This is a quick introduction to Step 2 of the Risk Management Framework NIST 800-37 process. Step 2 involves selection of NIST Special Publication 800-53 security controls. There are (3) main tasks that you must do in this step:
1) Select the applicable baseline controls. Selection of baseline controls is based on system categorization.
2) Tailor the Security Controls to the system. Not all security controls can be used because they may break your system. And in some cases they are simply not applicable. There are also Common Controls, Hybrid controls, and system specific controls.
3) Document the Security Controls. You must document the selected security controls in a system security plan and have the security controls reviewed.
Risk = ((Vulnerability * Threat) / Countermeasure) * Asset Value at Risk IT Risk
Risk is the likelihood that a threat will compromise the weakness of an asset. So risk identification is based on knowing the threat, the vulnerability and the asset.
The better you understand these factors, the better your chances of risk determination and risk identification.
NIST SP 800-39, Managing Information Security Risk is a document that defines risk management as the process of not only identifying risk but also, assessing risk, and taking steps to mitigate risks for fit within one of the types or risk, risk acceptance.
Risk identification starts off with identification of the asset.
1) System characterization – Gather information into a System Security Plan (SSP). Identifying risk requires a deep understanding of the asset and its environment. Asset information for the SSP will cover the following:
System interfaces (e.g., internal and external connectivity)
Data and information
Persons who support and use the IT system
System mission (e.g., the processes performed by the IT system)
System and data critical (e.g., the system’s value or importance to an organization)
System and data sensitivity
2) Threat Identification – Without a defined threat, there is no way to quantify or identify a threat. Threat identification starts with looking at the threat sources and events. An example to look for threat sources historical data. When has the asset or similar assets from other related organizations in the same industry been attacked or suffered disaster. Remember a threat source is not just criminal-hackers and malware, it can be a natural disaster or unintentional destruction of data or a power outage.
3) Vulnerability Identification – Once the asset and threat are identified, you can more easily determine if your system has a weakness for that particular disaster or exploit. Again, you can look at historical evidence that suggest weakness. You can use scanners to find open ports that are exposed to the Internet.
4) Security Control Analysis – If your system already has security controls in place, you must take that into account because this may lower your risk.
5) Likelihood determination – The probability that your asset will be exploited is based on the threat source motivation, threat capability, your vulnerability and the security controls you have in place. Based on all these factors you can calculated the likelihood of an attack or disaster.
6) Impact Analysis – This where you gather all the data from asset identification, threat source, vulnerability identification, security controls, likelihood of attack and figure you what would happen if something really did happen. How important is your system and its data? What would happen to the mission or bottom line or profits if the system went down for a few hours? a few days? a few weeks? Some system are so important that they cannot be down for even a minute. Impact is very important to the level of risk. The more important the system is, the high the risk.
7) Risk Determination / Risk Identification – Based on all the data gathered you can make a pretty good risk determination. You should have defined the systems components and what data is important, made a pretty good conclusion on threat sources and likelihood of the vulnerability exploits and know exactly what kind of impact there will be if the system goes down.
Who does risk identification:
Ultimately it is the information system owner and authorizing official that must make a determination on what kind of risk they will accept, but they rely heavily on the expertise of an information security engineer, information system security manager, information system security officer and technical professionals to articulate what is happening on the ground.
The ISSO/ISSM/ISSE typically document the process mentioned above or the DIARMF process. Security professionals coordinate with IT professionals to “get into the weeds” of technical security controls and vulnerabilities.
The risk management framework steps are detailed in NIST SP 800-37, Guide for Applying the Risk Management Framework to Federal Information Systems.
The DoD has recently adopted the Risk Management Framework steps (called the DIARMF process). There are 6 step: Categorize, Select, Implement, Assess, Authorize and Continuous Monitor.
risk management framework – Step 1. Categorize
The first risk management framework step is categorization. This step consists of classifying the importance of the information system. This is done by the system owner with FIPS 199 and NIST 800-60.
Categorization is based on how much negative impact the organization will receive if the information system lost is confidentiality, integrity or availability.
risk management framework – Step 2. Select
With FIPS 200 and NIST SP 800-53, the organization responsible for the systems security will select the security controls required to limit the risk to their organization. The selection of the controls is based on the categorization of your system. A system security plan is created as a guide to what will be installed and/or configured on the system.
Using the System Security Plan, the organization responsible for the categorized system can begin risk management framework step 3. This step is implementation which is installation and configuration of security patches, hotfixes and security devices where necessary. Guidance for actual implantation has to come from technical manuals, system administrators, system engineers and others technically competent enough to do the work.
The organization has to make sure that the security controls are implemented properly. This is done in risk management step 4, assess. Using NIST SP 800-53A, Guide for Assessing the Security Controls in Federal Information Systems and Organizations is used to determine which controls have been fully implemented to limit the risks to the organization.
Even after implementation and assessment of the security controls that limits the over all risk to the organization, there is some remaining (residual) risk. The organization must have someone who has enough authority of over the system to accept the residual risk. This person is known as the Authorizing Official.
In risk management framework step 5, an Authorizing Official makes a formal, written acceptance of the risks. The AO makes a decision on whether or not to accept the risk based on the authorization package. The authorization package consists of the system security plan, plan of action and milestone, security/risk assessment report and any other supporting documents.
After acceptance of risk by the organization, they must develop a program that monitors the ongoing changes to the systems security posture. They take a proactive approach to watching for advanced persistent threats, configuration changes and new vulnerabilities. Risk management framework step 6 handles all of this.
The documents for putting DIARMF Categorization and Selection of security controls together are FIPS 200, CNSSI-1253 and NIST SP 800-53. DoD 8510 is based on these documents.
FIPS 200, Minimum Security Requirements for Federal Information and Information Systems is a bridge between the FIPS 199 and the security controls documented in NIST SP 800-53. It sets forth the initial set of baseline security controls for your system based on the system impact level and minimum security requirements FIPS 200 is a very short document that explains the levels of impact that your system has based on your systems security categorization and how the security controls will be selected.
FIPS 200 mentions seventeen security-related areas (more current revisions of SP 800-53 have more security areas) with regard to protecting the confidentiality, integrity, and availability of data processed, transmitted and stored on government systems. The security-related areas include:
awareness and training
audit and accountability
certification, accreditation security assessments
identification and authentication
physical and environmental protection
systems and services acquisition
system and communications protection
system and information integrity
Addressing each of the 17 DIARMF areas will give the organization a comprehensive selection of security controls. These controls address technical, operational and management aspects of security and risk management.
The DIARMF Selected security controls come from NIST SP 800-53, Recommended Security Controls for Federal Information Systems and Organizations. Recommended Security controls are prescribe to the security categorization of your system.
Initial Set of Baseline Security Controls
The 17 DIARMF areas are broken out in NIST SP 800-53 Appendix D. Appendix D is the DIARMF Initial Set Baseline Security Controls. CNSSI 1253, Security Categorization and control Selection for National Security Systems is specifically for National Security systems and offers a few alternatives to normal federals systems.
What are National Security Systems?
(OMB Circular A-130, FIPS 200, NIST 137) Any telecommunications or information system operated by the United States Government, the function, operation, or use of which (1) involves intelligence activities; (2) involves cryptologic activities related to national security; (3) involves command and control of military forces; (4) involves equipment that is an integral part of a weapon or weapons system; or (5) is critical to the direct fulfillment of military or intelligence missions, but excluding any system that is to be used for routine administrative and business applications (including payroll, finance, logistics, and personnel management applications).
The differences are based on specific challenges of national security systems (aka mission systems) that set them apart from normal federals systems. The following adjustments are mentioned in CNSSI to deal with NSS systems:
NO “High Water Mark” – normally federal systems are take on impact level of their most important security factor. NSS systems don’t apply HWM. Instead, confidentiality, integrity and availability retain their impact levels.
Confidentiality is treated different – Other factors affect confidentiality: aggregation of information on the system, system environment, and attributes of users. For example, and IP address by itself is not classified, but an IP and a vulnerability and the system description and location is probably classified.
Appendix K, Overlay – NSS have to use security overlays in addition to the initial security baseline. See CNSSI for more details.
Reciprocity – One of the biggest problems with the old C&A was that when one military unit would have a need to connect to a military or intel unit from a different branch, they would end up having to do two or three different C&A processes. This was VERY costly.. and stupid. So the new DIARMF is pushing away from that with Reciprocity. Its something they started during DIACAP with different degrees of success.
Tailoring means aligning the selected security controls to your system. For example, if you have a bunch of security controls that apply to Internet browsers. And the controls are designed to minimize the possibility of your browser being hacked from the Internet. But your system HAS NO INTERNET CONNECTION and only uses browsers for reading HTML based manuals on your Internal network. If you are tailoring, you could cut all those security controls out because they are not applicable to your network.
You are applying ONLY security baseline controls that fit within the SCOPE of your systems capabilities. And you are applying the MOST critical controls over more expensive security controls that have little or no impact to minimizing the risk. So scoping and tailoring is something that requires a security professional and someone that knows the system well enough to know what controls are really needed and which one are not.
During the selection process your Information System Security Officer should be working on the System Security Plan. The details of what the SSP looks like is in NIST SP 800-18, Guide for Developing Security Plans for Federal Information Systems. The SSP is like your play book on what will be implemented and why.
It should be robust enough to guide the implementation of the security controls but flexible enough so that you can continue to add to it during the implementation process.
After the implementation of the security controls, the SSP will be needed for the Assessment of the security controls.