Fedora Red Team
The goal of the Fedora Red Team (FRT) is to become Red Hat's upstream cybersecurity community. This SIG is the community focal point for offensive tooling, exploit curation, standards, and reference architectures.
Get in contact with the SIG:
- IRC: #fedora-security
- List: firstname.lastname@example.org
First Friday of every month at 10am US/Eastern (1400 UTC) in #fedora-security.
To be the catalyst in the enterprise Linux community that enables Computer Network Operations, the open source way.
Defining the term "cyber"
The term "cyber" itself is somewhat overloaded.
For purposes of this SIG, let the term “cyber” be defined as:
- Being rooted in the Greek word, κυβερνάω (kubernao), literally, to steer or govern. Note that this is the same root word for Kubernetes.
- Based on the term “cybernetics,” which was coined by Norbert Wiener during WWII, and later morphed into “cyberspace,” which was popularized by William Gibson’s Neuromancer. Justice Department attorney Michael Vatis, in need of a term to describe the new domain, referred to Neuromancer during the development of the Marsh Report. See this blog post for additional information.
- Defined as the “fifth domain of warfare” by the US Department of Defense (the others being land, sea, air, and space)
- Used as an umbrella term to describe the Computer Network Operations (CNO) triplet of Computer Network Defense (CND), Computer Network Exploitation (CNE), and Computer Network Attack (CNA)
Where terms like security and infosec are broad and refer to a branch of computer science, cyber should be understood to refer to the antagonistic world of nation state electronic warfare, Advanced Persistent Threat (APT) actors, and cyber criminals.
In the context defined above, Red Teams are “friendlies” who attempt to break systems being defended by the Blue Teams in order to identify and remediate vulnerabilities. For those interested in the concept of "red teaming," refer to Micha Zenko's book, Red Team: How to Succeed By Thinking Like the Enemy.
It’s from this Red-vs-Blue concept that the Fedora Red Team draws its naming inspiration.
Relationship with Fedora Security Lab
Fedora Security Lab is a spin of Fedora that focuses on packaging security tools. Packaging of these tools is an important task, but with the advent and adoption rate of the Open Container Image Format, rpm packaging has become less critical.
Container images for discrete cyber tools may have a wider impact. See the section below on Red Container.
This SIG will collaborate with the Security Lab spin wherever appropriate.
FRT will make original tools, starting with those listed below.
Enterprise Linux Exploit Mapper (ELEM) uses the yum security plugin and Red Hat Security Data API to map CVEs to known exploits in the wild. The initial data source for exploits will be the Exploit Database, but other data sources can be added later.
An important aspect of this project is “exploit curation.” Many vulnerabilities discovered and documented via CVE contain theoretical vulnerabilities. These CVEs use language like “this vulnerability may permit an attacker to execute arbitrary code.” Other vulnerabilities are associated with exploits that have been tested and are known to permit remote code execution, local privilege escalation, or other classes of exploits.
The exploit curation process requires FRT testers to:
- Map local CVEs to exploits on Exploit DB
- Test those exploits on vulnerable systems
- Document which exploits are able to leverage the vulnerability
This project will initially curate CentOS vulnerabilities. CentOS CVEs and RHEL CVEs are mostly the same, and Fedora CVE mapping is more difficult than querying the Security Data API.
Fedora Security Data API
Fedora Security Data API provides CVE mappings and security data a la the Red Hat Security Data API.
Fedora Cyber Test Lab is an open source implementation of the Cyber-ITL dynamic analysis approach. Originally described at Def Con 24, the Cyber-ITL examines disassembled binaries for complexity, "application armoring," developer hygene, and susceptibility to fuzzing.
Aside from the obvious benefits of bug hunting at scale, the Cyber-ITLs reports will probably be used to quantify software risk and enable a burgeoning cyber insurance industry.
At Def Con 25, the Cyber-ITL provided an update including a histogram of the latest RHEL 7 AMI’s security performance. While RHEL’s performance wasn’t any worse than other Linuxes -- it actually appeared to be slightly better -- the results are not repeatable.
The Fedora Cyber Test Lab (FCTL) will endeavor to recreate the Cyber-ITL’s approach in an open source, repeatable fashion. The initial release of the FCTL tooling will include:
- Cyclomatic complexity
- Branch frequency
- Hardening features as identified by hardening-check(1)
- Known-bad functions
- Note: without the Cyber-ITLs mass fuzzing results, we have no way of scoring functions in terms of “badness.” Instead we’ll track the presence or absence of functions on a bad list and sum the hits. This is not as high-fidelity as the Cyber-ITL’s approach, but it’s a start.
The FCTL will make its findings open source, and provide Ansible test harnesses to recreate those findings. While the FRT will focus on RHEL, CentOS, and Fedora, the project will be open and receptive to community contributions for other OSes.
Adoption of the Open Container Image Format is changing the packaging landscape. RPMs are still useful, but getting community adoption and participation in a containerization effort is easier than that of a packaging effort.
The Red Container project will work on containerizing the popular tooling in Kali Linux so that they can be run from any Linux distro, including Fedora, CentOS, and RHEL.
Initial focus will be given to pentesting tools not already packaged in the Fedora repos or in EPEL such as:
Other tools will be added after that, including a controlled interface to install package groups similar to those of Kali, and to invoke those tools from the CLI without having to call
The Fedora Red Team will test vulnerability / exploit combinations and document the ones that can be leveraged by attackers. This curation will take place under the ELEM project, but if you want to find a place to jump in, check out our public Trello board.
Pen Testing Execution Standard (PTES)
The PTES project is very useful, although outdated. The FRT plans to help the founders re-invigorate the project, and to use it as our process and documentation standard. Stay tuned for updates.
A common thread in US DoD cyber procurements have been so-called “cyber ranges.” While the concept of a cyber range has yet to be officially defined by NIST, it can be thought of as a cyber analog to kinetic weapons ranges. New machine guns aren’t developed and tested near civilians, they’re developed in special labs and tested in safe, specialized locations. Cyber weapons and defensive technologies require their own safe places. We call those cyber ranges.
Technologies like RDO, Origin, ManageIQ, and Ansible fit perfectly into the problem space of cyber ranges, the gross functions for which could be described as:
- Training and Simulation: practicing Red Team and Blue Team operations
- Analysis: reverse engineering, sim/stim, C2 hijacking, attribution
- Research & Development: building new offensive and defensive capabilities
Building an environment that can support these features is very much an “infrastructure problem,” but it spans multiple project upstreams -- RDO, Origin, ManageIQ, etc. It makes sense for this SIG to be the fulcrum for cyber range designs, white papers, reference architectures, etc.
Red Team SIG reference architectures will be maintained under this FRT GitHub project.
How to contribute
Want to join the Fedora Red Team? Great! Basically, you just have to show up.
- Join us for our monthly meeting on #fedora-security on Freenode IRC
- Email us your thoughts at email@example.com
- Hit us up with pull requests on any of the GitHub projects