Introduction

pyrefs provides a series of libraries helpful to parse and analyze Resilient FileSystems (ReFS). An interactive application to analyze a ReFS dump named Igor is also provided.

pyrefs is still very immature, as of today it has been only tested with partitions using ReFS version 1.1. The code is neither far from perfect, and it would need a good refactoring. We are working on that, but be aware when you retrieve pyrefs.

Finally, pyrefs naming convention for ReFS structures is mainly guided by Henry Georges thesis work. You can find Henry Georges's thesis and the code he developed to analyze ReFS here https://brage.bibsys.no/xmlui/handle/11250/2502565.

Authors

Daniel Gracia Pérez <daniel.gracia-perez@cfa-afti.fr>
Eric Jouenne <eric.jouenne@cfa-afti.fr>
Tony Gerard <tony.gerard@cfa-afti.fr>
Francois-Xavier Babin <francois-xavier.babin@cfa-afti.fr>

Thanks

We would like to thank: Henry Georges, his thesis document has been the foundation of this project; Willi Ballenthin, for the ReFS dump images on his website and his ReFS articles; and all the other authors providing information on the ReFS filesystem.

We would like to also thank Juan Romero, Olivier Gilles, Isabelle Isabelle Moukala-Mouberi and Claudia Duhoo, who performed a similar work than ours. Your feedback was unvaluable.

Dependencies

pyrefs was developed using Python 3 (Python 3.6.6) at the time of this release, and has been tested to work on Windows and Linux platforms.

No special dependencies, if you find any please let us know.

Sources organization

.
+-- README.md
+-- LICENSE
+-- igor.py
+-- part
|   +-- refs
|       +-- allocator.py
|       +-- attribute.py
|       +-- entry_block.py
|       +-- object.py
|       +-- object_tree.py
|       +-- tree_control.py
|       +-- vol.py
+-- media
|   +-- gpt.py
|   +-- mbr.py
+-- util
|   +-- carving.py
|   +-- filetree.py
|   +-- hexdump.py
|   +-- table.py
|   +-- time.py
+-- examples
    +-- carving.rec
    +-- filetree.rec
    +-- tree_explorer.rec

Igor

An interactive command line tool is provided to facilitate the analysis of those not willing to play with the code. We call it Igor and it can be launched with Python.

Igor provides a limited toolset to analyze ReFS partitions exploiting the pyrefs library, the following sections enumerate them.

Examples

Some examples of Igor usage are provided in the examples folder as Igor records. To execute/playback the examples contained in the examples folder download the Windows8ReFSImages.zip file provided by Willi Ballenthin at http://www.williballenthin.com/forensics/refs/test_images/, copy it to the examples/dump folder and uncompress it. Then you can use igor to test the records in the examples folder with the command playback examples/<example_file>.

Command reference

file

Usage: file [-h] [-i] [-f] [-F] dump

Load the provided dump file for analysis, and automatically select the ReFS partition for you. You can also initialize the list of entryblocks of the partition.

Positional arguments:

dump: File to use as dump for the analysis.

Optional arguments:

-h, --help: show this help message and exit
-i, --initiliaze-entryblocks: find blocks in provided dump
-f, --files: find files in provided dump (only considered if -i defined)
-F, --folders: find folders in provided dump (only considered if -i defined)

vol

Usage: vol [-h]

Dump the volume record information from the current ReFS partition.

Optional arguments:

-h, --help: show this help message and exit

Dump the volume record information from the current ReFS partition.

part

Usage: part [-h] [partidx]

Show available partitions in the currently loaded dump if no parameter is given, switch to the provided partition if any provided.

Positional arguments:

partidx: Partition index of the partition to use for analysis

Optional arguments:

-h, --help: show this help message and exit

find_entryblocks

Usage: find_entryblocks [-h] [-f] [-F]

Find and show all the entryblocks in current partition. If requested number of files and folders information will be also collected.

Optional arguments:

-h, --help: show this help message and exit
-f, --files: Collect information on the number of files in the entryblocks.
-F, --folders: Collect information on the number of folders in the entryblocks.

find_pattern

Usage: find_pattern [-h] pattern

Find a data pattern in all the blocks of the current partition. Special characters (including spaces, carriage return, etc.) are not allowed in the pattern, think of escaping them.

Positional arguments:

pattern: Pattern to find in the current partition blocks.

Optional arguments:

-h, --help: show this help message and exit

list_filenames

Usage: list_filenames [-h]

List the found filenames from the list of entryblocks with filenames.

Optional arguments:

-h, --help: show this help message and exit

list_folders

Usage: list_folders [-h]

List the found filename folders from the list of entryblocks with folders.

Optional arguments:

-h, --help: show this help message and exit

entryblock

Usage: entryblock [-h] entryblock_identifier

Dump the provided entryblock identifier as EntryBlock.

Positional arguments:

entryblock_identifier: entryblock identifier of the EntryBlock to dump

Optional arguments:

-h, --help: show this help message and exit

tree_control

Usage: tree_control [-h] [entryblock_identifier]

Parse the given entryblock identifier as a TreeControl. If no entryblock identifier is provided it uses 0x1e as entryblock identifier.

Positional arguments:

entryblock_identifier: entryblock identifier of the TreeControl to dump

Optional arguments:

-h, --help: show this help message and exit

tree_control_extension

Usage: tree_control_extension [-h] entryblock_identifier

Parse the given entryblock identifier as a TreeControlExtension.

Positional arguments:

entryblock_identifier: entryblock identifier of the TreeControlExtension to dump

Optional arguments:

-h, --help: show this help message and exit

object_tree

Usage: object_tree [-h] entryblock_identifier

Parse the given entryblock identifier as an ObjectTree.

Positional arguments:

entryblock_identifier: entryblock identifier of the ObjectTree to dump

Optional arguments:

-h, --help: show this help message and exit

allocator

Usage: allocator [-h] entryblock_identifier

Parse the given entryblock identifier as an Allocator.

Positional arguments:

entryblock_identifier: entryblock identifier of the Allocator to dump

Optional arguments:

-h, --help: show this help message and exit

attribute

Usage: attribute [-h] dump_offset

Parse the given dump offset (in bytes) as an Attribute.

Positional arguments:

dump_offset: dump offset (in bytes) of the Attribute to dump

Optional arguments:

-h, --help: show this help message and exit

datastream

Usage:

datastream [-h] output_filename output_size
           entryblock_identifier,number_of_blocks
           [entryblock_identifier,number_of_blocks ...]

Extract data from dump from the given datarun. Dataruns can be extracted by exploring the EntryBlocks (command entryblock) or using the list_dataruns command.

Positional arguments:

output_filename: name of the generated output file
output_size: total amount of data to extract
entryblock_identifier,number_of_blocks: datarun to follow for the data extraction

Optional arguments:

-h, --help: show this help message and exit

list_dataruns

Usage: list_dataruns [-h]

Retrieve list of all the files dataruns.

Optional arguments:

-h, --help: show this help message and exit

filetree

Usage: filetree [-h] node_id

Extract the file tree structure from the given node (use node 0x600 by default).

Positional arguments:

node_id: node identifier of the node to extract the file tree structure from

Optional arguments:

-h, --help: show this help message and exit

hexdump

Usage: hexdump [-h] dump_offset size

Hexdump the number of bytes at the provided offset.

Positional arguments:

dump_offset: dump offset of the hexdump start
size: number of bytes to dump

Optional arguments:

-h, --help: show this help message and exit

hexblock

Usage: hexblock [-h] entryblock_id

Hexdump the block with the provided entryblock identifier.

Positional arguments:

entryblock_id: entryblock identifier to hexdump

Optional arguments:

-h, --help: show this help message and exit

bye or <ctrl-d>

Usage: bye [-h]

Exit the program. Are you sure?

Optional arguments:

-h, --help: show this help message and exit

record

Usage: record [-h] output_file

Save the following commands to selected file. Saved file can be used afterwards for replay with the playback command.

Positional arguments:

output_file: file onto which commands will be saved

Optional arguments:

-h, --help: show this help message and exit

playback

Usage: playback [-h] input_file

Execute the sequence of commands defined in the input file.

Positional arguments:

input_file: file from which commands to execute will be read

Optional arguments:

-h, --help: show this help message and exit

Dafti/pyrefs

Introduction

Authors

Thanks

Dependencies

Sources organization

Igor

Examples

Command reference

file

vol

part

find_entryblocks

find_pattern

list_filenames

list_folders

entryblock

tree_control

tree_control_extension

object_tree

allocator

attribute

datastream

list_dataruns

filetree

hexdump

hexblock

bye or <ctrl-d>

record

playback