Laika BOSS: Object Scanning System

Laika is an object scanner and intrusion detection system that strives to achieve the following goals:

• Scalable

  • Work across multiple systems
  • High volume of input from many sources

• Flexible

  • Modular architecture
  • Highly configurable dispatching and dispositioning logic
  • Tactical code insertion (without needing restart)

• Verbose

  • Generate more metadata than you know what to do with

Each scan does three main actions on each object:

• Extract child objects Some objects are archives, some are wrappers, and others are obfuscators. Whatever the case may be, find children objects that should be scanned recursively by extracting them out.

• Mark flags Flags provide a means for dispositioning objects and for pivoting on future analysis.

• Add metadata Discover as much information describing the object for future analysis.

Feel free to read the whitepaper!

Components

Laika is composed of the following pieces:

• Framework (laika.py) This is the core of Laika BOSS. It includes the object model and the dispatching logic.

• laikad This piece contains the code for running Laika as a deamonized, networked service using the ZeroMQ broker.

• cloudscan A command-line client for sending a local system file to a running service instance of Laika (laikad).

• modules The scan itself is composed of the running of modules. Each module is its own program that focuses on a particular sub-component of the overall file analysis.

Getting Started

Laika BOSS has been tested on the latest versions of CentOS and Ubuntu LTS

Installing on Ubuntu

1. Install framework dependencies:

   apt-get install yara python-yara python-progressbar python-pip
   pip install interruptingcow

2. Install network client and server dependencies:

   apt-get install libzmq3 python-zmq python-gevent python-pexpect

3. Install module dependencies:

   apt-get install python-ipy python-m2crypto python-pyclamd liblzma5 libimage-exiftool-perl python-msgpack libfuzzy-dev python-cffi python-dev unrar
   pip install fluent-logger olefile ssdeep py-unrar2 pylzma javatools
   wget https://github.com/smarnach/pyexiftool/archive/master.zip
   unzip master.zip
   cd pyexiftool-master
   python setup.py build
   python setup.py install
   wget https://github.com/erocarrera/pefile/archive/pefile-1.2.10-139.tar.gz
   tar vxzf pefile-1.2.10-139.tar.gz
   cd pefile-1.2.10-139
   python setup.py build
   python setup.py install

Installing on CentOS

1. Install framework dependencies

   sudo yum install -y epel-release
   sudo yum install -y autoconf automake libtool libffi-devel python-devel python-pip python-zmq ssdeep-devel swig

2. Install Python modules

   pip install IPy cffi interruptingcow fluent-logger javatools m2crypto olefile pylzma pyclamd py-unrar2
   pip install six --upgrade --force-reinstall
   pip install ssdeep

3. Install Yara

   There is no Yara package for CentOS, so we have to build it from source. You can't use a checkout from Github as it won't contain the Python code; you must download one of the release versions. The following uses Yara version 3.5.0

   wget https://github.com/VirusTotal/yara/archive/v3.5.0.zip
   unzip yara-3.5.0.zip
   cd yara-3.5.0
   chmod +x ./build.sh
   ./build.sh
   sudo make install
   cd yara-python
   python setup.py build
   sudo python setup.py install

4. Install pyexif

   wget https://github.com/smarnach/pyexiftool/archive/master.zip
   unzip master.zip
   python setup.py build
   sudo python setup.py install

5. Install pefile

   wget https://github.com/erocarrera/pefile/archive/pefile-1.2.10-139.tar.gz
   tar vxzf pefile-1.2.10-139.tar.gz
   cd pefile-1.2.10-139
   python setup.py build
   python setup.py install --user

You may need to set the LD_LIBRARY_PATH variable to include /usr/local/lib when running Laika.

Installing Laika BOSS (optional)

You may use the provided setup script to install the Laika BOSS framework, client library, modules and associated scripts (laika.py, laikad.py, cloudscan.py).

python setup.py install

Standalone instance

From the directory containing the framework code, you may run the standalone scanner, laika.py against any file you choose. If you move this file from this directory you'll have to specify various config locations. By default it uses the configurations in the ./etc directory.

We recommend using installing jq to parse Laika output.

$ ./laika.py ~/test_files/testfile.cws.swf | jq '.scan_result[] | { "file type" : .fileType, "flags" : .flags, "md5" : .objectHash }'
100%[############################################] Processed: 1/1 total files (Elapsed Time: 0:00:00) Time: 0:00:00
{
  "md5": "dffcc2464911077d8ecd352f3d611ecc",
  "flags": [],
  "file type": [
    "cws",
    "swf"
  ]
}
{
  "md5": "587c8ac651011bc23ecefecd4c253cd4",
  "flags": [],
  "file type": [
    "fws",
    "swf"
  ]
}

Networked instance

$ ./laikad.py

$ ./cloudscan.py ~/test_files/testfile.cws.swf | jq '.scan_result[] | { "file type" : .fileType, "flags" : .flags, "md5" : .objectHash }'
{
  "md5": "dffcc2464911077d8ecd352f3d611ecc",
  "flags": [],
  "file type": [
    "cws",
    "swf"
  ]
}
{
  "md5": "587c8ac651011bc23ecefecd4c253cd4",
  "flags": [],
  "file type": [
    "fws",
    "swf"
  ]
}

Milter

The Laika BOSS milter server allows you to integrate Laika BOSS with mail transfer agents such as Sendmail or Postfix. This enables better visibility (passive visibility can be hampered by TLS) and provides a means to block email according to Laika BOSS disposition.

+----------------+             +---------------+             +----------------+
|                |    email    |               |   email     |                |
|    sendmail    +------------->  laikamilter  +------------->     laikad     |
|                | accept/deny |               | scan result |                |
|                <-------------+               <-------------+                |
+----------------+             +---------------+             +----------------+

The Laika BOSS milter server requires the python-milter module and the Laika BOSS client library. Check out the comments in the source code for more details.

Suricata Integration Prototype

We have released a proof of concept feature for Suricata that allows it to store extracted files and their associated metadata in a Redis database. You will find this code under a new branch in our Suricata fork. We hope to refine the implementation and eventually have it accepted by the project.

Once you've enabled file extraction and the optional Redis integration in Suricata, you can extract these files from Redis and submit them to Laika BOSS for scanning by using the middleware script laika_redis_client.py as shown below. Note that it requires the python-redis module.

First, start laikad.py in async mode:

./laikad.py -a

Then launch the middleware script and give it the address of the laikad broker and Redis database (defaults shown below):

./laika_redis_client.py -b tcp://localhost:5558 -r localhost -p 6379

Note that you will need to use a logging module such as LOG_FLUENT to export the full scan result of the these file scans from laikad.

Licensing

The Laika framework and associated modules are released under the terms of the Apache 2.0 license.