,
offset=ehdr.e_shoff + (sindex * ehdr.e_shentsize),
)
self._segment_headers.append(shdr)
if sindex == ehdr.e_shstrndx:
segment_names = self.context.layers[self._base_layer].read(
shdr.sh_offset, shdr.sh_size
)
segment_names = segment_names.split(b"\x00")
if not segment_names:
raise elf.ElfFormatException("No segment names, not a Xen Core Dump")
try:
p2m_data = self._extract_result_array(
"xen_p2m", segment_names.index(b".xen_p2m")
)
except ValueError:
p2m_data = None
try:
pfn_data = self._extract_result_array(
"xen_pfn", segment_names.index(b".xen_pfn")
)
except ValueError:
pfn_data = None
pages_hdr = self._segment_headers[segment_names.index(b".xen_pages")]
page_size = 0x1000
if pfn_data and not p2m_data:
for entry_index in range(len(pfn_data.entries)):
entry = pfn_data.entries[entry_index]
# TODO: Don't hardcode the maximum value here
if entry and entry != 0xFFFFFFFF:
segments.append(
(
entry * page_size,
pages_hdr.sh_offset + (entry_index * page_size),
page_size,
page_size,
)
)
elif p2m_data and not pfn_data:
for entry_index in range(len(p2m_data.entries)):
entry = p2m_data.entries[entry_index]
# TODO: Don't hardcode the maximum value here
if entry.pfn != 0xFFFFFFFF:
segments.append(
(
entry.pfn * page_size,
pages_hdr.sh_offset + (entry_index * page_size),
page_size,
page_size,
)
)
elif p2m_data and pfn_data:
raise elf.ElfFormatException(
self.name, f"Both P2M and PFN in Xen Core Dump"
)
else:
raise elf.ElfFormatException(
self.name, f"Neither P2M nor PFN in Xen Core Dump"
)
if len(segments) == 0:
raise elf.ElfFormatException(
self.name, f"No ELF segments defined in {self._base_layer}"
)
self._segments = segments
@classmethod
def _check_header(
cls, base_layer: interfaces.layers.DataLayerInterface, offset: int = 0
) -> bool:
try:
header_data = base_layer.read(offset, cls._header_struct.size)
except exceptions.InvalidAddressException:
raise elf.ElfFormatException(
base_layer.name,
f"Offset 0x{offset:0x} does not exist within the base layer",
)
(magic, elf_class, elf_data_encoding, elf_version) = cls._header_struct.unpack(
header_data
)
if magic != cls.MAGIC:
raise elf.ElfFormatException(
base_layer.name, f"Bad magic 0x{magic:x} at file offset 0x{offset:x}"
)
if elf_class != cls.ELF_CLASS:
raise elf.ElfFormatException(
base_layer.name, f"ELF class is not 64-bit (2): {elf_class:d}"
)
# Virtualbox uses an ELF version of 0, which isn't to specification, but is ok to deal with
return True
class XenCoreDumpStacker(elf.Elf64Stacker):
stack_order = 10
@classmethod
def stack(
cls,
context: interfaces.context.ContextInterface,
layer_name: str,
progress_callback: constants.ProgressCallback = None,
) -> Optional[interfaces.layers.DataLayerInterface]:
try:
if not XenCoreDumpLayer._check_header(context.layers[layer_name]):
return None
except elf.ElfFormatException as excp:
vollog.log(constants.LOGLEVEL_VVVV, f"Exception: {excp}")
return None
new_name = context.layers.free_layer_name("XenCoreDumpLayer")
context.config[
interfaces.configuration.path_join(new_name, "base_layer")
] = layer_name
return XenCoreDumpLayer(context, new_name, new_name)
@@ -0,0 +1,208 @@
# This file is Copyright 2023 Volatility Foundation and licensed under the Volatility Software License 1.0
# which is available at https://www.volatilityfoundation.org/license/vsl-v1.0
#
import logging
from dataclasses import dataclass, astuple, fields
from typing import Iterable, List, Tuple
from volatility3.framework import interfaces, renderers, exceptions
from volatility3.framework.configuration import requirements
from volatility3.framework.interfaces import plugins
from volatility3.framework.objects import utility
from volatility3.framework.symbols.linux import extensions
from volatility3.plugins.linux import pslist
vollog = logging.getLogger(__name__)
@dataclass
class TaskData:
"""Stores basic information about a task"""
comm: str
pid: int
tgid: int
ppid: int
euid: int
@dataclass
class CapabilitiesData:
"""Stores each set of capabilties for a task"""
cap_inheritable: interfaces.objects.ObjectInterface
cap_permitted: interfaces.objects.ObjectInterface
cap_effective: interfaces.objects.ObjectInterface
cap_bset: interfaces.objects.ObjectInterface
cap_ambient: interfaces.objects.ObjectInterface
def astuple(self) -> Tuple:
"""Returns a shallow copy of the capability sets in a tuple.
Otherwise, when dataclasses.astuple() performs a deep-copy recursion on
ObjectInterface will take a substantial amount of time.
"""
return tuple(getattr(self, field.name) for field in fields(self))
class Capabilities(plugins.PluginInterface):
"""Lists process capabilities"""
_required_framework_version = (2, 0, 0)
_version = (1, 0, 0)
@classmethod
def get_requirements(cls) -> List[interfaces.configuration.RequirementInterface]:
return [
requirements.ModuleRequirement(
name="kernel",
description="Linux kernel",
architectures=["Intel32", "Intel64"],
),
requirements.PluginRequirement(
name="pslist", plugin=pslist.PsList, version=(2, 0, 0)
),
requirements.ListRequirement(
name="pids",
description="Filter on specific process IDs.",
element_type=int,
optional=True,
),
]
def _check_capabilities_support(
self,
context: interfaces.context.ContextInterface,
vmlinux_module_name: str,
):
"""Checks that the framework supports at least as much capabilities as
the kernel being analysed. Otherwise, it shows a warning for the
developers.
"""
vmlinux = context.modules[vmlinux_module_name]
try:
kernel_cap_last_cap = vmlinux.object_from_symbol(symbol_name="cap_last_cap")
except exceptions.SymbolError:
# It should be a kernel < 3.2
return
vol2_last_cap = extensions.kernel_cap_struct.get_last_cap_value()
if kernel_cap_last_cap > vol2_last_cap:
vollog.warning(
"Developers: The supported Linux capabilities of this plugin are outdated for this kernel"
)
@staticmethod
def _decode_cap(cap: interfaces.objects.ObjectInterface) -> str:
"""Returns a textual representation of the capability set.
The format is a comma-separated list of capabilitites. In order to
summarize the output and if all the capabilities are enabled, instead of
the individual capabilities, the special name "all" will be shown.
Args:
cap: Kernel capability object. Usually a 'kernel_cap_struct' struct
Returns:
str: A string with a comma separated list of decoded capabilities
"""
if isinstance(cap, renderers.NotAvailableValue):
return cap
cap_value = cap.get_capabilities()
if not cap_value:
return ""
if cap_value == cap.get_kernel_cap_full():
return "all"
return ", ".join(cap.enumerate_capabilities())
@classmethod
def get_task_capabilities(
cls, task: interfaces.objects.ObjectInterface
) -> Tuple[TaskData, CapabilitiesData]:
"""Returns a tuple with the task basic information along with its capabilities
Args:
task: A task object from where to get the fields.
Returns:
A tuple with the task basic information and its capabilities
"""
task_data = TaskData(
comm=utility.array_to_string(task.comm),
pid=int(task.pid),
tgid=int(task.tgid),
ppid=int(task.parent.pid),
euid=int(task.cred.euid),
)
task_cred = task.real_cred
capabilities_data = CapabilitiesData(
cap_inheritable=task_cred.cap_inheritable,
cap_permitted=task_cred.cap_permitted,
cap_effective=task_cred.cap_effective,
cap_bset=task_cred.cap_bset,
cap_ambient=renderers.NotAvailableValue(),
)
# Ambient capabilities were added in kernels 4.3.6
if task_cred.has_member("cap_ambient"):
capabilities_data.cap_ambient = task_cred.cap_ambient
return task_data, capabilities_data
@classmethod
def get_tasks_capabilities(
cls, tasks: List[interfaces.objects.ObjectInterface]
) -> Iterable[Tuple[TaskData, CapabilitiesData]]:
"""Yields a tuple for each task containing the task's basic information along with its capabilities
Args:
tasks: An iterable with the tasks to process.
Yields:
A tuple for each task containing the task's basic information and its capabilities
"""
for task in tasks:
yield cls.get_task_capabilities(task)
def _generator(
self, tasks: Iterable[interfaces.objects.ObjectInterface]
) -> Iterable[Tuple[int, Tuple]]:
for task_fields, capabilities_fields in self.get_tasks_capabilities(tasks):
task_fields = astuple(task_fields)
capabilities_text = tuple(
self._decode_cap(cap) for cap in capabilities_fields.astuple()
)
yield 0, task_fields + capabilities_text
def run(self):
self._check_capabilities_support(self.context, self.config["kernel"])
pids = self.config.get("pids")
pid_filter = pslist.PsList.create_pid_filter(pids)
tasks = pslist.PsList.list_tasks(
self.context, self.config["kernel"], filter_func=pid_filter
)
columns = [
("Name", str),
("Tid", int),
("Pid", int),
("PPid", int),
("EUID", int),
("cap_inheritable", str),
("cap_permitted", str),
("cap_effective", str),
("cap_bounding", str),
("cap_ambient", str),
]
return renderers.TreeGrid(columns, self._generator(tasks))
+1 -1
View File
@@ -48,7 +48,7 @@ class Elfs(plugins.PluginInterface):
name = utility.array_to_string(task.comm)
for vma in task.mm.get_mmap_iter():
for vma in task.mm.get_vma_iter():
hdr = proc_layer.read(vma.vm_start, 4, pad=True)
if not (
hdr[0] == 0x7F
+3 -3
View File
@@ -42,7 +42,7 @@ class IOMem(interfaces.plugins.PluginInterface):
Args:
context: The context to retrieve required elements (layers, symbol tables) from
vmlinux_module_name: The name of the kernel module on which to operate
resource_offset: The offset to the resouce to be parsed
resource_offset: The offset to the resource to be parsed
seen: The set of resource offsets that have already been parsed
depth: How deep into the resource structure we are
@@ -57,7 +57,7 @@ class IOMem(interfaces.plugins.PluginInterface):
except exceptions.InvalidAddressException:
vollog.warning(
f"Unable to create resource object at {resource_offset:#x}. This resource, "
"its sibling, and any of it's childern and will be missing from the output."
"its sibling, and any of it's children and will be missing from the output."
)
return None
@@ -66,7 +66,7 @@ class IOMem(interfaces.plugins.PluginInterface):
name = utility.pointer_to_string(resource.name, 128)
except exceptions.InvalidAddressException:
vollog.warning(
"Unable to follow pointer to name for resource object at {resource_offset:#x}, "
f"Unable to follow pointer to name for resource object at {resource_offset:#x}, "
"replaced with UnreadableValue"
)
name = renderers.UnreadableValue()
+12 -3
View File
@@ -3,7 +3,7 @@
#
from typing import List
import logging
from volatility3.framework import constants, interfaces
from volatility3.framework import renderers
from volatility3.framework.configuration import requirements
@@ -11,6 +11,8 @@ from volatility3.framework.objects import utility
from volatility3.framework.renderers import format_hints
from volatility3.plugins.linux import pslist
vollog = logging.getLogger(__name__)
class Malfind(interfaces.plugins.PluginInterface):
"""Lists process memory ranges that potentially contain injected code."""
@@ -46,8 +48,15 @@ class Malfind(interfaces.plugins.PluginInterface):
proc_layer = self.context.layers[proc_layer_name]
for vma in task.mm.get_mmap_iter():
if vma.is_suspicious() and vma.get_name(self.context, task) != "[vdso]":
for vma in task.mm.get_vma_iter():
vma_name = vma.get_name(self.context, task)
vollog.debug(
f"Injections : processing PID {task.pid} : VMA {vma_name} : {hex(vma.vm_start)}-{hex(vma.vm_end)}"
)
if (
vma.is_suspicious(proc_layer)
and vma.get_name(self.context, task) != "[vdso]"
):
data = proc_layer.read(vma.vm_start, 64, pad=True)
yield vma, data
@@ -9,8 +9,10 @@ from typing import Tuple, List, Iterable, Union
from volatility3.framework import renderers, interfaces
from volatility3.framework.configuration import requirements
from volatility3.framework.interfaces import plugins
from volatility3.framework.symbols import linux
from volatility3.plugins.linux import pslist
vollog = logging.getLogger(__name__)
MountInfoData = namedtuple(
@@ -48,6 +50,9 @@ class MountInfo(plugins.PluginInterface):
requirements.PluginRequirement(
name="pslist", plugin=pslist.PsList, version=(2, 0, 0)
),
requirements.VersionRequirement(
name="linuxutils", component=linux.LinuxUtilities, version=(2, 1, 0)
),
requirements.ListRequirement(
name="pids",
description="Filter on specific process IDs.",
@@ -71,37 +76,6 @@ class MountInfo(plugins.PluginInterface):
),
]
@classmethod
def _do_get_path(cls, mnt, fs_root) -> Union[None, str]:
"""It mimics the Linux kernel prepend_path function."""
vfsmnt = mnt.mnt
dentry = vfsmnt.get_mnt_root()
path_reversed = []
while dentry != fs_root.dentry or vfsmnt.vol.offset != fs_root.mnt:
if dentry == vfsmnt.get_mnt_root() or dentry.is_root():
parent = mnt.get_mnt_parent().dereference()
# Escaped?
if dentry != vfsmnt.get_mnt_root():
return None
# Global root?
if mnt.vol.offset != parent.vol.offset:
dentry = mnt.get_mnt_mountpoint()
mnt = parent
vfsmnt = mnt.mnt
continue
return None
parent = dentry.d_parent
dname = dentry.d_name.name_as_str()
path_reversed.append(dname.strip("/"))
dentry = parent
path = "/" + "/".join(reversed(path_reversed))
return path
@classmethod
def get_mountinfo(
cls, mnt, task
@@ -115,8 +89,8 @@ class MountInfo(plugins.PluginInterface):
if not mnt_root:
return None
path_root = cls._do_get_path(mnt, task.fs.root)
if path_root is None:
path_root = linux.LinuxUtilities.get_path_mnt(task, mnt)
if not path_root:
return None
mnt_root_path = mnt_root.path()
@@ -170,9 +144,11 @@ class MountInfo(plugins.PluginInterface):
)
def _get_tasks_mountpoints(
self, tasks: Iterable[interfaces.objects.ObjectInterface], per_namespace: bool
self,
tasks: Iterable[interfaces.objects.ObjectInterface],
filtered_by_pids: bool,
):
seen_namespaces = set()
seen_mountpoints = set()
for task in tasks:
if not (
task
@@ -181,19 +157,27 @@ class MountInfo(plugins.PluginInterface):
and task.nsproxy
and task.nsproxy.mnt_ns
):
# This task doesn't have all the information required
# This task doesn't have all the information required.
# It should be a kernel < 2.6.30
continue
mnt_namespace = task.nsproxy.mnt_ns
mnt_ns_id = mnt_namespace.get_inode()
if per_namespace:
if mnt_ns_id in seen_namespaces:
continue
else:
seen_namespaces.add(mnt_ns_id)
try:
mnt_ns_id = mnt_namespace.get_inode()
except AttributeError:
mnt_ns_id = renderers.NotAvailableValue()
for mount in mnt_namespace.get_mount_points():
# When PIDs are filtered, it makes sense that the user want to
# see each of those processes mount points. So we don't filter
# by mount id in this case.
if not filtered_by_pids:
mnt_id = int(mount.mnt_id)
if mnt_id in seen_mountpoints:
continue
else:
seen_mountpoints.add(mnt_id)
yield task, mount, mnt_ns_id
def _generator(
@@ -201,10 +185,20 @@ class MountInfo(plugins.PluginInterface):
tasks: Iterable[interfaces.objects.ObjectInterface],
mnt_ns_ids: List[int],
mount_format: bool,
per_namespace: bool,
filtered_by_pids: bool,
) -> Iterable[Tuple[int, Tuple]]:
for task, mnt, mnt_ns_id in self._get_tasks_mountpoints(tasks, per_namespace):
if mnt_ns_ids and mnt_ns_id not in mnt_ns_ids:
show_filter_warning = False
for task, mnt, mnt_ns_id in self._get_tasks_mountpoints(
tasks, filtered_by_pids
):
if mnt_ns_ids and isinstance(mnt_ns_id, renderers.NotAvailableValue):
show_filter_warning = True
if (
not isinstance(mnt_ns_id, renderers.NotAvailableValue)
and mnt_ns_ids
and mnt_ns_id not in mnt_ns_ids
):
continue
mnt_info = self.get_mountinfo(mnt, task)
@@ -242,12 +236,17 @@ class MountInfo(plugins.PluginInterface):
]
fields_values = [mnt_ns_id]
if not per_namespace:
if filtered_by_pids:
fields_values.append(task.pid)
fields_values.extend(extra_fields_values)
yield (0, fields_values)
if show_filter_warning:
vollog.warning(
"Could not filter by mount namespace id. This field is not available in this kernel."
)
def run(self):
pids = self.config.get("pids")
mount_ns_ids = self.config.get("mntns")
@@ -263,9 +262,9 @@ class MountInfo(plugins.PluginInterface):
# to displays the mountpoints per namespace.
if pids:
columns.append(("PID", int))
per_namespace = False
filtered_by_pids = True
else:
per_namespace = True
filtered_by_pids = False
if self.config.get("mount-format"):
extra_columns = [
@@ -292,5 +291,6 @@ class MountInfo(plugins.PluginInterface):
columns.extend(extra_columns)
return renderers.TreeGrid(
columns, self._generator(tasks, mount_ns_ids, mount_format, per_namespace)
columns,
self._generator(tasks, mount_ns_ids, mount_format, filtered_by_pids),
)
+171 -4
View File
@@ -4,18 +4,25 @@
"""A module containing a collection of plugins that produce data typically
found in Linux's /proc file system."""
from volatility3.framework import renderers
import logging
from typing import Callable, Generator, Type, Optional
from volatility3.framework import renderers, interfaces, exceptions
from volatility3.framework.configuration import requirements
from volatility3.framework.interfaces import plugins
from volatility3.framework.objects import utility
from volatility3.framework.renderers import format_hints
from volatility3.plugins.linux import pslist
vollog = logging.getLogger(__name__)
class Maps(plugins.PluginInterface):
"""Lists all memory maps for all processes."""
_required_framework_version = (2, 0, 0)
_version = (1, 0, 0)
MAXSIZE_DEFAULT = 1024 * 1024 * 1024 # 1 Gb
@classmethod
def get_requirements(cls):
@@ -35,16 +42,149 @@ class Maps(plugins.PluginInterface):
element_type=int,
optional=True,
),
requirements.BooleanRequirement(
name="dump",
description="Extract listed memory segments",
default=False,
optional=True,
),
requirements.ListRequirement(
name="address",
description="Process virtual memory addresses to include "
"(all other VMA sections are excluded). This can be any "
"virtual address within the VMA section.",
element_type=int,
optional=True,
),
requirements.IntRequirement(
name="maxsize",
description="Maximum size for dumped VMA sections "
"(all the bigger sections will be ignored)",
default=cls.MAXSIZE_DEFAULT,
optional=True,
),
]
@classmethod
def list_vmas(
cls,
task: interfaces.objects.ObjectInterface,
filter_func: Callable[
[interfaces.objects.ObjectInterface], bool
] = lambda _: True,
) -> Generator[interfaces.objects.ObjectInterface, None, None]:
"""Lists the Virtual Memory Areas of a specific process.
Args:
task: task object from which to list the vma
filter_func: Function to take a vma and return False if it should be filtered out
Returns:
Yields vmas based on the task and filtered based on the filter function
"""
if task.mm:
for vma in task.mm.get_vma_iter():
if filter_func(vma):
yield vma
else:
vollog.debug(
f"Excluded vma at offset {vma.vol.offset:#x} for pid {task.pid} due to filter_func"
)
else:
vollog.debug(
f"Excluded pid {task.pid} as there is no mm member. It is likely a kernel thread."
)
@classmethod
def vma_dump(
cls,
context: interfaces.context.ContextInterface,
task: interfaces.objects.ObjectInterface,
vm_start: int,
vm_end: int,
open_method: Type[interfaces.plugins.FileHandlerInterface],
maxsize: int = MAXSIZE_DEFAULT,
) -> Optional[interfaces.plugins.FileHandlerInterface]:
"""Extracts the complete data for VMA as a FileInterface.
Args:
context: The context to retrieve required elements (layers, symbol tables) from
task: an task_struct instance
vm_start: The start virtual address from the vma to dump
vm_end: The end virtual address from the vma to dump
open_method: class to provide context manager for opening the file
maxsize: Max size of VMA section (default MAXSIZE_DEFAULT)
Returns:
An open FileInterface object containing the complete data for the task or None in the case of failure
"""
pid = task.pid
try:
proc_layer_name = task.add_process_layer()
except exceptions.InvalidAddressException as excp:
vollog.debug(
"Process {}: invalid address {} in layer {}".format(
pid, excp.invalid_address, excp.layer_name
)
)
return None
vm_size = vm_end - vm_start
# check if vm_size is negative, this should never happen.
if vm_size < 0:
vollog.warning(
f"Skip virtual memory dump for pid {pid} between {vm_start:#x}-{vm_end:#x} as {vm_size} is negative."
)
return None
# check if vm_size is larger than the maxsize limit, and therefore is not saved out.
if maxsize <= vm_size:
vollog.warning(
f"Skip virtual memory dump for pid {pid} between {vm_start:#x}-{vm_end:#x} as {vm_size} is larger than maxsize limit of {maxsize}"
)
return None
proc_layer = context.layers[proc_layer_name]
file_name = f"pid.{pid}.vma.{vm_start:#x}-{vm_end:#x}.dmp"
try:
file_handle = open_method(file_name)
chunk_size = 1024 * 1024 * 10
offset = vm_start
while offset < vm_start + vm_size:
to_read = min(chunk_size, vm_start + vm_size - offset)
data = proc_layer.read(offset, to_read, pad=True)
file_handle.write(data)
offset += to_read
except Exception as excp:
vollog.debug(f"Unable to dump virtual memory {file_name}: {excp}")
return None
return file_handle
def _generator(self, tasks):
# build filter for addresses if required
address_list = self.config.get("address", None)
if not address_list:
# do not filter as no address_list was supplied
vma_filter_func = lambda _: True
else:
# filter for any vm_start that matches the supplied address config
def vma_filter_function(x: interfaces.objects.ObjectInterface) -> bool:
addrs_in_vma = [
addr for addr in address_list if x.vm_start <= addr <= x.vm_end
]
# if any of the user supplied addresses would fall within this vma return true
if addrs_in_vma:
return True
else:
return False
vma_filter_func = vma_filter_function
for task in tasks:
if not task.mm:
continue
name = utility.array_to_string(task.comm)
for vma in task.mm.get_mmap_iter():
for vma in self.list_vmas(task, filter_func=vma_filter_func):
flags = vma.get_protection()
page_offset = vma.get_page_offset()
major = 0
@@ -58,9 +198,34 @@ class Maps(plugins.PluginInterface):
major = inode_object.i_sb.major
minor = inode_object.i_sb.minor
inode = inode_object.i_ino
path = vma.get_name(self.context, task)
file_output = "Disabled"
if self.config["dump"]:
file_output = "Error outputting file"
try:
vm_start = vma.vm_start
vm_end = vma.vm_end
except AttributeError:
vollog.debug(
f"Unable to find the vm_start and vm_end for vma at {vma.vol.offset:#x} for pid {task.pid}"
)
vm_start = None
vm_end = None
if vm_start and vm_end:
# only attempt to dump the memory if we have vm_start and vm_end
file_handle = self.vma_dump(
self.context,
task,
vm_start,
vm_end,
self.open,
self.config["maxsize"],
)
if file_handle:
file_handle.close()
file_output = file_handle.preferred_filename
yield (
0,
(
@@ -74,6 +239,7 @@ class Maps(plugins.PluginInterface):
minor,
inode,
path,
file_output,
),
)
@@ -92,6 +258,7 @@ class Maps(plugins.PluginInterface):
("Minor", int),
("Inode", int),
("File Path", str),
("File output", str),
],
self._generator(
pslist.PsList.list_tasks(
@@ -52,7 +52,10 @@ class PsScan(interfaces.plugins.PluginInterface):
"""
pid = task.tgid
tid = task.pid
ppid = task.parent.tgid if task.parent else 0
ppid = 0
if task.parent.is_readable():
ppid = task.parent.tgid
name = utility.array_to_string(task.comm)
exit_state = DescExitStateEnum(task.exit_state).name
+44 -13
View File
@@ -28,7 +28,11 @@ class SockHandlers(interfaces.configuration.VersionableInterface):
self._vmlinux = vmlinux
self._task = task
netns_id = task.nsproxy.net_ns.get_inode()
try:
netns_id = task.nsproxy.net_ns.get_inode()
except AttributeError:
netns_id = NotAvailableValue()
self._netdevices = self._build_network_devices_map(netns_id)
self._sock_family_handlers = {
@@ -61,7 +65,10 @@ class SockHandlers(interfaces.configuration.VersionableInterface):
self._vmlinux.symbol_table_name + constants.BANG + "net_device"
)
for net_dev in net.dev_base_head.to_list(net_device_symname, "dev_list"):
if net.get_inode() != netns_id:
if (
isinstance(netns_id, NotAvailableValue)
or net.get_inode() != netns_id
):
continue
dev_name = utility.array_to_string(net_dev.name)
netdevices_map[net_dev.ifindex] = dev_name
@@ -76,7 +83,7 @@ class SockHandlers(interfaces.configuration.VersionableInterface):
sock: Kernel generic `sock` object
Returns a tuple with:
sock: The respective kernel's *_sock object for that socket family
sock: The respective kernel's \*_sock object for that socket family
sock_stat: A tuple with the source and destination (address and port) along with its state string
socket_filter: A dictionary with information about the socket filter
"""
@@ -143,19 +150,32 @@ class SockHandlers(interfaces.configuration.VersionableInterface):
return
bpfprog = sock_filter.prog
if bpfprog.type == 0:
# BPF_PROG_TYPE_UNSPEC = 0
BPF_PROG_TYPE_UNSPEC = 0 # cBPF filter
try:
bpfprog_type = bpfprog.get_type()
if bpfprog_type == BPF_PROG_TYPE_UNSPEC:
return # cBPF filter
except AttributeError:
# kernel < 3.18.140, it's a cBPF filter
return
BPF_PROG_TYPE_SOCKET_FILTER = 1 # eBPF filter
if bpfprog_type != BPF_PROG_TYPE_SOCKET_FILTER:
socket_filter["bpf_filter_type"] = f"UNK({bpfprog_type})"
vollog.warning(f"Unexpected BPF type {bpfprog_type} for a socket")
return
socket_filter["bpf_filter_type"] = "eBPF"
if not bpfprog.has_member("aux") or not bpfprog.aux:
return
return # kernel < 3.18.140
bpfprog_aux = bpfprog.aux
if bpfprog_aux.has_member("id"):
# `id` member was added to `bpf_prog_aux` in kernels 4.13
# `id` member was added to `bpf_prog_aux` in kernels 4.13.16
socket_filter["bpf_filter_id"] = str(bpfprog_aux.id)
if bpfprog_aux.has_member("name"):
# `name` was added to `bpf_prog_aux` in kernels 4.15
# `name` was added to `bpf_prog_aux` in kernels 4.15.18
bpfprog_name = utility.array_to_string(bpfprog_aux.name)
if bpfprog_name:
socket_filter["bpf_filter_name"] = bpfprog_name
@@ -227,14 +247,22 @@ class SockHandlers(interfaces.configuration.VersionableInterface):
if netlink_sock.groups:
groups_bitmap = netlink_sock.groups.dereference()
src_addr = f"groups:0x{groups_bitmap:08x}"
src_port = netlink_sock.portid
try:
# Kernel >= 3.7.10
src_port = netlink_sock.get_portid()
except AttributeError:
src_port = NotAvailableValue()
dst_addr = f"group:0x{netlink_sock.dst_group:08x}"
module = netlink_sock.module
if module and module.name:
module_name_str = utility.array_to_string(module.name)
dst_addr = f"{dst_addr},lkm:{module_name_str}"
dst_port = netlink_sock.dst_portid
try:
dst_port = netlink_sock.get_dst_portid()
except AttributeError:
dst_port = NotAvailableValue()
state = netlink_sock.get_state()
@@ -473,8 +501,7 @@ class Sockstat(plugins.PluginInterface):
family: Socket family string (AF_UNIX, AF_INET, etc)
sock_type: Socket type string (STREAM, DGRAM, etc)
protocol: Protocol string (UDP, TCP, etc)
sock_fields: A tuple with the *_sock object, the sock stats and the
extended info dictionary
sock_fields: A tuple with the \*_sock object, the sock stats and the extended info dictionary
"""
vmlinux = context.modules[symbol_table]
@@ -518,7 +545,11 @@ class Sockstat(plugins.PluginInterface):
protocol = child_sock.get_protocol()
net = task.nsproxy.net_ns
netns_id = net.get_inode()
try:
netns_id = net.get_inode()
except AttributeError:
netns_id = NotAvailableValue()
yield task, netns_id, fd_num, family, sock_type, protocol, sock_fields
def _format_fields(self, sock_stat, protocol):
+27 -6
View File
@@ -2,12 +2,14 @@
# which is available at https://www.volatilityfoundation.org/license/vsl-v1.0
#
import datetime
import logging
from typing import Callable, Iterable, List, Dict
from typing import Callable, Dict, Iterable, List
from volatility3.framework import renderers, interfaces, exceptions
from volatility3.framework import exceptions, interfaces, renderers
from volatility3.framework.configuration import requirements
from volatility3.framework.objects import utility
from volatility3.framework.renderers import format_hints
from volatility3.framework.symbols import mac
vollog = logging.getLogger(__name__)
@@ -105,10 +107,20 @@ class PsList(interfaces.plugins.PluginInterface):
self.config["kernel"],
filter_func=self.create_pid_filter(self.config.get("pid", None)),
):
pid = task.p_pid
ppid = task.p_ppid
offset = format_hints.Hex(task.vol.offset)
name = utility.array_to_string(task.p_comm)
yield (0, (pid, ppid, name))
pid = task.p_pid
uid = task.p_uid
gid = task.p_gid
start_time_seconds = task.p_start.tv_sec
start_time_microseconds = task.p_start.tv_usec
start_time = datetime.datetime.fromtimestamp(
start_time_seconds + start_time_microseconds / 1e6
)
ppid = task.p_ppid
yield (0, (offset, name, pid, uid, gid, start_time, ppid))
@classmethod
def list_tasks_allproc(
@@ -310,5 +322,14 @@ class PsList(interfaces.plugins.PluginInterface):
def run(self):
return renderers.TreeGrid(
[("PID", int), ("PPID", int), ("COMM", str)], self._generator()
[
("OFFSET", format_hints.Hex),
("NAME", str),
("PID", int),
("UID", int),
("GID", int),
("Start Time", datetime.datetime),
("PPID", int),
],
self._generator(),
)
@@ -14,6 +14,8 @@ vollog = logging.getLogger(__name__)
class Crashinfo(interfaces.plugins.PluginInterface):
"""Lists the information from a Windows crash dump."""
_required_framework_version = (2, 0, 0)
@classmethod
@@ -7,6 +7,8 @@ from volatility3.plugins.windows import pslist, vadinfo
class LdrModules(interfaces.plugins.PluginInterface):
"""Lists the loaded modules in a particular windows memory image."""
_required_framework_version = (2, 0, 0)
_version = (1, 0, 0)
@@ -42,6 +42,12 @@ class Modules(interfaces.plugins.PluginInterface):
default=False,
optional=True,
),
requirements.StringRequirement(
name="name",
description="module name/sub string",
optional=True,
default=None,
),
]
def _generator(self):
@@ -63,6 +69,9 @@ class Modules(interfaces.plugins.PluginInterface):
except exceptions.InvalidAddressException:
FullDllName = ""
if self.config["name"] and self.config["name"] not in BaseDllName:
continue
file_output = "Disabled"
if self.config["dump"]:
file_handle = dlllist.DllList.dump_pe(
@@ -149,9 +149,9 @@ class Strings(interfaces.plugins.PluginInterface):
for mapval in layer.mapping(0x0, layer.maximum_address, ignore_errors=True):
offset, _, mapped_offset, mapped_size, maplayer = mapval
for val in range(mapped_offset, mapped_offset + mapped_size, 0x1000):
cur_set = reverse_map.get(mapped_offset >> 12, set())
cur_set = reverse_map.get(val >> 12, set())
cur_set.add(("kernel", offset))
reverse_map[mapped_offset >> 12] = cur_set
reverse_map[val >> 12] = cur_set
if progress_callback:
progress_callback(
(offset * 100) / layer.maximum_address,
@@ -198,6 +198,7 @@ class VadInfo(interfaces.plugins.PluginInterface):
def _generator(self, procs):
kernel = self.context.modules[self.config["kernel"]]
kernel_layer = self.context.layers[kernel.layer_name]
def passthrough(_: interfaces.objects.ObjectInterface) -> bool:
return False
@@ -229,7 +230,7 @@ class VadInfo(interfaces.plugins.PluginInterface):
(
proc.UniqueProcessId,
process_name,
format_hints.Hex(vad.vol.offset),
format_hints.Hex(kernel_layer.canonicalize(vad.vol.offset)),
format_hints.Hex(vad.get_start()),
format_hints.Hex(vad.get_end()),
vad.get_tag(),
+16 -2
View File
@@ -34,13 +34,27 @@ class YaraScanner(interfaces.layers.ScannerInterface):
if rules is None:
raise ValueError("No rules provided to YaraScanner")
self._rules = rules
self.st_object = not tuple([int(x) for x in yara.__version__.split(".")]) < (
4,
3,
)
def __call__(
self, data: bytes, data_offset: int
) -> Iterable[Tuple[int, str, str, bytes]]:
for match in self._rules.match(data=data):
for offset, name, value in match.strings:
yield (offset + data_offset, match.rule, name, value)
if self.st_object:
for match_string in match.strings:
for instance in match_string.instances:
yield (
instance.offset + data_offset,
match.rule,
match_string.identifier,
instance.matched_data,
)
else:
for offset, name, value in match.strings:
yield (offset + data_offset, match.rule, name, value)
class YaraScan(plugins.PluginInterface):
+134 -64
View File
@@ -1,7 +1,7 @@
# This file is Copyright 2019 Volatility Foundation and licensed under the Volatility Software License 1.0
# which is available at https://www.volatilityfoundation.org/license/vsl-v1.0
#
from typing import Iterator, List, Tuple, Optional
from typing import Iterator, List, Tuple, Optional, Union
from volatility3 import framework
from volatility3.framework import constants, exceptions, interfaces, objects
@@ -28,8 +28,11 @@ class LinuxKernelIntermedSymbols(intermed.IntermediateSymbolTable):
self.set_type_class("fs_struct", extensions.fs_struct)
self.set_type_class("files_struct", extensions.files_struct)
self.set_type_class("kobject", extensions.kobject)
self.set_type_class("cred", extensions.cred)
self.set_type_class("kernel_cap_struct", extensions.kernel_cap_struct)
# Might not exist in the current symbols
self.optional_set_type_class("module", extensions.module)
self.optional_set_type_class("bpf_prog", extensions.bpf_prog)
# Mount
self.set_type_class("vfsmount", extensions.vfsmount)
@@ -50,92 +53,113 @@ class LinuxKernelIntermedSymbols(intermed.IntermediateSymbolTable):
self.optional_set_type_class("bt_sock", extensions.bt_sock)
self.optional_set_type_class("xdp_sock", extensions.xdp_sock)
# Only found in 6.1+ kernels
self.optional_set_type_class("maple_tree", extensions.maple_tree)
class LinuxUtilities(interfaces.configuration.VersionableInterface):
"""Class with multiple useful linux functions."""
_version = (2, 0, 0)
_version = (2, 1, 0)
_required_framework_version = (2, 0, 0)
framework.require_interface_version(*_required_framework_version)
# based on __d_path from the Linux kernel
@classmethod
def _do_get_path(cls, rdentry, rmnt, dentry, vfsmnt) -> str:
ret_path: List[str] = []
def _get_path_file(cls, task, filp) -> str:
"""Returns the file pathname relative to the task's root directory.
while dentry != rdentry or vfsmnt != rmnt:
dname = dentry.path()
if dname == "":
break
Args:
task (task_struct): A reference task
filp (file *): A pointer to an open file
ret_path.insert(0, dname.strip("/"))
if dentry == vfsmnt.get_mnt_root() or dentry == dentry.d_parent:
if vfsmnt.get_mnt_parent() == vfsmnt:
Returns:
str: File pathname relative to the task's root directory.
"""
rdentry = task.fs.get_root_dentry()
rmnt = task.fs.get_root_mnt()
vfsmnt = filp.get_vfsmnt()
dentry = filp.get_dentry()
return cls.do_get_path(rdentry, rmnt, dentry, vfsmnt)
@classmethod
def get_path_mnt(cls, task, mnt) -> str:
"""Returns the mount point pathname relative to the task's root directory.
Args:
task (task_struct): A reference task
mnt (vfsmount or mount): A mounted filesystem or a mount point.
- kernels < 3.3.8 type is 'vfsmount'
- kernels >= 3.3.8 type is 'mount'
Returns:
str: Pathname of the mount point relative to the task's root directory.
"""
rdentry = task.fs.get_root_dentry()
rmnt = task.fs.get_root_mnt()
vfsmnt = mnt.get_vfsmnt_current()
dentry = mnt.get_dentry_current()
return cls.do_get_path(rdentry, rmnt, dentry, vfsmnt)
@classmethod
def do_get_path(cls, rdentry, rmnt, dentry, vfsmnt) -> Union[None, str]:
"""Returns a pathname of the mount point or file
It mimics the Linux kernel prepend_path function.
Args:
rdentry (dentry *): A pointer to the root dentry
rmnt (vfsmount *): A pointer to the root vfsmount
dentry (dentry *): A pointer to the dentry
vfsmnt (vfsmount *): A pointer to the vfsmount
Returns:
str: Pathname of the mount point or file
"""
path_reversed = []
while dentry != rdentry or not vfsmnt.is_equal(rmnt):
if dentry == vfsmnt.get_mnt_root() or dentry.is_root():
# Escaped?
if dentry != vfsmnt.get_mnt_root():
break
dentry = vfsmnt.get_mnt_mountpoint()
vfsmnt = vfsmnt.get_mnt_parent()
# Global root?
if not vfsmnt.has_parent():
break
dentry = vfsmnt.get_dentry_parent()
vfsmnt = vfsmnt.get_vfsmnt_parent()
continue
parent = dentry.d_parent
dname = dentry.d_name.name_as_str()
path_reversed.append(dname.strip("/"))
dentry = parent
# if we did not gather any valid dentrys in the path, then the entire file is
# either 1) smeared out of memory or 2) de-allocated and corresponding structures overwritten
# we return an empty string in this case to avoid confusion with something like a handle to the root
# directory (e.g., "/")
if not ret_path:
return ""
ret_val = "/".join([str(p) for p in ret_path if p != ""])
if ret_val.startswith(("socket:", "pipe:")):
if ret_val.find("]") == -1:
try:
inode = dentry.d_inode
ino = inode.i_ino
except exceptions.InvalidAddressException:
ino = 0
ret_val = ret_val[:-1] + f":[{ino}]"
else:
ret_val = ret_val.replace("/", "")
elif ret_val != "inotify":
ret_val = "/" + ret_val
return ret_val
# method used by 'older' kernels
# TODO: lookup when dentry_operations->d_name was merged into the mainline kernel for exact version
@classmethod
def _get_path_file(cls, task, filp) -> str:
rdentry = task.fs.get_root_dentry()
rmnt = task.fs.get_root_mnt()
dentry = filp.get_dentry()
vfsmnt = filp.get_vfsmnt()
return LinuxUtilities._do_get_path(rdentry, rmnt, dentry, vfsmnt)
path = "/" + "/".join(reversed(path_reversed))
return path
@classmethod
def _get_new_sock_pipe_path(cls, context, task, filp) -> str:
"""Returns the sock pipe pathname relative to the task's root directory.
Args:
context: The context to retrieve required elements (layers, symbol tables) from
task (task_struct): A reference task
filp (file *): A pointer to a sock pipe open file
Returns:
str: Sock pipe pathname relative to the task's root directory.
"""
dentry = filp.get_dentry()
kernel_module = cls.get_module_from_volobj_type(context, dentry)
sym_addr = dentry.d_op.d_dname
symbol_table_arr = sym_addr.vol.type_name.split("!")
symbol_table = None
if len(symbol_table_arr) == 2:
symbol_table = symbol_table_arr[0]
for module_name in context.modules.get_modules_by_symbol_tables(symbol_table):
kernel_module = context.modules[module_name]
break
else:
raise ValueError(f"No module using the symbol table {symbol_table}")
symbs = list(kernel_module.get_symbols_by_absolute_location(sym_addr))
if len(symbs) == 1:
@@ -163,12 +187,26 @@ class LinuxUtilities(interfaces.configuration.VersionableInterface):
return ret
# a 'file' structure doesn't have enough information to properly restore its full path
# we need the root mount information from task_struct to determine this
@classmethod
def path_for_file(cls, context, task, filp) -> str:
"""Returns a file (or sock pipe) pathname relative to the task's root directory.
A 'file' structure doesn't have enough information to properly restore its
full path we need the root mount information from task_struct to determine this
Args:
context: The context to retrieve required elements (layers, symbol tables) from
task (task_struct): A reference task
filp (file *): A pointer to an open file
Returns:
str: A file (or sock pipe) pathname relative to the task's root directory.
"""
# Memory smear protection: Check that both the file and dentry pointers are valid.
try:
dentry = filp.get_dentry()
dentry.is_root()
except exceptions.InvalidAddressException:
return ""
@@ -348,3 +386,35 @@ class LinuxUtilities(interfaces.configuration.VersionableInterface):
return vmlinux.object(
object_type=type_name, offset=container_addr, absolute=True
)
@classmethod
def get_module_from_volobj_type(
cls,
context: interfaces.context.ContextInterface,
volobj: interfaces.objects.ObjectInterface,
) -> interfaces.context.ModuleInterface:
"""Get the vmlinux from a vol obj
Args:
context: The context to retrieve required elements (layers, symbol tables) from
volobj (vol object): A vol object
Raises:
ValueError: If it cannot obtain any module from the symbol table
Returns:
A kernel object (vmlinux)
"""
symbol_table_arr = volobj.vol.type_name.split("!", 1)
symbol_table = symbol_table_arr[0] if len(symbol_table_arr) == 2 else None
module_names = context.modules.get_modules_by_symbol_tables(symbol_table)
module_names = list(module_names)
if not module_names:
raise ValueError(f"No module using the symbol table '{symbol_table}'")
kernel_module_name = module_names[0]
kernel = context.modules[kernel_module_name]
return kernel
@@ -5,7 +5,7 @@
import collections.abc
import logging
import socket as socket_module
from typing import Generator, Iterable, Iterator, Optional, Tuple
from typing import Generator, Iterable, Iterator, Optional, Tuple, List
from volatility3.framework import constants
from volatility3.framework.constants.linux import SOCK_TYPES, SOCK_FAMILY
@@ -13,6 +13,7 @@ from volatility3.framework.constants.linux import IP_PROTOCOLS, IPV6_PROTOCOLS
from volatility3.framework.constants.linux import TCP_STATES, NETLINK_PROTOCOLS
from volatility3.framework.constants.linux import ETH_PROTOCOLS, BLUETOOTH_STATES
from volatility3.framework.constants.linux import BLUETOOTH_PROTOCOLS, SOCKET_STATES
from volatility3.framework.constants.linux import CAPABILITIES
from volatility3.framework import exceptions, objects, interfaces, symbols
from volatility3.framework.layers import linear
from volatility3.framework.objects import utility
@@ -34,10 +35,8 @@ class module(generic.GenericIntelProcess):
def get_init_size(self):
if self.has_member("init_layout"):
return self.init_layout.size
elif self.has_member("init_size"):
return self.init_size
raise AttributeError(
"module -> get_init_size: Unable to determine .init section size of module"
)
@@ -45,10 +44,8 @@ class module(generic.GenericIntelProcess):
def get_core_size(self):
if self.has_member("core_layout"):
return self.core_layout.size
elif self.has_member("core_size"):
return self.core_size
raise AttributeError(
"module -> get_core_size: Unable to determine core size of module"
)
@@ -58,7 +55,6 @@ class module(generic.GenericIntelProcess):
return self.core_layout.base
elif self.has_member("module_core"):
return self.module_core
raise AttributeError("module -> get_module_core: Unable to get module core")
def get_module_init(self):
@@ -66,7 +62,6 @@ class module(generic.GenericIntelProcess):
return self.init_layout.base
elif self.has_member("module_init"):
return self.module_init
raise AttributeError("module -> get_module_core: Unable to get module init")
def get_name(self):
@@ -88,7 +83,6 @@ class module(generic.GenericIntelProcess):
idx = 0
while arr[idx]:
idx = idx + 1
return idx
def get_sections(self):
@@ -97,7 +91,6 @@ class module(generic.GenericIntelProcess):
num_sects = self.sect_attrs.nsections
else:
num_sects = self._get_sect_count(self.sect_attrs.grp)
arr = self._context.object(
self.get_symbol_table().name + constants.BANG + "array",
layer_name=self.vol.layer_name,
@@ -116,7 +109,6 @@ class module(generic.GenericIntelProcess):
prefix = "Elf64_"
else:
prefix = "Elf32_"
elf_table_name = intermed.IntermediateSymbolTable.create(
self.context,
self.config_path,
@@ -155,7 +147,6 @@ class module(generic.GenericIntelProcess):
return self.kallsyms.symtab
elif self.has_member("symtab"):
return self.symtab
raise AttributeError("module -> symtab: Unable to get symtab")
@property
@@ -164,7 +155,6 @@ class module(generic.GenericIntelProcess):
return int(self.kallsyms.num_symtab)
elif self.has_member("num_symtab"):
return int(self.num_symtab)
raise AttributeError(
"module -> num_symtab: Unable to determine number of symbols"
)
@@ -177,7 +167,6 @@ class module(generic.GenericIntelProcess):
# Older kernels
elif self.has_member("strtab"):
return self.strtab
raise AttributeError("module -> strtab: Unable to get strtab")
@@ -195,19 +184,15 @@ class task_struct(generic.GenericIntelProcess):
pgd = self.mm.pgd
except exceptions.InvalidAddressException:
return None
if not isinstance(parent_layer, linear.LinearlyMappedLayer):
raise TypeError(
"Parent layer is not a translation layer, unable to construct process layer"
)
dtb, layer_name = parent_layer.translate(pgd)
if not dtb:
return None
if preferred_name is None:
preferred_name = self.vol.layer_name + f"_Process{self.pid}"
# Add the constructed layer and return the name
return self._add_process_layer(
self._context, dtb, config_prefix, preferred_name
@@ -229,7 +214,6 @@ class task_struct(generic.GenericIntelProcess):
vollog.info(
f"adding vma: {start:x} {self.mm.brk:x} | {end:x} {self.mm.start_brk:x}"
)
yield (start, end - start)
@property
@@ -282,7 +266,6 @@ class fs_struct(objects.StructType):
return self.root
elif self.root.has_member("dentry"):
return self.root.dentry
raise AttributeError("Unable to find the root dentry")
def get_root_mnt(self):
@@ -291,17 +274,135 @@ class fs_struct(objects.StructType):
return self.rootmnt
elif self.root.has_member("mnt"):
return self.root.mnt
raise AttributeError("Unable to find the root mount")
class maple_tree(objects.StructType):
# include/linux/maple_tree.h
# Mask for Maple Tree Flags
MT_FLAGS_HEIGHT_MASK = 0x7C
MT_FLAGS_HEIGHT_OFFSET = 0x02
# Shift and mask to extract information from maple tree node pointers
MAPLE_NODE_TYPE_SHIFT = 0x03
MAPLE_NODE_TYPE_MASK = 0x0F
MAPLE_NODE_POINTER_MASK = 0xFF
# types of Maple Tree Nodes
MAPLE_DENSE = 0
MAPLE_LEAF_64 = 1
MAPLE_RANGE_64 = 2
MAPLE_ARANGE_64 = 3
def get_slot_iter(self):
"""Parse the Maple Tree and return every non zero slot."""
maple_tree_offset = self.vol.offset & ~(self.MAPLE_NODE_POINTER_MASK)
expected_maple_tree_depth = (
self.ma_flags & self.MT_FLAGS_HEIGHT_MASK
) >> self.MT_FLAGS_HEIGHT_OFFSET
yield from self._parse_maple_tree_node(
self.ma_root, maple_tree_offset, expected_maple_tree_depth
)
def _parse_maple_tree_node(
self,
maple_tree_entry,
parent,
expected_maple_tree_depth,
seen=set(),
current_depth=1,
):
"""Recursively parse Maple Tree Nodes and yield all non empty slots"""
# protect against unlikely loop
if maple_tree_entry in seen:
vollog.warning(
f"The mte {hex(maple_tree_entry)} has all ready been seen, no further results will be produced for this node."
)
return
else:
seen.add(maple_tree_entry)
# check if we have exceeded the expected depth of this maple tree.
# e.g. when current_depth is larger than expected_maple_tree_depth there may be an issue.
# it is normal that expected_maple_tree_depth is equal to current_depth.
if expected_maple_tree_depth < current_depth:
vollog.warning(
f"The depth for the maple tree at {hex(self.vol.offset)} is {expected_maple_tree_depth}, however when parsing the nodes "
f"a depth of {current_depth} was reached. This is unexpected and may lead to incorrect results."
)
# parse the mte to extract the pointer value, node type, and leaf status
pointer = maple_tree_entry & ~(self.MAPLE_NODE_POINTER_MASK)
node_type = (
maple_tree_entry >> self.MAPLE_NODE_TYPE_SHIFT
) & self.MAPLE_NODE_TYPE_MASK
# create a pointer object for the node parent mte (note this will include flags in the low bits)
symbol_table_name = self.get_symbol_table_name()
node_parent_mte = self._context.object(
symbol_table_name + constants.BANG + "pointer",
layer_name=self.vol.native_layer_name,
offset=pointer,
)
# extract the actual pointer to the parent of this node
node_parent_pointer = node_parent_mte & ~(self.MAPLE_NODE_POINTER_MASK)
# verify that the node_parent_pointer correctly points to the parent
assert node_parent_pointer == parent
# create a node object
node = self._context.object(
symbol_table_name + constants.BANG + "maple_node",
layer_name=self.vol.layer_name,
offset=pointer,
)
# parse the slots based on the node type
if node_type == self.MAPLE_DENSE:
for slot in node.alloc.slot:
if (slot & ~(self.MAPLE_NODE_TYPE_MASK)) != 0:
yield slot
elif node_type == self.MAPLE_LEAF_64:
for slot in node.mr64.slot:
if (slot & ~(self.MAPLE_NODE_TYPE_MASK)) != 0:
yield slot
elif node_type == self.MAPLE_RANGE_64:
for slot in node.mr64.slot:
if (slot & ~(self.MAPLE_NODE_TYPE_MASK)) != 0:
yield from self._parse_maple_tree_node(
slot,
pointer,
expected_maple_tree_depth,
seen,
current_depth + 1,
)
elif node_type == self.MAPLE_ARANGE_64:
for slot in node.ma64.slot:
if (slot & ~(self.MAPLE_NODE_TYPE_MASK)) != 0:
yield from self._parse_maple_tree_node(
slot,
pointer,
expected_maple_tree_depth,
seen,
current_depth + 1,
)
else:
# unkown maple node type
raise AttributeError(
f"Unkown Maple Tree node type {node_type} at offset {hex(pointer)}."
)
class mm_struct(objects.StructType):
def get_mmap_iter(self) -> Iterable[interfaces.objects.ObjectInterface]:
"""Returns an iterator for the mmap list member of an mm_struct."""
if not self.has_member("mmap"):
raise AttributeError(
"get_mmap_iter called on mm_struct where no mmap member exists."
)
if not self.mmap:
return
yield self.mmap
seen = {self.mmap.vol.offset}
@@ -312,6 +413,33 @@ class mm_struct(objects.StructType):
seen.add(link.vol.offset)
link = link.vm_next
def get_maple_tree_iter(self) -> Iterable[interfaces.objects.ObjectInterface]:
"""Returns an iterator for the mm_mt member of an mm_struct."""
if not self.has_member("mm_mt"):
raise AttributeError(
"get_maple_tree_iter called on mm_struct where no mm_mt member exists."
)
symbol_table_name = self.get_symbol_table_name()
for vma_pointer in self.mm_mt.get_slot_iter():
# convert pointer to vm_area_struct and yield
vma = self._context.object(
symbol_table_name + constants.BANG + "vm_area_struct",
layer_name=self.vol.native_layer_name,
offset=vma_pointer,
)
yield vma
def get_vma_iter(self) -> Iterable[interfaces.objects.ObjectInterface]:
"""Returns an iterator for the VMAs in an mm_struct. Automatically choosing the mmap or mm_mt as required."""
if self.has_member("mmap"):
yield from self.get_mmap_iter()
elif self.has_member("mm_mt"):
yield from self.get_maple_tree_iter()
else:
raise AttributeError("Unable to find mmap or mm_mt in mm_struct")
class super_block(objects.StructType):
# include/linux/kdev_t.h
@@ -418,7 +546,6 @@ class vm_area_struct(objects.StructType):
retval = retval + char
else:
retval = retval + "-"
return retval
# only parse the rwx bits
@@ -432,7 +559,6 @@ class vm_area_struct(objects.StructType):
def get_page_offset(self) -> int:
if self.vm_file == 0:
return 0
return self.vm_pgoff << constants.linux.PAGE_SHIFT
def get_name(self, context, task):
@@ -449,21 +575,36 @@ class vm_area_struct(objects.StructType):
fname = "[vdso]"
else:
fname = "Anonymous Mapping"
return fname
# used by malfind
def is_suspicious(self):
def is_suspicious(self, proclayer=None):
ret = False
flags_str = self.get_protection()
if flags_str == "rwx":
ret = True
elif flags_str == "r-x" and self.vm_file.dereference().vol.offset == 0:
ret = True
elif proclayer and "x" in flags_str:
for i in range(self.vm_start, self.vm_end, 1 << constants.linux.PAGE_SHIFT):
try:
if proclayer.is_dirty(i):
vollog.warning(
f"Found malicious (dirty+exec) page at {hex(i)} !"
)
# We do not attempt to find other dirty+exec pages once we have found one
ret = True
break
except (
exceptions.PagedInvalidAddressException,
exceptions.InvalidAddressException,
) as excp:
vollog.debug(f"Unable to translate address {hex(i)} : {excp}")
# Abort as it is likely that other addresses in the same range will also fail
ret = False
break
return ret
@@ -473,12 +614,10 @@ class qstr(objects.StructType):
str_length = self.len + 1 # Maximum length should include null terminator
else:
str_length = 255
try:
ret = objects.utility.pointer_to_string(self.name, str_length)
except (exceptions.InvalidAddressException, ValueError):
ret = ""
return ret
@@ -509,7 +648,6 @@ class dentry(objects.StructType):
"""
if self.vol.offset == old_dentry:
return True
return self.d_ancestor(old_dentry)
def d_ancestor(self, ancestor_dentry):
@@ -527,10 +665,8 @@ class dentry(objects.StructType):
):
if current_dentry.d_parent == ancestor_dentry.vol.offset:
return current_dentry
dentry_seen.add(current_dentry.vol.offset)
current_dentry = current_dentry.d_parent
return None
@@ -544,6 +680,7 @@ class struct_file(objects.StructType):
raise AttributeError("Unable to find file -> dentry")
def get_vfsmnt(self) -> interfaces.objects.ObjectInterface:
"""Returns the fs (vfsmount) where this file is mounted"""
if self.has_member("f_vfsmnt"):
return self.f_vfsmnt
elif self.has_member("f_path"):
@@ -587,12 +724,10 @@ class list_head(objects.StructType, collections.abc.Iterable):
link = getattr(self, direction).dereference()
except exceptions.InvalidAddressException:
return
if not sentinel:
yield self._context.object(
symbol_type, layer, offset=self.vol.offset - relative_offset
)
seen = {self.vol.offset}
while link.vol.offset not in seen:
obj = self._context.object(
@@ -675,11 +810,70 @@ class mount(objects.StructType):
raise AttributeError("Unable to find mount -> mount flags")
def get_mnt_parent(self):
"""Gets the fs where we are mounted on
Returns:
A mount pointer
"""
return self.mnt_parent
def get_mnt_mountpoint(self):
"""Gets the dentry of the mountpoint
Returns:
A dentry pointer
"""
return self.mnt_mountpoint
def get_parent_mount(self):
return self.mnt.get_parent_mount()
def has_parent(self) -> bool:
"""Checks if this mount has a parent
Returns:
bool: 'True' if this mount has a parent
"""
return self.mnt_parent != self.vol.offset
def get_vfsmnt_current(self):
"""Returns the fs where we are mounted on
Returns:
A 'vfsmount'
"""
return self.mnt
def get_vfsmnt_parent(self):
"""Gets the parent fs (vfsmount) to where it's mounted on
Returns:
A 'vfsmount'
"""
return self.get_mnt_parent().get_vfsmnt_current()
def get_dentry_current(self):
"""Returns the root of the mounted tree
Returns:
A dentry pointer
"""
vfsmnt = self.get_vfsmnt_current()
dentry = vfsmnt.mnt_root
return dentry
def get_dentry_parent(self):
"""Returns the parent root of the mounted tree
Returns:
A dentry pointer
"""
return self.get_mnt_parent().get_dentry_current()
def get_flags_access(self) -> str:
return "ro" if self.get_mnt_flags() & self.MNT_READONLY else "rw"
@@ -703,9 +897,6 @@ class mount(objects.StructType):
def get_devname(self) -> str:
return utility.pointer_to_string(self.mnt_devname, count=255)
def has_parent(self) -> bool:
return self.vol.offset != self.mnt_parent
def get_dominating_id(self, root) -> int:
"""Get ID of closest dominating peer group having a representative under the given root."""
mnt_seen = set()
@@ -718,7 +909,6 @@ class mount(objects.StructType):
peer = current_mnt.get_peer_under_root(self.mnt_ns, root)
if peer and peer.vol.offset != 0:
return peer.mnt_group_id
mnt_seen.add(current_mnt.vol.offset)
current_mnt = current_mnt.mnt_master
return 0
@@ -734,12 +924,10 @@ class mount(objects.StructType):
current_mnt.mnt.mnt_root, root
):
return current_mnt
mnt_seen.add(current_mnt.vol.offset)
current_mnt = current_mnt.next_peer()
if current_mnt.vol.offset == self.vol.offset:
break
return None
def is_path_reachable(self, current_dentry, root):
@@ -756,7 +944,6 @@ class mount(objects.StructType):
current_dentry = current_mnt.mnt_mountpoint
mnt_seen.add(current_mnt.vol.offset)
current_mnt = current_mnt.mnt_parent
return current_mnt.mnt.vol.offset == root.mnt and current_dentry.is_subdir(
root.dentry
)
@@ -783,24 +970,121 @@ class vfsmount(objects.StructType):
and self.get_mnt_parent() != 0
)
def _get_real_mnt(self):
table_name = self.vol.type_name.split(constants.BANG)[0]
mount_struct = f"{table_name}{constants.BANG}mount"
offset = self._context.symbol_space.get_type(
mount_struct
).relative_child_offset("mnt")
def _is_kernel_prior_to_struct_mount(self) -> bool:
"""Helper to distinguish between kernels prior to version 3.3.8 that
lacked the 'mount' structure and later versions that have it.
return self._context.object(
mount_struct, self.vol.layer_name, offset=self.vol.offset - offset
The 'mnt_parent' member was moved from struct 'vfsmount' to struct
'mount' when the latter was introduced.
Alternatively, vmlinux.has_type('mount') can be used here but it is faster.
Returns:
bool: 'True' if the kernel
"""
return self.has_member("mnt_parent")
def is_equal(self, vfsmount_ptr) -> bool:
"""Helper to make sure it is comparing two pointers to 'vfsmount'.
Depending on the kernel version, the calling object (self) could be
a 'vfsmount \*' (<3.3.8) or a 'vfsmount' (>=3.3.8). This way we trust
in the framework "auto" dereferencing ability to assure that when we
reach this point 'self' will be a 'vfsmount' already and self.vol.offset
a 'vfsmount \*' and not a 'vfsmount \*\*'. The argument must be a 'vfsmount \*'.
Typically, it's called from do_get_path().
Args:
vfsmount_ptr (vfsmount \*): A pointer to a 'vfsmount'
Raises:
exceptions.VolatilityException: If vfsmount_ptr is not a 'vfsmount \*'
Returns:
bool: 'True' if the given argument points to the the same 'vfsmount'
as 'self'.
"""
if type(vfsmount_ptr) == objects.Pointer:
return self.vol.offset == vfsmount_ptr
else:
raise exceptions.VolatilityException(
"Unexpected argument type. It has to be a 'vfsmount *'"
)
def _get_real_mnt(self):
"""Gets the struct 'mount' containing this 'vfsmount'.
It should be only called from kernels >= 3.3.8 when 'struct mount' was introduced.
Returns:
mount: the struct 'mount' containing this 'vfsmount'.
"""
vmlinux = linux.LinuxUtilities.get_module_from_volobj_type(self._context, self)
return linux.LinuxUtilities.container_of(
self.vol.offset, "mount", "mnt", vmlinux
)
def get_vfsmnt_current(self):
"""Returns the current fs where we are mounted on
Returns:
A vfsmount pointer
"""
return self.get_mnt_parent()
def get_vfsmnt_parent(self):
"""Gets the parent fs (vfsmount) to where it's mounted on
Returns:
For kernels < 3.3.8: A vfsmount pointer
For kernels >= 3.3.8: A vfsmount object
"""
if self._is_kernel_prior_to_struct_mount():
return self.get_mnt_parent()
else:
return self._get_real_mnt().get_vfsmnt_parent()
def get_dentry_current(self):
"""Returns the root of the mounted tree
Returns:
A dentry pointer
"""
if self._is_kernel_prior_to_struct_mount():
return self.get_mnt_mountpoint()
else:
return self._get_real_mnt().get_dentry_current()
def get_dentry_parent(self):
"""Returns the parent root of the mounted tree
Returns:
A dentry pointer
"""
if self._is_kernel_prior_to_struct_mount():
return self.get_mnt_mountpoint()
else:
return self._get_real_mnt().get_mnt_mountpoint()
def get_mnt_parent(self):
if self.has_member("mnt_parent"):
"""Gets the mnt_parent member.
Returns:
For kernels < 3.3.8: A vfsmount pointer
For kernels >= 3.3.8: A mount pointer
"""
if self._is_kernel_prior_to_struct_mount():
return self.mnt_parent
else:
return self._get_real_mnt().mnt_parent
return self._get_real_mnt().get_mnt_parent()
def get_mnt_mountpoint(self):
"""Gets the dentry of the mountpoint
Returns:
A dentry pointer
"""
if self.has_member("mnt_mountpoint"):
return self.mnt_mountpoint
else:
@@ -809,6 +1093,41 @@ class vfsmount(objects.StructType):
def get_mnt_root(self):
return self.mnt_root
def has_parent(self) -> bool:
if self._is_kernel_prior_to_struct_mount():
return self.mnt_parent != self.vol.offset
else:
return self._get_real_mnt().has_parent()
def get_mnt_sb(self):
return self.mnt_sb
def get_flags_access(self) -> str:
return "ro" if self.mnt_flags & mount.MNT_READONLY else "rw"
def get_flags_opts(self) -> Iterable[str]:
flags = [
mntflagtxt
for mntflag, mntflagtxt in mount.MNT_FLAGS.items()
if mntflag & self.mnt_flags != 0
]
return flags
def get_mnt_flags(self):
return self.mnt_flags
def is_shared(self) -> bool:
return self.get_mnt_flags() & mount.MNT_SHARED
def is_unbindable(self) -> bool:
return self.get_mnt_flags() & mount.MNT_UNBINDABLE
def is_slave(self) -> bool:
return self.mnt_master and self.mnt_master.vol.offset != 0
def get_devname(self) -> str:
return utility.pointer_to_string(self.mnt_devname, count=255)
class kobject(objects.StructType):
def reference_count(self):
@@ -817,7 +1136,6 @@ class kobject(objects.StructType):
ret = refcnt.counter
else:
ret = refcnt.refs.counter
return ret
@@ -825,7 +1143,7 @@ class mnt_namespace(objects.StructType):
def get_inode(self):
if self.has_member("proc_inum"):
return self.proc_inum
elif self.ns.has_member("inum"):
elif self.has_member("ns") and self.ns.has_member("inum"):
return self.ns.inum
else:
raise AttributeError("Unable to find mnt_namespace inode")
@@ -836,7 +1154,6 @@ class mnt_namespace(objects.StructType):
if not self._context.symbol_space.has_type(mnt_type):
# Old kernels ~ 2.6
mnt_type = table_name + constants.BANG + "vfsmount"
for mount in self.list.to_list(mnt_type, "mnt_list"):
yield mount
@@ -844,10 +1161,13 @@ class mnt_namespace(objects.StructType):
class net(objects.StructType):
def get_inode(self):
if self.has_member("proc_inum"):
# 3.8.13 <= kernel < 3.19.8
return self.proc_inum
elif self.ns.has_member("inum"):
elif self.has_member("ns") and self.ns.has_member("inum"):
# kernel >= 3.19.8
return self.ns.inum
else:
# kernel < 3.8.13
raise AttributeError("Unable to find net_namespace inode")
@@ -861,7 +1181,6 @@ class socket(objects.StructType):
)
if not module_names:
raise ValueError(f"No module using the symbol table {symbol_table}")
kernel_module_name = module_names[0]
kernel = self._context.modules[kernel_module_name]
return kernel
@@ -871,7 +1190,6 @@ class socket(objects.StructType):
kernel = self._get_vol_kernel()
except ValueError:
return 0
socket_alloc = linux.LinuxUtilities.container_of(
self.vol.offset, "socket_alloc", "socket", kernel
)
@@ -897,7 +1215,6 @@ class sock(objects.StructType):
def get_inode(self):
if not self.sk_socket:
return 0
return self.sk_socket.get_inode()
def get_protocol(self):
@@ -907,7 +1224,6 @@ class sock(objects.StructType):
# Return the generic socket state
if self.has_member("sk"):
return self.sk.sk_socket.get_state()
return self.sk_socket.get_state()
@@ -915,7 +1231,6 @@ class unix_sock(objects.StructType):
def get_name(self):
if not self.addr:
return
sockaddr_un = self.addr.name.cast("sockaddr_un")
saddr = str(utility.array_to_string(sockaddr_un.sun_path))
return saddr
@@ -951,7 +1266,6 @@ class inet_sock(objects.StructType):
protocol = IP_PROTOCOLS.get(self.sk.sk_protocol)
if self.get_family() == "AF_INET6":
protocol = IPV6_PROTOCOLS.get(self.sk.sk_protocol, protocol)
return protocol
def get_state(self):
@@ -982,7 +1296,6 @@ class inet_sock(objects.StructType):
dport_le = sk_common.skc_dport
else:
return
return socket_module.htons(dport_le)
def get_src_addr(self):
@@ -1001,7 +1314,6 @@ class inet_sock(objects.StructType):
saddr = self.pinet6.saddr
else:
return
parent_layer = self._context.layers[self.vol.layer_name]
try:
addr_bytes = parent_layer.read(saddr.vol.offset, addr_size)
@@ -1010,7 +1322,6 @@ class inet_sock(objects.StructType):
f"Unable to read socket src address from {saddr.vol.offset:#x}"
)
return
return socket_module.inet_ntop(family, addr_bytes)
def get_dst_addr(self):
@@ -1032,7 +1343,6 @@ class inet_sock(objects.StructType):
addr_size = 16
else:
return
parent_layer = self._context.layers[self.vol.layer_name]
try:
addr_bytes = parent_layer.read(daddr.vol.offset, addr_size)
@@ -1041,7 +1351,6 @@ class inet_sock(objects.StructType):
f"Unable to read socket dst address from {daddr.vol.offset:#x}"
)
return
return socket_module.inet_ntop(family, addr_bytes)
@@ -1055,6 +1364,26 @@ class netlink_sock(objects.StructType):
# Return the generic socket state
return self.sk.sk_socket.get_state()
def get_portid(self):
if self.has_member("pid"):
# kernel < 3.7.10
return self.pid
if self.has_member("portid"):
# kernel >= 3.7.10
return self.portid
else:
raise AttributeError("Unable to find a source port id")
def get_dst_portid(self):
if self.has_member("dst_pid"):
# kernel < 3.7.10
return self.dst_pid
if self.has_member("dst_portid"):
# kernel >= 3.7.10
return self.dst_portid
else:
raise AttributeError("Unable to find a destination port id")
class vsock_sock(objects.StructType):
def get_protocol(self):
@@ -1101,3 +1430,154 @@ class xdp_sock(objects.StructType):
def get_state(self):
# xdp_sock.state is an enum
return self.state.lookup()
class bpf_prog(objects.StructType):
def get_type(self):
# The program type was in `bpf_prog_aux::prog_type` from 3.18.140 to
# 4.1.52 before it was moved to `bpf_prog::type`
if self.has_member("type"):
# kernel >= 4.1.52
return self.type
if self.has_member("aux") and self.aux:
if self.aux.has_member("prog_type"):
# 3.18.140 <= kernel < 4.1.52
return self.aux.prog_type
# kernel < 3.18.140
raise AttributeError("Unable to find the BPF type")
class cred(objects.StructType):
# struct cred was added in kernels 2.6.29
def _get_cred_int_value(self, member: str) -> int:
"""Helper to obtain the right cred member value for the current kernel.
Args:
member (str): The requested cred member name to obtain its value
Raises:
AttributeError: When the requested cred member doesn't exist
AttributeError: When the cred implementation is not supported.
Returns:
int: The cred member value
"""
if not self.has_member(member):
raise AttributeError(f"struct cred doesn't have a '{member}' member")
cred_val = self.member(member)
if hasattr(cred_val, "val"):
# From kernels 3.5.7 on it is a 'kuid_t' type
value = cred_val.val
elif isinstance(cred_val, objects.Integer):
# From at least 2.6.30 and until 3.5.7 it was a 'uid_t' type which was an 'unsigned int'
value = cred_val
else:
raise AttributeError("Kernel struct cred is not supported")
return int(value)
@property
def euid(self):
"""Returns the effective user ID
Returns:
int: the effective user ID value
"""
return self._get_cred_int_value("euid")
class kernel_cap_struct(objects.StructType):
# struct kernel_cap_struct was added in kernels 2.5.0
@classmethod
def get_last_cap_value(cls) -> int:
"""Returns the latest capability ID supported by the framework.
Returns:
int: The latest supported capability ID supported by the framework.
"""
return len(CAPABILITIES) - 1
def get_kernel_cap_full(self) -> int:
"""Return the maximum value allowed for this kernel for a capability
Returns:
int: _description_
"""
vmlinux = linux.LinuxUtilities.get_module_from_volobj_type(self._context, self)
try:
cap_last_cap = vmlinux.object_from_symbol(symbol_name="cap_last_cap")
except exceptions.SymbolError:
# It should be a kernel < 3.2, let's use our list of capabilities
cap_last_cap = self.get_last_cap_value()
return (1 << cap_last_cap + 1) - 1
@classmethod
def capabilities_to_string(cls, capabilities_bitfield: int) -> List[str]:
"""Translates a capability bitfield to a list of capability strings.
Args:
capabilities_bitfield (int): The capability bitfield value.
Returns:
List[str]: A list of capability strings.
"""
capabilities = []
for bit, name in enumerate(CAPABILITIES):
if capabilities_bitfield & (1 << bit) != 0:
capabilities.append(name)
return capabilities
def get_capabilities(self) -> int:
"""Returns the capability bitfield value
Returns:
int: The capability bitfield value.
"""
if isinstance(self.cap, objects.Array):
# In 2.6.25.x <= kernels < 6.3 kernel_cap_struct::cap is a two
# elements __u32 array that constitutes a 64bit bitfield.
# Technically, it can also be an array of 1 element if
# _KERNEL_CAPABILITY_U32S = _LINUX_CAPABILITY_U32S_1
# However, in the source code, that never happens.
# From 2.6.24 to 2.6.25 cap became an array of 2 elements.
cap_value = (self.cap[1] << 32) | self.cap[0]
else:
# In kernels < 2.6.25.x kernel_cap_struct::cap was a __u32
# In kernels >= 6.3 kernel_cap_struct::cap is a u64
cap_value = self.cap
return cap_value & self.get_kernel_cap_full()
def enumerate_capabilities(self) -> List[str]:
"""Returns the list of capability strings.
Returns:
List[str]: The list of capability strings.
"""
capabilities_value = self.get_capabilities()
return self.capabilities_to_string(capabilities_value)
def has_capability(self, capability: str) -> bool:
"""Checks if the given capability string is enabled.
Args:
capability (str): A string representing the capability i.e. dac_read_search
Raises:
AttributeError: If the given capability is unknown to the framework.
Returns:
bool: "True" if the given capability is enabled.
"""
if capability not in CAPABILITIES:
raise AttributeError(f"Unknown capability with name '{capability}'")
cap_value = 1 << CAPABILITIES.index(capability)
return cap_value & self.get_capabilities() != 0
@@ -0,0 +1,115 @@
{
"symbols": {
},
"user_types": {
"xen_p2m": {
"fields":{
"entries": {
"offset": 0,
"type": {
"count": 1,
"kind": "array",
"subtype": {
"kind": "base",
"name": "unsigned long long"
}
}
}
},
"kind": "struct",
"size": 8
},
"xen_pfn":{
"fields":{
"entries": {
"offset": 0,
"type": {
"count": 1,
"kind": "array",
"subtype": {
"kind": "base",
"name": "unsigned long long"
}
}
}
},
"kind": "struct",
"size": 16
},
"xen_pfn_entry":{
"fields":{
"pfn":{
"offset": 0,
"type": {
"kind": "base",
"name": "unsigned long long"
}
},
"gmfn":{
"offset": 8,
"type": {
"kind": "base",
"name": "unsigned long long"
}
}
},
"kind": "struct",
"size": 16
}
},
"enums": {
},
"base_types": {
"unsigned char": {
"endian": "little",
"kind": "char",
"signed": false,
"size": 1
},
"unsigned short": {
"endian": "little",
"kind": "int",
"signed": false,
"size": 2
},
"long": {
"endian": "little",
"kind": "int",
"signed": true,
"size": 4
},
"char": {
"endian": "little",
"kind": "char",
"signed": true,
"size": 1
},
"unsigned long": {
"endian": "little",
"kind": "int",
"signed": false,
"size": 4
},
"long long": {
"endian": "little",
"kind": "int",
"signed": true,
"size": 8
},
"unsigned long long": {
"endian": "little",
"kind": "int",
"signed": false,
"size": 8
}
},
"metadata": {
"producer": {
"version": "0.0.1",
"name": "ikelos-by-hand",
"datetime": "2023-01-09T00:51:00"
},
"format": "6.1.0"
}
}
@@ -397,12 +397,9 @@ class OBJECT_HEADER(objects.StructType):
symbol_table_name = self.vol.type_name.split(constants.BANG)[0]
try:
header_offset = self.NameInfoOffset
except AttributeError:
# http://codemachine.com/article_objectheader.html (Windows 7 and later)
name_info_bit = 0x2
if symbol_table_name in self._context.modules:
ntkrnlmp = self._context.modules[symbol_table_name]
else:
layer = self._context.layers[self.vol.native_layer_name]
kvo = layer.config.get("kernel_virtual_offset", None)
@@ -411,16 +408,24 @@ class OBJECT_HEADER(objects.StructType):
f"Could not find kernel_virtual_offset for layer: {self.vol.layer_name}"
)
# We know this symbol table name can't exist because we checked for it earlier
ntkrnlmp = self._context.module(
symbol_table_name, layer_name=self.vol.layer_name, offset=kvo
)
try:
header_offset = self.NameInfoOffset
except AttributeError:
# http://codemachine.com/article_objectheader.html (Windows 7 and later)
name_info_bit = 0x2
address = ntkrnlmp.get_symbol("ObpInfoMaskToOffset").address
calculated_index = self.InfoMask & (name_info_bit | (name_info_bit - 1))
header_offset = self._context.object(
symbol_table_name + constants.BANG + "unsigned char",
header_offset = ntkrnlmp.object(
"unsigned char",
layer_name=self.vol.native_layer_name,
offset=kvo + address + calculated_index,
offset=address + calculated_index,
)
if header_offset == 0:
@@ -430,10 +435,11 @@ class OBJECT_HEADER(objects.StructType):
)
)
header = self._context.object(
symbol_table_name + constants.BANG + "_OBJECT_HEADER_NAME_INFO",
header = ntkrnlmp.object(
"_OBJECT_HEADER_NAME_INFO",
layer_name=self.vol.layer_name,
offset=self.vol.offset - header_offset,
native_layer_name=self.vol.native_layer_name,
absolute=True,
)
return header
@@ -13,6 +13,8 @@ vollog = logging.getLogger(__name__)
class Statistics(plugins.PluginInterface):
"""Lists statistics about the memory space."""
_required_framework_version = (2, 0, 0)
@classmethod
Regular → Executable
View File