CHAPTER 1 - PART 1
1 BACKGROUND
Networks
and information technology (IT) resources are continually vulnerable to
illegal/malicious activity or exploitation by internal and external
sources. Cyber Security(CS) incident
handling is an important and required component of USDA’s CS program. CS related threats can exploit
vulnerabilities in new or rapidly changing IT.
The most common security threats are those that travel through and to
networked systems. While it is
impossible to eliminate all CS incidents, proactive incident prevention is a
critical element of a mature incident management capability.
Preventative
procedures such as patch management, firewalls, risk and vulnerability
assessments and mitigation can reduce incidents. Not all incidents can be prevented. A flexible and adaptable incident response
capability is a necessary part of managing network security threats. Damage to IT systems from a CS incident can
occur in a short period. It is
essential that all USDA organizations (agencies, staff offices, projects,
mission areas, and contractor managed locations) have procedures in place that
can be activated immediately. The
inability of any USDA organization to recognize and promptly report incidents
impacts and potentially compromises the information systems security program
(ISSP) efforts of other USDA organizations and their customers.
The
Federal Information Security Management Act (FISMA) of 2002 requires Federal
agencies to establish incident response and handling capabilities. The law also requires USDA to report
incidents to United
States Computer Emergency Response Team (US-CERT) (formerly FedCIRC)
in the Department of Homeland Security (DHS).
Each Federal agency is required to designate a primary and secondary
Point of Contact (POC) with US-CERT. The
USDA US-CERT POC is located in OCIO CS.
Each USDA agency, mission area and staff office is required to
communicate with US-CERT through OCIO CS.
The
need for an incident handling capability within USDA organizations that crosses
agency boundaries has never been greater.
This need will continue as long as those who exploit IT exist. Standard reporting and uniform operating
procedures permit USDA and US-CERT to be better positioned for assessing risks,
addressing vulnerabilities, reducing overall costs and meeting the security
challenges of USDA’s information infrastructure.
2 POLICY
This
chapter establishes the minimum policy and procedures for CS incident handling
in USDA. A Department-wide incident
handling and tracking capability will be supported and maintained by OCIO
CS. Each agency is expected to
establish, support and maintain their own internal policies, procedures or team
to support prompt, effective and efficient resolution of CS incidents in
accordance with the process outlined below.
USDA organizations must acknowledge and respond to all CS incidents in
accordance with the timeframes in the procedures below. A critical component of successful incident
handling is a comprehensive knowledge and inventory of all Internet Protocol
(IP) addresses that were delegated to agencies by Telecommunications Service
Organization (TSO). Each USDA
organization is also expected to control, allocate and maintain accurate
electronic records of all assigned IP addresses as required by DR 3300 and
assist with notification of emergency personnel. OCIO CS has documented its responsibilities
and role to be the POC to US-CERT. OCIO
CS will be responsible for notifying OIG and US-CERT of USDA incidents and
their closure. US-CERT will acknowledge
closure of incidents assigned their tracking number. All USDA organizations will ensure that all
incident procedures are followed and that incident reporting is accomplished by
the ISSPM through OCIO CS for all OCIO CS assigned incidents, even if they have
their own incident response team (IRT).
ISSPMs shall be responsible for certifying the accuracy of incident
reports.
Policy
Exception Requirements – There are no exceptions to
the requirement that all agencies report incidents. However, USDA organizations that cannot comply with this policy are required to
document shortcomings as formal policy exceptions. The CIO of the agency/staff office/mission
area will submit all policy exception requests directly to the ACIO CS. Exceptions to policy will be considered only
in terms of implementation timeframes; exceptions will not be granted to the
requirement to conform to this policy.
USDA organizations cannot wait until CS incidents occur or cannot be
closed to request an exception to policy requirements. Exceptions that are approved will be interim
in nature and will require that each agency report this Granted Policy
Exception (GPE) as a Plan of Action & Milestones (POA&M) in their FISMA
reporting, with a GPE notation, until full compliance is achieved. Interim exceptions expire with each fiscal
year. Compliance exceptions that require
longer durations must be submitted to the USDA CIO for approval and contain a
convincing case for the extension with an updated timeline for completion. Any approved extensions must continue to be
documented in the agency’s annual FISMA report and quarterly POA&Ms. OCIO CS will monitor all approved exceptions.
3 PROCEDURES
An incident
is the act of violating an explicit or implied security policy. The types of activity that are widely
recognized as being CS incidents are violations categorized as, but are not
limited to, attempts (either failed or successful) to gain unauthorized access
to a system or its data, unwanted disruption or denial of service, the
unauthorized use of a system for the processing or storage of data, or changes
to system hardware, firmware or software characteristics without the owner’s
knowledge, instructions, and approval.
The level of consequence of an incident refers to the relative impact it
has on an organization. The types of
impact include: loss of data; the loss
or theft of information, IT resources, revenue or confidence in a USDA agency
or mission area by the general public or customers; or a high level of damage
that must be corrected prior to system restoration.
a In USDA, CS incidents shall be declared for the following
reasons:
(1) Analysis of
intrusion detection system (IDS) reports that are rated as High: Internal, or
High: External, and show system compromises in the logs;
(2) Notification by
US-CERT of a USDA IP or e-mail address being the cause or victim of malicious
or questionable activity;
(3)
Alert, notification, or warning from other U.
S. Government agencies that USDA IP address(s) is the target or originator of
malicious activity;
(4) Notification by the
USDA OIG of a complaint that requires CS investigation or technical support;
(5) Complaints by an
Internet Service Provider (ISP) that detail specific, prohibited activities by
a USDA host, IP address or e-mail address;
(6) Complaints by
organizations and companies that exist to ensure copyright protection. These include the Business Software Alliance
(BSA), Software & Information Industry Association (SIIA), Recording
Industry Association of America (RIAA), The Motion Picture Association of
America (MPAA), and companies that monitor the Internet on behalf of movie,
video, and music copyright holders;
(7) Floods of viruses,
worms and Trojan Horses for which anti-malicious code/anti-virus software is
not available. In attacks such as Code
Red, Nimda, Slammer, and Blaster One, one USDA incident number will be assigned
for the entire process;
(8)
Complaints from the public, or
other employees that include specific examples or references of inappropriate
or illegal use by USDA employees,
cooperators, partners or contractors utilizing USDA IT; and
(9) A self-discovery by
a USDA organization that meets the
definition
of an incident (i.e., virus discoveries, criminal actions, etc.)

Figure 1
b Cyber Security Incidents are to be
declared when they are serious and considered major in nature. They are declared based on the
assessment of the
gravity of the situation, sensitivity of information threatened or compromised
and the potential for harm to USDA.
Outlined below are criteria for the high-level incidents or medium and
low events:
Figure 1
b CS incidents
are to be declared when they are serious and considered major in nature. They are declared based on the
assessment
of the gravity of the situation, sensitivity of information threatened or
compromised and the potential for harm to USDA.
Outlined
below are criteria for the CS incidents (High Level Events) or medium and low
level events:
(1) Cyber Security
(CS) incidents are High Level Events or US-CERT Priority Level 1 and 2
disruptions that are the most serious and considered ‘major’ in nature. Because of the gravity of the situation and
the high potential for harm to USDA, these incidents should be handled
immediately. USDA CS incidents include
events, activities, and violations such as:
possible life threatening activity, compromise of critical systems or
information, root compromise, child pornography, pornographic trafficking,
music/unauthorized software trafficking, any violation of law or agency
specific policies or statute. Any
activities that are not normally reported to US-CERT but are reported to OIG,
Human Resources or law enforcement are defined as CS incidents and will be
assigned an incident tracking number (ITN).
These incidents will be handled using an accelerated and principals
only/limited distribution CS incident response.
If criminal proceedings are initiated, the USDA incident handler may not
have a need-to- know further details.
Agency
ISSPMs who have suspected or confirmed incidents in this category are to
immediately report the severity and coordinate the incident response with the
ACIO for CS or designate. If the
incident remains open for more than 15 days, ACIO CS will send the agency CIO a
one-time notification of open incident(s).
Each USDA organization’s CIO will respond with corrective actions; a
POA&M will also be initiated until incident(s) are closed.
CS incidents include:
Other
types of incidents are categorized as adverse CS events and shall not be
declared CS incidents unless there is a confirmed compromise of sensitive
information, a threat to USDA IT resources or subsequent escalation to a CS
Incident.
(2) Medium level
Cyber Security (CS) events are potentially serious and should be handled
the same day the event occurs or notification of the event is made to USDA
organization (normally in two to four hours of the event). These events can be reported to the agency
ISSPM by OCIO CS (when detected in USDA/OCIO), the helpdesk, system
administrator (SA) or incident handler(s) or incident response team
(CSIRT).
These include:
in which
the Government computer is neither the tool or target of the action;
·
Unconfirmed computer
virus/worms (depending on impact to Agency/Department and if the infection is
the result of a security policy violation); and
·
Undocumented or unapproved
vulnerability scans.
(3) Low level Cyber
Security (CS) events are the least severe and should be investigated within
three working days after the event occurs.
These events can be reported to the agency ISSPM by OCIO CS (when
detected in USDA), the helpdesk, SA or incident handler or incident response
team (IRT).
Low level CS events include:
·
Loss or compromise of
a personal password;
·
Minor misuse of Government
property, facilities and services;
·
US-CERT Priority Level 4
Incident Reporting Guideline events;
·
Unsuccessful scans/probes
(internal & external); and
·
Computer virus/worms
(depending on impact to Agency/Department).
Agencies and staff
offices shall not be required to report actions taken to mitigate adverse
events unless requested or instructed to by ACIO CS.
c Incident Handling Phases – The incident handling process is
comprised
of seven phases that compose an effective response to the overall
incident. These phases are designed to
ensure that no portion of the process is overlooked and consistency in incident
handling is maintained. The steps in
each phase are listed below:

Figure 2
(1)
Incident Prevention – NIST Special Publication 800-61 reminds Federal
agencies that keeping the number of incidents low is important to protect their
business processes, mission and reputation.
If security controls are insufficient or security policies are not
enforced large numbers of incidents can occur with overwhelming consequences
for the agency and USDA as an organization.
In addition, to prevent incidents each agency and staff office must
conduct and keep current risk assessments of systems and applications. These assessments should determine what
risks, if any, the combinations of threats and vulnerabilities pose to those
systems.
Incident
Indications, Alerts & Warnings – OCIO CS will analyze suspected events,
complaints and findings from a variety of sources and notify agencies of these
occurrences. These sources include: the
IDS, US-CERT, other Federal agencies, Federal Trade Commission, OIG, ISP,
internal audit or assessment, and private copyright protection organizations. ACIO CS does not automatically declare those
communications to be incidents. When
OCIO CS and/ or TSO cannot adequately or promptly determine the accuracy of the
indications, alerts and warnings by providing their own findings they will
defer to the USDA organization to make a finding. When USDA organizations do not respond with a
finding within 48 hours, ACIO CS will declare these to be a CS Incident.
(2)
Incident Notification - Incident notification is a multi-stage process. Suspected events, complaints and incidents
can occur anytime during a 24-hour period.
For this reason, USDA has established an Incident Handling Program
Manager in OCIO CS. The Incident
Handling Program Manager will ensure that USDA organizational personnel are
provided with notification of suspected intrusions and receive and document the
suspected incident regardless of the source.
Each USDA organization will ensure that OCIO CS has a current electronic
list of Agency incident contacts in order to ensure that USDA organizations can
be reached promptly to resolve incidents effectively. This list will include the agency ISSPM,
Deputy ISSPM and the CIO. The ISSPM will
be the individual who is responsible for the overall management and resolution
of all suspected incidents in agencies and staff offices.
Each USDA
organization will establish internal IRT to handle incident data, determine the
impact of the incident and act appropriately to limit the damage to the
organization and restore normal services.
In OCIO, there is a coordinating team Led by OCIO CS staff who will act
as the Incident Handling Program Manager.
This coordinating team can elect to activate the “Ad Hoc IRT” from all
areas of USDA, as required, to assist USDA organizations in responding to major
incidents that threaten department resources.
Outside resources often provide objectivity and can be helpful to the
internal team under pressure to resolve the crisis. The primary role of the coordinating team is
to provide guidance and advice to the agency internal IRT without having
authority over the team. The agency
ISSPM or Deputy ISSPM will notify the agency IRT when a suspected incident is
reported by OCIO CS for response and action.
Agencies can respond to these incidents using a team already established
for this purpose or assign individuals based on the action needed in an ad hoc
fashion. However, the designated team
should be part of a centralized response by the agency to ensure that the
process is consistent across the organization and information is shared at all
layers rapidly and effectively.
(3)
Incident
Identification/Declaration – ACIO CS does not
automatically declare findings to be incidents.
However, USDA organizations must respond in 48 hours or ACIO CS will
declare an incident. ACIO CS will need
a finding or status report to prevent their declaration. When ACIO CS declares an incident, a USDA
Incident Tracking Number (ITN) is assigned by which the department tracks and
responds to requests for information concerning the incident. Agency internal IRTs may also assign their
own internal number for tracking purposes.
However, all reports must reference the USDA and US-CERT tracking number
for reporting purposes. ACIO CS is still
the departmental POC for all incidents and is responsible for providing
notifications, status reports and close out recommendations to US-CERT, OIG and
other oversight authorities. In
addition, ACIO CS acts as the POC for notification of the CIO, responds to
requests for status and to Secretarial inquires. OCIO, in coordination with the Office of
Communication (OC), is responsible for all dealings with the media and
public. USDA agencies are to direct
inquiries from these sources to OCIO for response and resolution.
During this
phase an incident or incidents may be cancelled. Cancellation occurs when investigations
determine that no incident occurred, the IDS provided a “false positive”, or
information related to the incident was incorrect. A cancelled incident is the same as a closed
incident.
(4)
Incident Preparation – Each USDA organization will develop their own incident handling
procedures and notification trees.
Documentation and forms should be available at the outset of each formal
incident or event that shall be updated at each stage of the incident and shall
be finalized at incident/event conclusion.
The USDA CIO, through ACIO CS, will be kept abreast of the status of
ongoing major incidents at regular intervals (as events change or progress is
made) by the agency until resolution of the incident.
(5)
Incident Response – This phase includes the analysis of how the incident happened,
how to handle the situation so that it is resolved quickly and to ensure that
it does not reoccur. Each USDA
organization will develop internal response procedures that support the actions
that must be taken in responding to incidents.
At a minimum, the internal procedures will include a reporting chain and
require the involvement of organizational personnel and OCIO CS. These procedures will also require the
preservation of evidence, assessment, containment and recover actions, damage
determination, report documentation, lessons learned and the identification of
corrective actions required by the agency security program managers and CIO.
There are
three definitive sub-phases of this process: assessment and containment, recovery
operations, and damage analysis.
Assessment
and containment – This process begins as soon
as suspicious activity is detected and personnel are designated to take
immediate action to resolve the incident.
The IRT(s) must be empowered to take containment actions up to and
including the immediate shut down of the system to prevent further intrusion or
damage to the agency system or other department networks or resources. The Department CIO also has the authority to
issue a “Cease and Desist” order to bring a system down should the
circumstances dictate or the agency not respond in a expeditious manner to the
incident (normally 12 hours).
Additionally, the department may issue a port or IP address block
internally or externally. This block
will remain in place until the incident is officially closed by OCIO. Reporting through the agency ISSPM to OCIO CS
will occur simultaneously when accurate information is available, particularly
in cases where the preliminary assessment indicates that significant damage to
USDA resources may have occurred.
Unavailability of any official in the organizational reporting chain is
not to delay the continuation of the incident notification or response process.
Recovery
operations - Each USDA organization should
prioritize those actions that support the smooth recovery of a compromised
system(s). In no case should a
compromised system, web page or application be returned to normal operation
without the approval of ACIO CS. The
ISSPM will request that OCIO CS permit the system(s), web page, or application
to resume normal operation. OCIO CS
reserves the right to further scan the system to ensure that appropriate
security is in place to protect the Department. The agency may resume normal operation of
the restored system, upon ACIO CS approval and the completion of the IT
incident report. The ACIO CS will have 1
working day to respond with the approval or disapproval to return the system to
normal operation. If a system is mission
critical, the USDA organization can coordinate directly with the ACIO CS for a
more immediate system restoration, on a case-by-case basis. If the USDA organization does not receive a
response within that time, they can return the system to normal operation
provided that they feel adequate security protection is in place to prevent
future incidents.
Damage
analysis - An analysis of all CS incidents is
to be initiated immediately after assessment, containment and recovery actions
are completed by each agency ISSPM. The
ISSPM will determine if the incident is confined to one agency or multiple
agencies and if there is impact to organizations outside USDA. The impact to each system will be analyzed to
determine if the control of the system has been compromised. All compromised systems will be disconnected
from external communications as soon as possible, but not later than 12 hours
from discovery of the incident. Control
of a system is lost when the intruder obtains control of the root or system
accounts with administrative privileges.
A determination is to be made if log files have been erased or
compromised.
The
ISSPM will initiate the process of estimating the overall economic impact of
the incident to the USDA organization and Department in coordination with the
system owner/business manager. At a
minimum, the estimate will be quantified in terms of loss of system(s)
availability, loss of response capability to customers, cost of
equipment/software to repair, and hours of personnel associated with the repair
or restoration of the system(s). The
damage assessment report will be reviewed and concurred on by the system
owner/business manager prior to inclusion in the CS Incident report. This information will then be updated in the
CS Incident report.
(6)
Incident Reporting – involves formal documentation that a CS Incident occurred using
the departmental formal reporting process established in this policy. All USDA completed incident report
documentation is to be reported to OCIO CS.
OCIO CS is responsible for incident reporting to the OIG, US-CERT, and
law enforcement for any violation of law.
CS incidents are to be tracked and closed in accordance with the
requirements of this policy. However, CS
incidents that involve violations of the law or investigation will be
separately tracked as resolution may not occur for a protracted period of time.