Skip to content
ATO Pathways
Log In
Overview
Search
Catalogs
SCAP
OSCAL
Catalogs
Profiles
Resources
Documents
Publishers
References
Knowledge Base
Platform Documentation
Compliance Dictionary
Platform Changelog
About
Catalogs
XCCDF
Network Infrastructure Policy Security Technical Implementation Guide
Network Infrastructure Policy Security Technical Implementation Guide
An XCCDF Benchmark
Details
Profiles
Items
Prose
File Metadata
67 rules organized in 67 groups
NET0140
1 Rule
The connection between the Channel Service Unit/Data Service Unit (CSU/DSU) and the Local Exchange Carriers (LEC) data service jack (i.e., demarc) as well as any service provider premise equipment must be located in a secure environment.
Low Severity
DOD leased lines carry an aggregate of sensitive and non-sensitive data; therefore unauthorized access must be restricted. Inadequate cable protection can lead to damage and denial of service attacks against the site and the LAN infrastructure.
NET-IDPS-016
1 Rule
An Intrusion Detection and Prevention System (IDPS) sensor must be deployed to monitor all Demilitarized Zone (DMZ) segments housing public servers.
Medium Severity
The initial step in IDPS deployment is determining where sensors should be placed. Because attacks originate at the enclave perimeter and within the enclave boundary an IDPS implementation at the enclave perimeter only will not suffice. By placing IDPS technology throughout the Enterprise Regional enclaves and stand-alone enclaves, system administrators can track the spread of attacks and take corrective actions to prevent attacks reaching critical resources.
NET-IDPS-018
1 Rule
An Intrusion Detection and Prevention System (IDPS) sensor must be deployed to monitor the network segment hosting web, application, and database servers.
Medium Severity
Attacks can originate within the enclave boundary. Hence, deploying an IDPS on the network segment hosting web, application, and database servers is imperative. The servers are critical resource and the network segment hosting them will receive the most traffic within the enclave. Deploying IDPS on this network is promotes defense-in-depth principles that will enable operations to detect attacks quickly and take corrective actions.
NET-IDPS-019
1 Rule
An Intrusion Detection and Prevention System (IDPS) sensor must be deployed to monitor network segments that house network security management servers.
Medium Severity
The initial step in IDPS deployment is determining where sensors should be placed. Because attacks originate at the enclave perimeter and within the enclave boundary an IDPS implementation at the enclave perimeter only will not suffice. By placing IDPS technology throughout the Enterprise Regional enclaves and stand-alone enclaves, system administrators can track the spread of attacks and take corrective actions to prevent attacks reaching critical resources.
NET-IDPS-021
1 Rule
An Intrusion Detection and Prevention System (IDPS) must be deployed to monitor all unencrypted traffic entering and leaving the enclave.
Medium Severity
Per CJCSI 6510.01F, Enclosure A-5, Paragraph 8, "DOD ISs (e.g., enclaves, applications, outsourced IT-based process, and platform IT interconnections) shall be monitored to detect and react to incidents, intrusions, disruption of services, or other unauthorized activities (including insider threat) that threaten the security of DOD operations or IT resources, including internal misuse." An Intrusion Prevention System (IPS) allows the sensor to monitor, alert, and actively attempt to drop/block malicious traffic. An Intrusion Detection System (IDS) uses a passive method; receiving a copy of the packets to analyze and alert authorized persons about any malicious activity. While an IDS or an IPS in a passive role cannot stop the attack itself, it can typically notify and dynamically assign ACLs or other rules to a firewall or router for filtering. The preferred method of installation is to have the IDPS configured for inline mode. Only when there is a valid technical reason, should the IDPS be placed into a passive or IDS mode. For a full uninhibited view of the traffic, the IDPS must sit behind the enclave's firewall. This will allow the IDPS to monitor all traffic unencrypted, entering or leaving the enclave.
NET-IDPS-024
1 Rule
Sensor traffic in transit must be protected at all times via an Out-of-Band (OOB) network or an encrypted tunnel between site locations.
Medium Severity
User interface services must be physically or logically separated from data storage and management services. Data from IDS sensors must be protected by confidentiality controls; from being lost and altered.
NET-IDPS-025
1 Rule
Intrusion Detection and Prevention System (IDPS) traffic between the sensor and the security management or sensor data collection servers must traverse a dedicated Virtual Local Area Network (VLAN) logically separating IDPS traffic from all other enclave traffic.
Medium Severity
All IDPS data collected by agents in the enclave at required locations must also be protected by logical separation when in transit from the agent to the management or database servers located on the Network Management subnet.
NET-IDPS-027
1 Rule
Products collecting baselines for anomaly-based detection must have their baselines rebuilt based on changes to mission requirements such as Information Operations Conditions (INFOCON) levels and when the traffic patterns are expected to change significantly.
Low Severity
Administrators should ensure that any products collecting baselines for anomaly-based detection have their baselines rebuilt periodically as needed to support accurate detection. The ISSM is required to have the enclave prepared for readiness by raising INFOCON levels prior to an activity to ensure the network is as ready as possible when the operation or exercise begins. Because system and network administrators implement many of the INFOCON measures over a period of time in a pre-determined operational rhythm, commanders should raise INFOCON levels early enough to ensure completion of at least one cycle before the operational activity begins. Recommendations for possible INFOCON changes should be written into Operation Plans (OPLAN) and Concept Plans (CONPLAN). Guidelines can be found in Strategic Command Directive (SD) 527-1.
NET-IDPS-029
1 Rule
If a Secure File Transfer Protocol (SFTP) server is used to provide updates to the sensors, the server must be configured to allow read-only access to the files within the directory on which the signature packs are placed.
Medium Severity
In a large scale IDPS deployment, it is common to have an automated update process implemented. This is accomplished by having the updates downloaded on a dedicated SFTP server within the management network. The SFTP server should be configured to allow read-only access to the files within the directory on which the signature packs are placed, and then only from the account that the sensors will use. The sensors can then be configured to automatically check the SFTP server periodically to look for the new signature packs and to update themselves once they have been tested.
NET-IDPS-030
1 Rule
If an automated scheduler is used to provide updates to the sensors, an account on the file server must be defined that will provide access to the signatures only to the sensors.
Medium Severity
In a large scale IDPS deployment, it is common to have an automated update process implemented. This is accomplished by having the updates downloaded on a dedicated secure file server within the management network. The file server should be configured to allow read-only access to the files within the directory on which the signature packs are placed, and then only from the account that the sensors will use. The sensors can then be configured to automatically check the secure file server periodically to look for the new signature packs and to update themselves.
NET-IDPS-031
1 Rule
The Intrusion Detection and Prevention System (IDPS) configuration must be backed up before applying software or signature updates, or when making changes to the configuration.
Low Severity
There are two types of IDPS updates: software updates and signature updates. Software updates fix bugs in the IDPS software or add new functionality, while signature updates add new detection capabilities or refine existing detection capabilities (e.g., reducing false positives). For many IDPSs, signature updates cause program code to be altered or replaced, so they are really a specialized form of software update. For other IDPSs, signatures are not written in code, so a signature update is a change to the configuration data for the IDPS. Software updates can include any or all IDPS components, including sensors, agents, management servers, and consoles. Software updates for sensors and management servers, particularly appliance-based devices, are often applied by replacing an existing IDPS CD with a new one and rebooting the device. Many IDPSs run the software directly from the CD, so that no software installation is required. Other components, such as agents, require an administrator to install software or apply patches, either manually on each host or automatically through IDPS management software. Some vendors make software and signature updates available for download from their Web sites or other servers; often, the administrator interfaces for IDPSs have features for downloading and installing such updates. Administrators should verify the integrity of updates before applying them, because updates could have been inadvertently or intentionally altered or replaced. The recommended verification method depends on the update's format, as follows: Files downloaded from a Web site or FTP site. Administrators should compare file checksums provided by the vendor with checksums that they compute for the downloaded files. Update downloaded automatically through the IDPS user interface. If an update is downloaded as a single file or a set of files, either checksums provided by the vendor should be compared to checksums generated by the administrator, or the IDPS user interface itself should perform some sort of integrity check. In some cases, updates might be downloaded and installed as one action, precluding checksum verification; the IDPS user interface should check each update's integrity as part of this. Removable media (e.g., CD, DVD). Vendors may not provide a specific method for customers to verify the legitimacy of removable media apparently sent by the vendors. If media verification is a concern, administrators should contact their vendors to determine how the media can be verified, such as comparing vendor-provided checksums to checksums computed for files on the media, or verifying digital signatures on the media's contents to ensure they are valid. Administrators should also consider scanning the media for malware, with the caveat that false positives might be triggered by IDPS signatures for malware on the media.
NET-IDPS-032
1 Rule
The Intrusion Detection and Prevention System (IDPS) file checksums provided by the vendor must be compared and verified with checksums computed from CD or downloaded files.
Low Severity
There are two types of IDPS updates: software updates and signature updates. Software updates fix bugs in the IDPS software or add new functionality, while signature updates add new detection capabilities or refine existing detection capabilities (e.g., reducing false positives). For many IDPSs, signature updates cause program code to be altered or replaced, so they are really a specialized form of software update. For other IDPSs, signatures are not written in code, so a signature update is a change to the configuration data for the IDPS. Software updates can include any or all IDPS components, including sensors, agents, management servers, and consoles. Software updates for sensors and management servers, particularly appliance-based devices, are often applied by replacing an existing IDPS CD with a new one and rebooting the device. Many IDPSs run the software directly from the CD, so that no software installation is required. Other components, such as agents, require an administrator to install software or apply patches, either manually on each host or automatically through IDPS management software. Some vendors make software and signature updates available for download from their Web sites or other servers; often, the administrator interfaces for IDPSs have features for downloading and installing such updates. Administrators should verify the integrity of updates before applying them, because updates could have been inadvertently or intentionally altered or replaced. The recommended verification method depends on the update's format, as follows: Files downloaded from a Web site or FTP site. Administrators should compare file checksums provided by the vendor with checksums that they compute for the downloaded files. Update downloaded automatically through the IDPS user interface. If an update is downloaded as a single file or a set of files, either checksums provided by the vendor should be compared to checksums generated by the administrator, or the IDPS user interface itself should perform some sort of integrity check. In some cases, updates might be downloaded and installed as one action, precluding checksum verification; the IDPS user interface should check each update's integrity as part of this. Removable media (e.g., CD, DVD). Vendors may not provide a specific method for customers to verify the legitimacy of removable media apparently sent by the vendors. If media verification is a concern, administrators should contact their vendors to determine how the media can be verified, such as comparing vendor-provided checksums to checksums computed for files on the media, or verifying digital signatures on the media's contents to ensure they are valid. Administrators should also consider scanning the media for malware, with the caveat that false positives might be triggered by IDPS signatures for malware on the media.
NET-IDPS-033
1 Rule
The organization must establish weekly data backup procedures for the network Intrusion Detection and Prevention System (IDPS) data.
Medium Severity
IDPS data needs to be backed up to ensure preservation in the case a loss of data due to hardware failure or malicious activity.
NET-IDPS-035
1 Rule
The Intrusion Detection and Prevention System (IDPS) software and signatures must be updated when updates are provided by the vendor.
Low Severity
Keeping the IDPS software updated with the latest engine and attack signatures will allow for the IDPS to detect all forms of known attacks. Not maintaining the IDPS properly could allow for attacks to go unnoticed.
NET-TUNL-026
1 Rule
Encapsulated and/or encrypted traffic received from another enclave must not bypass the network perimeter defense without being terminated and inspected before entering the enclaves private network.
High Severity
Allowing encapsulated traffic to bypass the enclave's network perimeter without being filtered and inspected leaves the enclave vulnerable to malicious traffic that could result in compromise and denial of service. The destination of these packets could be servers that provide mission critical services and data.
NET-TUNL-028
1 Rule
Tunneling of classified traffic across an unclassified IP transport network or service provider backbone must be documented in the enclaves security authorization package and an Approval to Connect (ATC), or an Interim ATC must be issued by DISA prior to implementation.
High Severity
CJCSI 6211.02D instruction establishes policy and responsibilities for the connection of any information systems to the Defense Information Systems Network (DISN) provided transport. Enclosure E mandates that the CC/S/A document all IP tunnels transporting classified communication traffic in the enclave's security authorization package prior to implementation. An ATC or IATC amending the current connection approval must be in place prior to implementation. Enclosure D of the CJCSI 6211.02D also provides guidance on the requirements of tunneling classified data (section 15.a), which helps a CC/S/A determine applicability to their mission. Items include but are not limited to: - minimize tunneling of classified data over transport other than DISN provided transport (i.e., SIPRNET); - ensure the Authorizing Official (DAA) validates all requirements to tunnel classified information across unclassified IP infrastructure; - obtain DSAWG approval before tunneling classified data across unclassified IP infrastructure; - ensure transmission of classified information is secured through use of authorized cryptographic equipment and algorithms and/or PDSs; - document IP tunnels transporting classified communication traffic in the enclave’s security authorization package prior to implementation; - an ATC or IATC amending the current connection approval must be in place prior to implementation.
NET-TUNL-030
1 Rule
DSAWG approval must be obtained before tunneling classified traffic outside the components local area network boundaries across a non-DISN or OCONUS DISN unclassified IP wide area network transport infrastructure.
High Severity
CJCSI 6211.02D instruction establishes policy and responsibilities for the connection of any information systems to the Defense Information Systems Network (DISN) provided transport. Enclosure E mandates that the CC/S/A obtain DSAWG approval before tunneling classified data outside component's local area network boundaries across a non-DISN or OCONUS DISN unclassified IP-wide area transport infrastructure.
NET-TUNL-031
1 Rule
Tunneling of classified traffic across an unclassified IP transport network must employ cryptographic algorithms in accordance with CNSS Policy No. 15.
High Severity
When transporting classified data over an unclassified IP network, it is imperative that traffic from the classified enclave or community of interest is encrypted prior reaching the point of presence or service delivery node of the unclassified network. Confidentiality and integrity of the classified traffic must be preserved by employing cryptographic algorithms in accordance with CNSS Policy No. 15 which requires the appropriate Suite B cryptographic algorithms listed in ANNEX B or a commensurate suite of NSA-approved cryptographic algorithms.
NET-VLAN-001
1 Rule
The organization must ensure all switches and associated cross-connect hardware are kept in a secure Intermediate Distribution Frame (IDF) or an enclosed cabinet that is kept locked.
Medium Severity
Since the IDF includes all hardware required to connect horizontal wiring to the backbone, it is imperative that all switches and associated cross-connect hardware are kept in a secured IDF or an enclosed cabinet that is kept locked. This will also prevent an attacker from gaining privilege mode access to the switch. Several switch products only require a reboot of the switch in order to reset or recover the password.
NET0090
1 Rule
Network topology diagrams for the enclave must be maintained and up to date at all times.
Medium Severity
To assist in the management, auditing, and security of the network infrastructure facility drawings and topology maps are a necessity. Topology maps are important because they show the overall layout of the network infrastructure and where devices are physically located. They also show the relationship and interconnectivity between devices and where possible intrusive attacks could take place. Having up to date network topology diagrams will also help show what the security, traffic, and physical impact of adding a new user(s) will be on the network.
NET0130
1 Rule
All external connections must be validated and approved by the Authorizing Official (AO) and the Connection Approval Office (CAO) and meeting Connection Approval Process (CAP) requirements.
Medium Severity
Every site must have a security policy to address filtering of the traffic to and from those connections. This documentation along with diagrams of the network topology is required to be submitted to the Connection Approval Process (CAP) for approval to connect to the NIPRNet or SIPRNet. SIPRNet connections must also comply with the documentation required by the Classified Connection Approval Office (CCAO) to receive the SIPRNet Interim Approval to Connect (IATC) or final Approval to Connect (ATC). Also any additional requirements must be met as outlined in the Interim Authority to Operate (IATO) or Authority to Operate (ATO) forms signed by the Authorizing Official (AO).
NET0131
1 Rule
Prior to having external connection provisioned between enclaves, a Memorandum of Agreement (MOA) or Memorandum of Understanding (MOU) must be established.
Medium Severity
Prior to establishing a connection with another activity, a Memorandum of Understanding (MOU) or Memorandum of Agreement (MOA) must be established between the two sites prior to connecting with each other.
NET0135
1 Rule
External connections to the network must be reviewed and the documentation updated semi-annually.
Medium Severity
A network is only as secure as its weakest link. It is imperative that all external connections be reviewed and kept to a minimum needed for operations. All external connections should be treated as untrusted networks. Reviewing who or what the network is connected to empowers the security manager to make sound judgements and security recommendations. Minimizing backdoor circuits and connections reduces the risk for unauthorized access to network resources.
NET0168
1 Rule
If the site has a non-DoD external connection (i.e. Approved Gateway), an Intrusion Detection and Prevention System (IDPS) must be located between the sites Approved Gateway and the perimeter router.
Medium Severity
The incorrect placement of the external IDPS may allow unauthorized access to go undetected and limit the ability of security personnel to stop malicious or unauthorized use of the network. In order to ensure that an attempted or existing attack goes unnoticed, the data from the sensors must be monitored continuously.
NET0170
1 Rule
External network connections must not bypass the enclaves perimeter security.
Medium Severity
Without taking the proper safeguards, external networks connected to the organization will impose security risks unless properly routed through the perimeter security devices. Since external networks to the organization are considered to be untrusted, this could prove detrimental since there is no way to verify traffic inbound or outbound on this backdoor connection. An attacker could carry out attacks or steal data from the organization without any notification. An external connection is considered to be any link from the organization's perimeter to the NIPRNet, SIPRNet, Commercial ISP, or other untrusted network outside the organization's defined security policy. The DREN and SREN are DoD's Research & Engineering Network. A DoD Network that is the official DoD long-haul network for computational scientific research, engineering, and testing in support of DoD's S&T and T&E communities. It has also been designated as a DoD IPv6 pilot network by the Assistant Secretary of Defense (Networks & Information Integration)/DoD Chief Information Officer ASD (NII)/DoD CIO. A DISN enclave should not have connectivity to the DREN unless approved by the AO and the requirements have been met for all external connections described in NET0130.
NET0180
1 Rule
All global address ranges used on unclassified and classified networks must be properly registered with the DoD Network Information Center (NIC).
Medium Severity
If network address space is not properly configured, managed, and controlled, the network could be accessed by unauthorized personnel resulting in security compromise of site information and resources. Allowing subscribers onto the network whose IP addresses are not registered with the .Mil NIC may allow unauthorized users access into the network. These unauthorized users could then monitor the network, steal passwords, and access classified information.
NET0185
1 Rule
Network Address Translation (NAT) and private IP address space must not be deployed within the SIPRNet enclave.
Medium Severity
The DoD has an enterprise level security-focused configuration management (SecCM) requirement to support end-to-end monitoring of SIPRNet, as a National Security System (NSS). The use of NAT and private IP address space inhibits the view of specialized DISN enterprise tools in tracking client level enclave to enclave traffic, monitoring client use of enterprise level application services, and detecting anomalies and potential malicious attacks in SIPRNet client application traffic flows. Enclave nodes that communicate outside the organization's enclave to other SIPRNet enclaves or enterprise services cannot use NATd private addresses via an enclave proxy without the permission of the SIPRNet DISN Authorizing Official, the DISA AO.
NET0198
1 Rule
Dynamic Host Configuration Protocol (DHCP) audit and event logs must record sufficient forensic data to be stored online for thirty days and offline for one year.
Medium Severity
In order to identify and combat IP address spoofing, it is highly recommended that the DHCP server logs MAC addresses and hostnames on the DHCP server, in addition to standard data such as IP address and date/time.
NET0199
1 Rule
Dynamic Host Configuration Protocol (DHCP) servers used within SIPRNet infrastructure must be configured with a minimum lease duration time of 30 days.
Low Severity
In order to trace, audit, and investigate suspicious activity, DHCP servers within the SIPRNet infrastructure must have the minimum lease duration time configured to 30 or more days.
NET0210
1 Rule
All network infrastructure devices must be located in a secure room with limited access.
Medium Severity
If all communications devices are not installed within controlled access areas, risk of unauthorized access and equipment failure exists, which could result in denial of service or security compromise. It is not sufficient to limit access to only the outside world or non-site personnel. Not everyone within the site has the need-to-know or the need-for-access to communication devices.
NET0346
1 Rule
All hosted NIPRNet-only applications must be located in a local enclave Demilitarized Zone (DMZ).
Medium Severity
Without the protection of a DMZ, production networks will be prone to outside attacks as they are allowing externally accessible services to be accessed on the internal LAN. This can cause many undesired consequences such as access to the entire network, Denial of Service attacks, or theft of sensitive information.
NET0348
1 Rule
All Internet-facing applications must be hosted in a DoD Demilitarized Zone (DMZ) Extension.
Medium Severity
Without the protection of a DMZ, production networks will be prone to outside attacks as they are allowing externally accessible services to be accessed on the internal LAN. This can cause many undesired consequences such as access to the entire network, Denial of Service attacks, or theft of sensitive information.
NET0351
1 Rule
When protecting the boundaries of a network, the firewall must be placed between the private network and the perimeter router and the Demilitarized Zone (DMZ).
Medium Severity
The only way to mediate the flow of traffic between the inside network, the outside connection, and the DMZ is to place the firewall into the architecture in a manner that allows the firewall the ability to screen content for all three destinations.
NET0365
1 Rule
The organization must implement a deep packet inspection solution when protecting perimeter boundaries.
High Severity
Deep packet inspection (DPI) examines the packet beyond the Layer 4 header by examining the payload to identify the application or service. DPI searches for illegal statements, predefined criteria, malformed packets, and malicious code, thereby enabling the IA appliances to make a more informed decision on whether to allow or not allow the packet through. DPI engines can delve into application centric information to allow different applications to be protected in different ways from different threats. Examples of DPI appliances include next-generation firewalls, application layer gateways as well as specific gateways for web, email and SSL traffic.
NET0369
1 Rule
A deny-by-default security posture must be implemented for traffic entering and leaving the enclave.
High Severity
To prevent malicious or accidental leakage of traffic, organizations must implement a deny-by-default security posture at the network perimeter. Such rulesets prevent many malicious exploits or accidental leakage by restricting the traffic to only known sources and only those ports, protocols, or services that are permitted and operationally necessary. Applications, protocols, TCP/UDP ports, and endpoints (specific hosts or networks) are identified and used to develop rulesets and access control lists to restrict traffic to and from an enclave.
NET0445
1 Rule
Two-factor authentication must be implemented to restrict access to all network elements.
Medium Severity
Without secure management implemented with authenticated access controls, strong two-factor authentication, encryption of the management session and audit logs, unauthorized users may gain access to network managed devices compromised, large parts of the network could be incapacitated with only a few commands.
NET0810
1 Rule
Two Network Time Protocol (NTP) servers must be deployed in the management network.
Low Severity
NTP provides an efficient and scalable method for managed network elements to actively synchronize to an accurate time source. Insuring that there are always NTP servers available to provide time is critical. It is imperative that all single points of failure for the NTP infrastructure are eliminated. Knowing the correct time is not only crucial for proper network functioning but also for security. Compromising an NTP server opens the door to more sophisticated attacks that include NTP poisoning, replay attacks, and denial of service. Where possible, deploy multiple gateways with diverse paths to the NTP servers. An alternative design is to have one server connected to a reference clock and the other server reference an external stratum-1 server. With this scenario, the NTP clients should be configured to prefer the stratum-1 server over the stratum-2 server. The NTP servers should be configured to easily scale by creating a hierarchy of lower level (stratum-2 to stratum-15) servers to accommodate the workload. The width and depth of the hierarchy is dependent on the number of NTP clients as well as the amount of redundancy that is required.
NET0928
1 Rule
A policy must be implemented to keep Bogon/Martian rulesets up to date.
Medium Severity
A Bogon route or Martian address is a type of packet that should never be routed inbound through the perimeter device. Bogon routes and Martian addresses are commonly found as the source addresses of DDoS attacks. By not having a policy implemented to keep these addresses up to date, the enclave will run the risk of allowing illegitimate traffic into the enclave or even blocking legitimate traffic. Also, if there are rulesets with "any" as the source address then Bogons/Martians must be applied. Bogons and Martian addresses can be kept up to date routinely checking the IANA website or creating an account with Team Cymru to retrieve these lists in one of many ways. http://www.iana.org/assignments/ipv4-address-space/ipv4-address-space.xml http://www.team-cymru.org/Services/Bogons/
NET0998
1 Rule
A dedicated management network must be implemented.
Medium Severity
To deploy a management network for the purpose of controlling, monitoring, and restricting management traffic, a separate management subnet must be implemented. Define a large enough address block that will enable the management network to scale in proportion to the managed network.
NET1025
1 Rule
A minimum of two syslog servers must be deployed in the management network.
Low Severity
Maintaining an audit trail of system activity logs can help identify configuration errors, understand past intrusions, troubleshoot service disruptions, and react to probes and scans of the network.
NET1026
1 Rule
Syslog messages must be retained for a minimum of 30 days online and then stored offline for one year.
Low Severity
Logging is a critical part of router security. Maintaining an audit trail of system activity logs (syslog) can help identify configuration errors, understand past intrusions, troubleshoot service disruptions, and react to probes and scans of the network.
NET1040
1 Rule
Current and previous network element configurations must be stored in a secured location.
Low Severity
If the network element's non-volatile memory is lost without a recent configuration stored in an offline location, it may take time to recover that segment of the network. Users connected directly to the switch or router will be without service for a longer than acceptable time.
NET1050
1 Rule
The organization must encrypt all network device configurations while stored offline.
Medium Severity
If a network device's non-volatile memory is lost without a recent configuration stored in an offline location, it may take time to recover that segment of the network. Users connected directly to the switch or router will be without service for a longer than acceptable time. Encrypting the configuration stored offline protects the data at rest and provides additional security to prevent tampering and potentially cause a network outage if the configuration were to be put into service.
NET1622
1 Rule
An Out-of-Band (OOB) management network must be deployed or 24x7 personnel must have console access for device management.
Medium Severity
From an architectural point of view, providing Out-Of-Band (OOB) management of network systems is the best first step in any management strategy. No production traffic resides on an out-of-band network. The biggest advantage to implementation of an OOB network is providing support and maintenance to the network that has become degraded or compromised. During an outage or degradation period the in band management link may not be available. The consequences of loss of availability is unacceptable and could include the immediate and sustained loss of mission effectiveness. Maintenance support for key IT assets must be available to respond 24x7 immediately upon failure.
NET1815
1 Rule
All Releasable Local Area Network (REL LAN) environments must be documented in the System Security Authorization Agreement (SSAA).
Medium Severity
The ISSM will ensure Releasable Local Area Network (REL LAN) environments are documented in the SSAA.
NET1816
1 Rule
Annual reviews must be performed on all Releasable Local Area Network (REL LAN) environments.
Medium Severity
The ISSM will ensure Releasable Local Area Network (REL LAN) reviews are performed annually.
NET1826
1 Rule
Enabling a connection that extends DISN IP network connectivity (e.g., NIPRNet and SIPRNet) to any DoD Vendor, Foreign, or Federal Mission Partner enclave or network without a signed DoD CIO approved sponsorship memo is prohibited. For classified connectivity it must be to a DSS approved contractor facility or DoD Component approved foreign government facility.
High Severity
Having a circuit provisioned that connects the SIPRNet enclave to a non-DoD, foreign, or contractor network puts the enclave and the entire SIPRNet at risk. If the termination point is not operated by the government, there is no control to ensure that the network element at the remote facility is not compromised or connected to another network.
NET1827
1 Rule
Command and Control (C2) and non-C2 exceptions of SIPRNet must be documented in the enclaves accreditation package and an Authority to Connect (ATC) or Interim ATC amending the connection approval received prior to implementation.
Medium Severity
Any exception to use SIPRNet must be documented in an update to the enclave's accreditation package and an Authority to Connect (ATC) or Interim ATC amending the connection approval received prior to implementation.
NET1832
1 Rule
VPN gateways used to create IP tunnels to transport classified traffic across an unclassified IP network must comply with appropriate physical security protection standards for processing classified information.
Medium Severity
When transporting classified data over an unclassified IP network, it is imperative that the network elements deployed to provision the encrypted tunnels are located in a facility authorized to process the data at the proper classification level.
NET2000
1 Rule
Multi-Protocol Labeled Switching (MPLS) protocols deployed to build Label-Switch Path (LSP) tunnels must authenticate all messages with a hash function using the most secured cryptographic algorithm available.
Medium Severity
Spoofed TCP segments could be introduced into the connection streams for LDP sessions used to build LSPs. By configuring strict authentication between LSR peers, LDP TCP sessions can be restricted and the integrity of LSPs can be guarded using the TCP MD5 Signature Option. The LSR ignores LDP Hellos from any LSR for which a password has not been configured. This ensures that the LSR establishes LDP TCP connections only with LSRs for which the shared secret has been configured. RSVP messages are used to control resource reservations for MPLS TE tunnels inside the MPLS core. The RSVP message authentication permits neighbors to use a secure hash to digitally sign all RSVP signaling messages, thus allowing the receiver of an RSVP message to verify the sender. By protecting against corruption and spoofing of RSVP messages, the integrity of the LSPs for bandwidth provisioning, path setup, and path teardown is maintained.
NET2001
1 Rule
Multi-Protocol Labeled Switching (MPLS) labels must not be exchanged between the enclaves edge routers and any external neighbor routers.
Medium Severity
MPLS label exchange via Label Distribution Protocol (LDP) or Resource Reservation Protocol (RSVP) with any external neighbor creates the risk of label spoofing that could disrupt optimum routing, or even drop packets that are encapsulated with a label that is not in the MPLS forwarding table.
NET2002
1 Rule
Label Distribution Protocol (LDP) must be synchronized with the Interior Gateway Protocol (IGP) to minimize packet loss when an IGP adjacency is established prior to LDP peers completing label exchange.
Low Severity
Packet loss can occur when an IGP adjacency is established and the router begins forwarding packets using the new adjacency before the LDP label exchange completes between the peers on that link. Packet loss can also occur if an LDP session closes and the router continues to forward traffic using the link associated with the LDP peer rather than an alternate pathway with a fully synchronized LDP session. The MPLS LDP-IGP Synchronization feature provides a means to synchronize LDP with OSPF or IS-IS to minimize MPLS packet loss. When an IGP adjacency is established on a link but LDP-IGP synchronization is not yet achieved or is lost, the IGP will advertise the max-metric on that link.
NET2004
1 Rule
Rapid Spanning Tree Protocol (STP) must be implemented at the access and distribution layers where Virtual Local Area Networks (VLANs) span multiple switches.
Low Severity
Spanning Tree Protocol (STP) is implemented on bridges and switches to prevent Layer 2 loops when a broadcast domain spans multiple bridges and switches and when redundant links are provisioned to provide high availability in case of link failures. Convergence time can be significantly reduced using Rapid STP (802.1w) instead of STP (802.1d), resulting in improved availability. Rapid STP should be deployed by implementing either Rapid Per-VLAN-Spanning-Tree (Rapid-PVST) or Multiple Spanning-Tree Protocol (MSTP), the later scales much better when there are many VLANs.
NET2005
1 Rule
A Quality of Service (QoS) policy must be implemented to provide preferred treatment for Command and Control (C2) real-time services and control plane traffic.
Low Severity
Different applications have unique requirements and toleration levels for delay, jitter, packet loss, and availability. To manage the multitude of applications and services, a network requires a Quality of Service (QoS) framework to differentiate traffic and provide a method to manage network congestion. The Differentiated Services Model (DiffServ) is based on per-hop behavior by categorizing traffic into different classes and enabling each node to enforce a forwarding treatment to each packet as dictated by a service policy. Packet markings such as IP Precedence and its successor, Differentiated Services Code Points (DSCP), were defined along with specific per-hop behaviors for key traffic types to enable a scalable QoS solution. DiffServ QoS categorizes network traffic, prioritizes it according to its relative importance, and provides priority treatment based on the classification. It is imperative that end-to-end QoS is implemented to guarantee the required bandwidth for control plane traffic and C2 real-time services during periods of congestion within the JIE WAN IP network.
NET2006
1 Rule
Protocol Independent Multicast (PIM) must be disabled on all router interfaces that are not required to support multicast routing.
Medium Severity
PIM is a routing protocol that is used by the IP core for forwarding multicast traffic. PIM operates independent of any particular IP routing protocol but makes use of the IP unicast routing table--PIM does not keep a separate multicast routing table. The multicast tree is built by first allowing a flood of traffic from the source to every dense mode router in the network. For a brief time, unnecessary traffic is allowed. As each router receives traffic for the group, it will decide whether it has active recipients wanting to receive the multicast data. If so, the router will let the flow continue. If no hosts have registered for the multicast group, the router sends a prune message to its neighbor toward the source. That branch of the tree is then pruned off so that the unnecessary traffic does not continue. Dense mode is viewed as a "flood and prune" implementation. With PIM Sparse Mode (PIM-SM), the multicast tree is not extended to a router unless a local host has already joined the group. The multicast tree is built by beginning with group members at the end leaf nodes and extending back toward a central root point--the tree is built from the bottom up. In either case, if an interface is not going to be supporting any of the multicast traffic--that is, join a multicast tree, PIM should be disabled.
NET2007
1 Rule
A Protocol Independent Multicast (PIM) neighbor filter must be implemented to restrict and control multicast traffic.
Low Severity
Protocol Independent Multicast (PIM) is a routing protocol that is used by the IP core for forwarding multicast traffic. PIM traffic must be limited to only known PIM neighbors by configuring and binding a PIM neighbor filter to those interfaces that have PIM enabled.
NET2008
1 Rule
The multicast domain must block inbound and outbound administratively-scoped multicast traffic at the edge.
Low Severity
A multicast boundary must be established to ensure that administratively-scoped multicast traffic does not flow into or out of the IP core. The multicast boundary can be created by ensuring that COI-facing interfaces on all PIM routers are configured to block inbound and outbound administratively-scoped multicast traffic.
NET2009
1 Rule
The multicast domain must block inbound and outbound Auto-RP discovery and announcement messages at the edge.
Low Severity
With static RP, the RP address for any multicast group must be consistent across all routers in a multicast domain. A static configuration is simple and convenient. However, if the statically defined RP router becomes unreachable, there is no automatic failover to another RP router. Auto-RP distributes information to routers as to which RP address must be used for various multicast groups. Auto-RP eliminates inconsistencies and enables scalability and automatic failover. All PIM-enabled routers join the RP discovery group (224.0.1.40), which allows them to receive all group-to-RP mapping information. This information is distributed by an entity called RP mapping agent. Mapping agents themselves join the RP announce group (224.0.1.39). All candidate RPs advertise themselves periodically using the RP announce group address. The mapping agent listens to all RP candidate announcements and determines which routers will be used for each multicast group. It then advertises the RP and its associate multicast groups to all PIM routers in the network using an RP discovery message. Auto-RP announcement and discovery messages provide information (i.e., IP addresses of the RP candidates, multicast groups, etc.) vital to the multicast domain and should not be leaked out of the multicast domain. Using this information, a malicious user could disrupt multicast services by attacking the RP or flooding bogus traffic destined to the learned multicast groups.
NET2010
1 Rule
Protocol Independent Multicast (PIM) register messages received from a downstream multicast Designated Routers (DR) must be filtered for any reserved or any other undesirable multicast groups.
Low Severity
Customer networks that do not maintain a multicast domain and only require the IP multicast service will be required to stand up a PIM-SM router that will be incorporated into the JIE shared tree structure by establishing a peering session with an RP router. Both of these implementations expose several risks that must be mitigated to provide a secured IP core network. All RP routers that are peering with customer PIM-SM routers must implement a PIM import policy to block multicast registration requests for reserved or any other undesirable multicast groups.
NET2011
1 Rule
Protocol Independent Multicast (PIM) join messages received from a downstream multicast Designated Routers (DR) must be filtered for any reserved or any other undesirable multicast groups.
Low Severity
Customer networks that do not maintain a multicast domain and only require the IP multicast service will be required to stand up a PIM-SM router that will be incorporated into the JIE shared tree structure by establishing a peering session with an RP router. Both of these implementations expose several risks that must be mitigated to provide a secure IP core network. All RP routers that are peering with customer PIM-SM routers must implement a PIM import policy to block multicast join requests for reserved or any other undesirable multicast groups.
NET2012
1 Rule
Multicast register messages must be rate limited per each source-group (S, G) entry.
Medium Severity
When a new source starts transmitting in a PIM Sparse Mode network, the DR will encapsulate the multicast packets into register messages and forward them to the Rendezvous Point (RP) using unicast. This process can be taxing on the CPU for both the DR and the RP if the source is running at a high data rate and there are many new sources starting at the same time. This scenario can potentially occur immediately after a network failover. The rate limit for the number of register messages should be set to a relatively low value based on the known number of multicast sources within the multicast domain.
NET2013
1 Rule
Internet Group Management Protocol (IGMP) and Multicast Listener Discovery (MLD) report messages must be filtered to allow hosts to join only those multicast groups that have been approved by the organization.
Low Severity
Real-time multicast traffic can entail multiple large flows of data. Large unicast flows tend to be fairly isolated (e.g., someone doing a file download here or there), whereas multicast can have broader impact on bandwidth consumption resulting in extreme network congestion. Hence, it is imperative that there is multicast admission control to restrict which multicast groups that hosts are allowed to join via IGMP (IPv4) or MLD (IPv6).
NET2014
1 Rule
The number of mroute states resulting from Internet Group Management Protocol (IGMP) or Multicast Listener Discovery (MLD) membership reports must be limited.
Medium Severity
The current multicast paradigm can let any host join any multicast group at any time by sending an Internet Group Management Protocol (IGMP) or Multicast Listener Discovery (MLD) membership report to the Designated Router (DR). In a PIM Sparse Mode network, the DR will send a PIM Join message for the group to the Rendezvous Point (RP). Without any form of admission control, this can pose a security risk to the entire multicast domain, specifically the multicast routers along the shared tree from the DR to the RP that must maintain the mroute state information for each group join request. Hence, it is imperative that the DR is configured to limit the number of mroute state information that must be maintained to mitigate the risk of IGMP (IPv4) or MLD (IPv6) flooding.
NET2015
1 Rule
The number of source-group (SG) states must be limited within the multicast topology where Any Source Multicast (ASM) is deployed.
Medium Severity
Any Source Multicast (ASM) can have many sources for the same groups (many-to-many). For many receivers, the path via the Rendezvous Point (RP) may not be ideal compared with the shortest path from the source to the receiver. By default, the last-hop router will initiate a switch from the shared tree to a source-specific shortest-path tree (SPT) to obtain lower latencies. This is accomplished by the last-hop router sending an (S, G) PIM Join towards S (the source). When the last-hop router begins to receive traffic for the group from the source via the SPT, it will send a PIM Prune message to the RP for the (S, G). The RP will then send a Prune message towards the source. The SPT switchover becomes a scaling issue for large multicast topologies that have many receivers and many sources for many groups because (S, G) entries require more memory than (*, G). Hence, it is imperative to minimize the amount of (S, G) state to be maintained by increasing the threshold that determines when the SPT switchover occurs.
NET2016
1 Rule
Internet Group Management Protocol (IGMP) or Multicast Listener Discovery (MLD) snooping must be implemented within the network access layer.
Low Severity
The last-hop router sends the multicast packet out the interface towards the LAN containing interested receivers. The default behavior for a Layer 2 switch is to forward all multicast traffic out every access switch port that belongs to the VLAN. IGMP snooping is a mechanism used by "Layer 3 aware" switches to maintain a Layer 2 multicast table by examining all IGMP join and leave messages (destined to the all router's multicast address 224.0.0.2) sent between hosts and the multicast routers on the LAN. This will enable the switch to only forward multicast packets out the access switch ports that have connected hosts that have subscribed to the multicast group, thereby reducing the load on the switching backplane as well as eliminating unwanted traffic to uninterested hosts.
NET2017
1 Rule
First-hop redundancy services must be configured to delay any preempt to provide enough time for the Internet Gateway Protocol (IGP) to stabilize.
Low Severity
The Layer 2 connection between the nodes providing first-hop redundancy comes up quickly. If the preemption takes effect prior to the routing protocol converging, traffic is black holed. Traffic will go to the active router that does not have full routing information. It may take several seconds for the IGP to exchange all the routes, longer than the Hot Standby Router Protocol (HSRP), Virtual Router Redundancy Protocol (VRRP), or Gateway Load Balancing Protocol (GLPB) transition. The recommended practice is to delay the preemption action until after the IGP has a chance to stabilize.
NET0160
1 Rule
Written mission justification approval must be obtained from the Office of the DoD CIO prior to establishing a direct connection to the Internet via commercial service provider outside DoD CIO approved Internet access points (e.g. DISA IAP, Cloud Access Point, NIPRnet Federated Gateway, DREN IAP, etc.).
High Severity
Analysis of DoD reported incidents reveal current protective measures at the NIPRNet boundary points are insufficient. Documented ISPs and validated architectures for DMZs are necessary to protect internal network resources from cyber attacks originating from external Internet sources by protective environments.