Internet-Draft | A Policy-based Network Access Control | June 2024 |
Ma, et al. | Expires 27 December 2024 | [Page] |
This document defines a YANG data model for policy-based network access control, which provides consistent and efficient enforcement of network access control policies based on group identity. Moreover, this document defines a mechanism to ease the maintenance of the mapping between a user group identifier and a set of IP/MAC addresses to enforce policy-based network access control.¶
In addition, the document defines a Remote Authentication Dial-in User Service (RADIUS) attribute that is used to communicate the user group identifier as part of identification and authorization information.¶
This note is to be removed before publishing as an RFC.¶
Discussion of this document takes place on the Operations and Management Area Working Group Working Group mailing list (opsawg@ietf.org), which is archived at https://mailarchive.ietf.org/arch/browse/opsawg/.¶
Source for this draft and an issue tracker can be found at https://github.com/boucadair/policy-based-network-acl.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 27 December 2024.¶
Copyright (c) 2024 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
With the increased adoption of remote access technologies (e.g., Virtual Private Networks (VPNs)) and Bring Your Own Device (BYOD) policies, enterprises adopted more flexibility related to how, where, and when employees work and collaborate. However, more flexibility comes with increased risks. Enabling office flexibility (e.g., mobility across many access locations) introduces a set of challenges for large-scale enterprises compared to conventional network access management approaches. Examples of such challenges are listed below:¶
Endpoints do not have stable IP addresses. For example, Wireless LAN (WLAN) and VPN clients, as well as back-end Virtual Machine (VM)-based servers, can move; their IP addresses could change as a result. This means that relying on IP/transport fields (e.g., the 5-tuple) is inadequate to ensure consistent and efficient security policy enforcement. IP address-based policies may not be flexible enough to accommodate endpoints with volatile IP addresses.¶
With the massive adoption of teleworking, there is a need to apply different security policies to the same set of users under different circumstances (e.g., prevent relaying attacks against a local attachment point to the enterprise network). For example, network access might be granted based upon criteria such as users' access location, source network reputation, users' role, time-of- day, type of network device used (e.g., corporate issued device versus personal device), device's security posture, etc. This means that the network needs to recognize the users' identity and their current context, and map the users to their correct access entitlement to the network.¶
This document defines a YANG data model (Section 6.1) for policy-based network access control, which extends the IETF Access Control Lists (ACLs) module defined in [RFC8519]. This module can be used to ensure consistent enforcement of ACL policies based on the group identity.¶
The ACL concept has been generalized to be device-nonspecific, and can be defined at network/administrative domain level [I-D.ietf-netmod-acl-extensions]. To allow for all applications of ACLs, the YANG module for policy-based network ACL defined in Section 6.1 does not limit how it can be used.¶
This document also defines a mechanism to establish a mapping between (1) the user group identifier (ID) and (2) common IP packet header fields and other encapsulating packet data (e.g., MAC address) to execute the policy-based access control.¶
Additionally, the document defines a Remote Authentication Dial-in User Service (RADIUS) [RFC2865] attribute that is used to communicate the user group identifier as part of identification and authorization information (Section 7).¶
Although the document cites MAC addresses as an example in some sections, the document does not make assumptions about which identifiers are used to trigger ACLs. These examples should not be considered as recommendations. Readers should be aware that MAC-based ACLs can be bypassed by flushing out the MAC address. Other implications related to the change of MAC addresses are discussed in [I-D.ietf-madinas-use-cases].¶
The document does not specify how to map the policy group identifiers to dedicated fields (e.g., Group Based Policy (GBP) discussed in Section 6.2.3 of [I-D.ietf-nvo3-encap]).¶
The YANG data model in this document conforms to the Network Management Datastore Architecture (NMDA) defined in [RFC8342].¶
Note to the RFC Editor: This section is to be removed prior to publication.¶
This document contains placeholder values that need to be replaced with finalized values at the time of publication. This note summarizes all of the substitutions that are needed. No other RFC Editor instructions are specified elsewhere in this document.¶
Please apply the following replacements:¶
XXXX --> the assigned RFC number for this document¶
SSSS --> the assigned RFC number for [I-D.ietf-netmod-schedule-yang]¶
2023-01-19 --> the actual date of the publication of this document¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
The meanings of the symbols in tree diagrams are defined in [RFC8340].¶
The document uses the following definition in [RFC3198]:¶
policy¶
The document uses the following definitions and acronyms defined in [RFC8519]:¶
The following definitions and acronyms are used throughout this document:¶
A YANG data model for policy-based network access control that specifies an extension to the "ietf-access-control-list" model [RFC8519]. It allows policy enforcement based on the group identity, which can be used both at the network device level and at the network/administrative domain level.¶
refers to an end-user, host device, or application that actually connects to a network. An end-user is defined as a person. A host device provides compute, memory, storage and networking capabilities and connects to the network without any user intervention. Host devices refer to servers, IoTs and other devices owned by the enterprise. An application is a software program used for a specific service.¶
Access to some networks (e.g., enterprise networks) requires to recognize the users’ identities no matter how, where, and when they connect to the network resources. Then, the network maps the (connecting) users to their access authorization rights. Such rights are defined following local policies. As discussed in Section 1, because (1) there is a large number of users and (2) a user may have different source IP addresses in different network segments, deploying a network access control policy for each IP address or network segment is a heavy workload. An alternate approach is to configure endpoint groups to classify users and enterprise devices and associate ACLs with endpoint groups so that endpoints in each group can share a group of ACL rules. This approach greatly reduces the workload of the administrators and optimizes the ACL resources.¶
The network ACLs can be provisioned on devices using specific mechanisms, such as [RFC8519] or [I-D.ietf-netmod-acl-extensions].¶
Different policies may need to be applied in different contextual situations. For example, companies may restrict (or grant) employees access to specific internal or external resources during work hours, while another policy is adopted during off-hours and weekends. A network administrator may also require to enforce traffic shaping (Section 2.3.3.3 of [RFC2475]) and policing ( Section 2.3.3.4 of [RFC2475]) during peak hours in order not to affect other data services.¶
The architecture of a system that provides real-time and consistent enforcement of access control policies is shown in Figure 1. This architecture includes the following functional entities and interfaces:¶
A service orchestrator which coordinates the overall service, including security policies. The service may be connectivity or any other access to resources that can be hosted and offered by a network.¶
A software-defined networking (SDN) [RFC7149] [RFC7426] controller which is responsible for maintaining endpoint-group based ACLs and mapping the endpoint-group to the associated attributes information (e.g., IP/MAC address). An SDN controller also behaves as a Policy Decision Point (PDP) [RFC3198] and pushes the required access control policies to relevant Policy Enforcement Points (PEPs) [RFC3198]. A PDP is also known as "policy server" [RFC2753].¶
An SDN controller may interact with an Authentication, Authorization and Accounting (AAA) [RFC3539] server or a Network Access Server (NAS) [RFC7542].¶
A Network Access Server (NAS) entity which handles authentication requests. The NAS interacts with an AAA server to complete user authentication using protocols like RADIUS [RFC2865]. When access is granted, the AAA server provides the group identifier (group ID) to which the user belongs when the user first logs onto the network. A new RADIUS attribute is defined in Section 7 for this purpose.¶
The AAA server provides a collection of authentication, authorization, and accounting functions. The AAA server is responsible for centralized user information management. The AAA server is preconfigured with user credentials (e.g., user name and password), possible group identities and related user attributes (users may be divided into different groups based on different user attributes).¶
The Policy Enforcement Point (PEP) is the central entity which is responsible for enforcing appropriate access control policies. A first deployment scenario assumes that the SDN controller maps the group ID to the related common packet header and delivers IP/MAC address based ACL policies to the required PEPs. Another deployment scenario may require that PEPs map incoming packets to their associated source and/or destination endpoint-group IDs, and acts upon the endpoint-group ID based ACL policies (e.g., a group identifier may be carried in packet headers such as discussed in Section 6.2.3 of [I-D.ietf-nvo3-encap]). More details are provided in Section 9.¶
Multiple PEPs may be involved in a network.¶
A PEP exposes a NETCONF interface [RFC6241] to an SDN controller.¶
Figure 1 provides the overall architecture and procedure for policy-based access control management.¶
In reference to Figure 1, the following typical flow is experienced:¶
Administrators (or a service orchestrator) configure an SDN controller with network-level ACLs using the YANG module defined in Section 6.1. An example is provided in Appendix A.1.¶
When a user first logs onto the network, he/she is required to be authenticated (e.g., using user name and password) at the NAS.¶
The authentication request is then relayed to the AAA server using a protocol such as RADIUS [RFC2865]. It is assumed that the AAA server has been appropriately configured to store user credentials, e.g., user name, password, group information, and other user attributes. This document does not restrict what authentication method is used. Administrators may refer to, e.g., Section 7.4 of [I-D.ietf-radext-deprecating-radius] for authentication method recommendations. If the authentication request succeeds, the user is placed in a user group the identity of which is returned to the network access server as the authentication result (see Section 7). If the authentication fails, the user is not assigned any user group, which also means that the user has no access; or the user is assigned a special group with very limited access permissions for the network (as a function of the local policy). ACLs are enforced so that flows from that IP address are discarded (or rate-limited) by the network. In some implementations, AAA server can be integrated with an SDN controller.¶
Either the AAA server or the NAS notifies an SDN controller of the mapping between the user group ID and related common packet header attributes (e.g., IP/MAC address).¶
Either group or IP/MAC address based access control policies are maintained on relevant PEPs under the SDN controller's management. Whether the PEP enforces the group or IP/MAC address based ACL is implementation specific. Both types of ACL policy may exist on the PEP. Appendix A.2 and Appendix A.3 elaborate on each case.¶
The user group is determined by a set of predefined policy criteria (e.g., source IP address, geolocation data, time of day, or device certificate). It uses an identifier (user group ID) to represent the collective identity of a group of users. Users may be moved to different user-groups if their composite attributes, environment, and/or local enterprise policy change.¶
A user is authenticated, and classified at the AAA server, and assigned to a user group. A user's group membership may change as aspects of the user change. For example, if the user group membership is determined solely by the source IP address, then a given user's group ID will change when the user moves to a new IP address that falls outside of the range of addresses of the previous user group.¶
This document does not make any assumption about how user groups are defined. Such considerations are deployment specific and are out of scope. However, and for illustration purposes, Table 1 shows an example of how user group definitions may be characterized. User groups may share several common criteria. That is, user group criteria are not mutually exclusive. For example, the policy criteria of user groups R&D Regular and R&D BYOD may share the same set of users that belong to the R&D organization, and differ only in the type of clients (firm-issued clients vs. users' personal clients). Likewise, the same user may be assigned to different user groups depending on the time of day or the type of day (e.g., weekdays versus weekends), etc.¶
Group Name | Group ID | Group Description |
---|---|---|
R&D | foo-10 | R&D employees |
R&D BYOD | foo-11 | Personal devices of R&D employees |
Sales | foo-20 | Sales employees |
VIP | foo-30 | VIP employees |
The device-group ID is an identifier that represents the collective identity of a group of enterprise end devices. An enterprise device could be a server that hosts applications or software that deliver services to enterprise users. It could also be an enterprise IoT device that serve a limited purpose, e.g., a printer that allows users to scan, print and send emails. Table 2 shows an example of how device-group definitions may be characterized.¶
Group Name | Group ID | Group Description |
---|---|---|
Workflow | bar-40 | Workflow resource servers |
R&D Resource | bar-50 | R&D resource servers |
Printer Resource | bar-60 | Printer resources |
Users accessing an enterprise device should be strictly controlled. Matching abstract device group ID instead of specified addresses in ACL polices helps shield the consequences of address change (e.g., back-end VM-based server migration).¶
An application group is a collection of applications that share a common access control policies. A device may run multiple applications, and different policies might need to be applied to the applications and device. A single application may need to run on multiple devices/VMs/containers, the abstraction of application group eases the process of application migration. For example, the policy does not depend on the transport coordinates (i.e., 5-tuple). Table 3 shows an example of how application-group definitions may be characterized.¶
Group Name | Group ID | Group Description |
---|---|---|
Audio/Video Streaming | baz-70 | Audio/Video conferecing application |
Instant messaging | baz-80 | Messaging application |
document collaboration | baz-90 | Real-time document editing application |
This module specifies an extension to the "ietf-access-control-list" module [RFC8519]. This extension adds endpoint groups so that an endpoint group identifier can be matched upon, and also enable access control policy activation based on date and time conditions.¶
Figure 2 provides the tree structure of the "ietf-ucl-acl" module.¶
The first part of the data model augments the "acl" list in the "ietf-access-control-list" model [RFC8519] with a "endpoint-groups" container having a list of "endpoint group" inside, each entry has a "group-id" that uniquely identifies the endpoint group and a "group-type" parameter to specify the endpoint group type.¶
"group-id" is defined as a string rather than uint to accommodate deployments which require some identification hierarchy within a domain. Such a hierarchy is meant to ease coordination within an administrative domain. There might be cases where a domain needs to tag packets with the group they belong to. The tagging does not need to mirror exactly the "group id" used to populate the policy. How the "group-id" string is mapped to the tagging or field in the packet header in encapsulation scenario is outside the scope of this document. Augmentation may be considered in the future to cover encapsulation considerations.¶
The second part of the data model augments the "matches" container in the IETF ACL model [RFC8519] so that a source and/or destination endpoint group index can be referenced as the match criteria.¶
The third part of the data model augments the "ace" list in the "ietf-access-control-list" model [RFC8519] with date and time specific parameters to allow ACE to be activated based on a date/time condition. Two types of time range are defined, which reuse "recurrence" and "period" groupings defined in the "ietf-schedule" YANG module in [I-D.ietf-netmod-schedule-yang], respectively. Note that the data model augments the definition of "recurrence" grouping with a "duration" data node to specify the duration of time for each occurrence the policy activation is triggered.¶
This module imports types and groupings defined in the "ietf-schedule" module [I-D.ietf-netmod-schedule-yang]. It also augments the "ietf-access-control-list" module [RFC8519].¶
<CODE BEGINS> file "ietf-ucl-acl@2023-01-19.yang" module ietf-ucl-acl { yang-version 1.1; namespace "urn:ietf:params:xml:ns:yang:ietf-ucl-acl"; prefix uacl; import ietf-access-control-list { prefix acl; reference "RFC 8519: YANG Data Model for Network Access Control Lists (ACLs)"; } import ietf-schedule { prefix schedule; reference "RFC SSSS: A Common YANG Data Model for Scheduling"; } organization "IETF OPSWG Working Group"; contact "WG Web: <https://datatracker.ietf.org/wg/opsawg/> WG List: <mailto:opsawg@ietf.org> Editor: Qiufang Ma <mailto:maqiufang1@huawei.com Author: Qin Wu <mailto:bill.wu@huawei.com> Editor: Mohamed Boucadair <mailto:mohamed.boucadair@orange.com> Author: Daniel King <mailto:d.king@lancaster.ac.uk>"; description "This YANG module augments the IETF access control lists (ACLs) module and is meant to ensure consistent enforcement of ACL policies based on the group identity. Copyright (c) 2024 IETF Trust and the persons identified as authors of the code. All rights reserved. Redistribution and use in source and binary forms, with or without modification, is permitted pursuant to, and subject to the license terms contained in, the Revised BSD License set forth in Section 4.c of the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info). This version of this YANG module is part of RFC XXXX (https://www.rfc-editor.org/info/rfcXXXX); see the RFC itself for full legal notices."; revision 2023-01-19 { description "Initial revision."; reference "RFC XXXX: A YANG Data Model and RADIUS Extension for Policy-based Network Access Control"; } feature schedule { description "Indicates support of schedule-based ACEs."; } feature match-on-group { description "Indicates support of matching on endpoint groups."; } feature group { if-feature "uacl:match-on-group"; description "Indicates support of group-based ACLs."; } feature mixed-ipv4-group { if-feature "acl:match-on-ipv4 and uacl:match-on-group"; description "IPv4 and group ACL combinations supported."; } feature mixed-ipv6-group { if-feature "acl:match-on-ipv6 and uacl:match-on-group"; description "IPv6 and group ACL combinations supported."; } feature mixed-ipv4-ipv6-group { if-feature "acl:match-on-ipv4 and acl:match-on-ipv6 and " + "uacl:match-on-group"; description "IPv4, IPv6, and group ACL combinations supported."; } feature mixed-eth-group { if-feature "acl:match-on-eth and uacl:match-on-group"; description "Eth and group ACL combinations supported."; } feature mixed-eth-ipv4-group { if-feature "acl:match-on-eth and acl:match-on-ipv4 and " + "uacl:match-on-group"; description "Eth, IPv4, and group ACL combinations supported."; } feature mixed-eth-ipv6-group { if-feature "acl:match-on-eth and acl:match-on-ipv6 and " + "uacl:match-on-group"; description "Eth, IPv6, and group ACL combinations supported."; } feature mixed-eth-ipv4-ipv6-group { if-feature "acl:match-on-eth and acl:match-on-ipv4 and " + "acl:match-on-ipv6 and uacl:match-on-group"; description "Eth, IPv4, IPv6, and group ACL combinations supported."; } identity group-acl-type { if-feature "group"; base acl:acl-base; description "An Access Control List (ACL) that matches based on an endpoint group identity, which can represent the collective identity of a group of authenticated users, end devices, or applications. An endpoint group identity may be carried in the outer/inner packet header (e.g., via NVO3 encapsulation), may also not correspond to any field in the packet header. Matching on Layer 4 header fields may also exist in the Access Control Entries (ACEs)."; } identity mixed-ipv4-group-type { if-feature "mixed-ipv4-group"; base acl:ipv4-acl-type; base uacl:group-acl-type; description "An ACL that contains a mix of entries that match on fields in the IPv4 header and endpoint group identities, which can represent the collective identity of a group of authenticated users, end devices, or applications. Matching on Layer 4 header fields may also exist in the ACEs."; } identity mixed-ipv6-group-type { if-feature "mixed-ipv6-group"; base acl:ipv6-acl-type; base uacl:group-acl-type; description "An ACL that contains a mix of entries that match on fields in the IPv6 header and endpoint group identities, which can represent the collective identity of a group of authenticated users, end devices, or applications. Matching on Layer 4 header fields may also exist in the ACEs."; } identity mixed-ipv4-ipv6-group-type { if-feature "mixed-ipv4-ipv6-group"; base acl:ipv4-acl-type; base acl:ipv6-acl-type; base uacl:group-acl-type; description "An ACL that contains a mix of entries that match on fields in the IPv4 header, IPv6 header, and endpoint group identities, which can represent the collective identity of a group of authenticated users, end devices, or applications. Matching on Layer 4 header fields may also exist in the ACEs."; } identity mixed-eth-group-type { if-feature "mixed-eth-group"; base acl:eth-acl-type; base uacl:group-acl-type; description "An ACL that contains a mix of entries that match on fields in the Ethernet header and endpoint group identities, which can represent the collective identity of a group of authenticated users, end devices, or applications. Matching on Layer 4 header fields may also exist in the ACEs."; } identity mixed-eth-ipv4-group-type { if-feature "mixed-eth-ipv4-group"; base acl:eth-acl-type; base acl:ipv4-acl-type; base uacl:group-acl-type; description "An ACL that contains a mix of entries that match on fields in the Ethernet header, IPv4 header, and endpoint group identities, which can represent the collective identity of a group of authenticated users, end devices or applications. Matching on Layer 4 header fields may also exist in the ACEs."; } identity mixed-eth-ipv6-group-type { if-feature "mixed-eth-ipv6-group"; base acl:eth-acl-type; base acl:ipv6-acl-type; base uacl:group-acl-type; description "An ACL that contains a mix of entries that match on fields in the Ethernet header, IPv6 header, and endpoint group identities, which can represent the collective identity of a group of authenticated users, end devices or applications. Matching on Layer 4 header fields may also exist in the ACEs."; } identity mixed-eth-ipv4-ipv6-group-type { if-feature "mixed-eth-ipv4-ipv6-group"; base acl:eth-acl-type; base acl:ipv4-acl-type; base acl:ipv6-acl-type; base uacl:group-acl-type; description "An ACL that contains a mix of entries that match on fields in the Ethernet header, IPv4 header, IPv6 header, and endpoint group identities, which can represent the collective identity of a group of authenticated users, end devices or applications. Matching on Layer 4 header fields may also exist in the ACEs."; } identity endpoint-group-type { description "Identity for the type of endpoint group."; } identity user-group { base uacl:endpoint-group-type; description "Identity for the user endpoint group."; } identity device-group { base uacl:endpoint-group-type; description "Identity for the device endpoint group."; } identity application-group { base uacl:endpoint-group-type; description "Identity for the application endpoint group."; } typedef group-id-reference { type leafref { path "/acl:acls/uacl:endpoint-groups" + "/uacl:endpoint-group/uacl:group-id"; } description "Defines a reference to a group identifier."; } augment "/acl:acls" { if-feature "uacl:group"; description "Adds a container for endpoint group definition."; container endpoint-groups { description "Defines a container for the endpoint group list."; list endpoint-group { key "group-id"; description "Definition of the endpoint group list."; leaf group-id { type string { length "1..64"; } description "The endpoint group identifier that uniquely identifies an endpoint group."; } leaf group-type { type identityref { base endpoint-group-type; } description "Specifies the endpoint group type."; } } } } augment "/acl:acls/acl:acl/acl:aces/acl:ace/acl:matches" { if-feature "uacl:match-on-group"; description "Specifies how a source and/or destination endpoint group index can be referenced as the match criteria in the ACEs."; container endpoint-group { when "derived-from-or-self(/acl:acls/acl:acl/acl:type, " + "'uacl:group-acl-type')"; description "Adds new match types."; leaf source-group-id { type group-id-reference; description "The matched source endpoint group ID."; } leaf destination-group-id { type group-id-reference; description "The matched destination endpoint group ID."; } } } augment "/acl:acls/acl:acl/acl:aces/acl:ace" { if-feature "uacl:schedule"; description "Adds schedule parameters to allow the ACE to take effect based on date and time."; container effective-schedule { description "Defines when the access control entry rules are in effect based on date and time condition. If it is not configured, the access control entry is immediately and always in effect."; choice schedule-type { description "Choice based on the type of the time range."; case period { description "The ACE takes effect based on a precise period of time."; uses schedule:period-of-time; } case recurrence { if-feature "schedule:icalendar-recurrence-supported"; description "The ACE takes effect based on a recurrence rule."; uses schedule:icalendar-recurrence; } } } } } <CODE ENDS>¶
The User-Access-Group-ID RADIUS attribute is defined with a globally unique name. The definition of the attribute follows the guidelines in Section 2.7.1 of [RFC6929]. This attribute is used to indicate the user group ID to be used by the NAS. When the User-Access-Group-ID RADIUS attribute is present in the RADIUS Access-Accept, the system applies the related access control to the users after it authenticates.¶
The User-Access-Group-ID Attribute is of type "string" as defined in Section 3.5 of [RFC8044].¶
The User-Access-Group-ID Attribute MAY appear in a RADIUS Access-Accept packet. It MAY also appear in a RADIUS Access-Request packet as a hint to the RADIUS server to indicate a preference. However, the server is not required to honor such a preference. If more than one instance of the User-Access-Group-ID Attribute appears in a RADIUS Access-Accept packet, this means that the user is a member of many groups.¶
The User-Access-Group-ID Attribute MAY appear in a RADIUS CoA-Request packet.¶
The User-Access-Group-ID Attribute MAY appear in a RADIUS Accounting-Request packet. Specifically, this may be used by a NAS to acknowledge that the attribute was received in the RADIUS Access-Request and the NAS is enforcing that policy.¶
The User-Access-Group-ID Attribute MUST NOT appear in any other RADIUS packet.¶
The User-Access-Group-ID Attribute is structured as follows:¶
Table 4 provides a guide as what type of RADIUS packets that may contain User-Access-Group-ID Attribute, and in what quantity.¶
Access-Request | Access-Accept | Access-Reject | Challenge | Attribute |
---|---|---|---|---|
0+ | 0+ | 0 | 0 | User-Access-Group-ID |
Accounting-Request | CoA-Request | CoA-ACK | CoA-NACK | Attribute |
0+ | 0+ | 0 | 0 | User-Access-Group-ID |
The UCL model can be implemented in different ways.¶
In some cases, the UCL model is implemented at the network/administrative domain level with an SDN controller maintaining the dynamical mapping from endpoint group ID to IP/transport fields (e.g., the 5-tuple) and programing the PEPs with IP address/5-tuple based ACLs. In such cases, PEPs do not require to implement specific logic (including hardware) compared to the enforcement of conventional ACLs.¶
It is possible for the UCL model to be implemented at the network device level. While it eliminates the need for an SDN controller to interact frequently with the PEPs for reasons like the user's context of network connection change or VM/application migration, dedicated hardware/software support might be needed for PEPs to understand the endpoint group identifier. In scenrios where the NAS behaves as the PEP which acquires the source and/or destination endpoint group ID from the AAA server, ACL policy enforcement based on the group identity without being encapsulated into packet headers might affect the forwarding performance. Implementations need to evaluate the operational tradeoff (flexibility brought to the network vs. the complexity of implementation) carefully. Such assessment is out of scope of this document.¶
This section uses the template described in Section 3.7 of [I-D.ietf-netmod-rfc8407bis].¶
The YANG modules specified in this document defines schema for data that is designed to be accessed via network management protocols such as NETCONF [RFC6241] or RESTCONF [RFC8040]. The lowest NETCONF layer is the secure transport layer, and the mandatory-to-implement secure transport is Secure Shell (SSH) [RFC6242]. The lowest RESTCONF layer is HTTPS, and the mandatory-to-implement secure transport is TLS [RFC8446].¶
The Network Configuration Access Control Model (NACM) [RFC8341] provides the means to restrict access for particular NETCONF or RESTCONF users to a preconfigured subset of all available NETCONF or RESTCONF protocol operations and content.¶
There are a number of data nodes defined in the "ietf-ucl-acl" YANG module that are writable, creatable, and deletable (i.e., config true, which is the default). These data nodes may be considered sensitive or vulnerable in some network environments. Write operations (e.g., edit-config) and delete operations to these data nodes without proper protection or authentication can have a negative effect on network operations. Specifically, the following subtrees and data nodes have particular sensitivities/vulnerabilities:¶
This list specifies all the endpoint group entries. Unauthorized write access to this list can allow intruders to modify the entries so as to forge an endpoint group that does not exist or maliciously delete an existing endpoint group, which could be used to craft an attack.¶
This subtree specifies a source and/or endpoint group index as match criteria in the ACEs. Unauthorized write access to this data node may allow intruders to modify the group identity so as to permit access that should not be permitted, or deny access that should be permitted.¶
Some of the readable data nodes in the "ietf-ucl-acl" YANG module may be considered sensitive or vulnerable in some network environments. It is thus important to control read access (e.g., via get, get-config, or notification) to these data nodes. Specifically, the following subtrees and data nodes have particular sensitivities/vulnerabilities:¶
This subtree specifies when the access control entry rules are in effect. An unauthorized read access of the list will allow the attacker to determine which rules are in effect, to better craft an attack.¶
This document registers the following URIs in the "IETF XML Registry" [RFC3688].¶
URI: urn:ietf:params:xml:ns:yang:ietf-ucl-acl Registrant Contact: The IESG. XML: N/A, the requested URI is an XML namespace.¶
This document registers the following YANG modules in the "YANG Module Names" registry [RFC6020].¶
name: ietf-ucl-acl prefix: uacl namespace: urn:ietf:params:xml:ns:yang:ietf-ucl-acl maintained by IANA? N reference: RFC XXXX¶
This document requests IANA to assign a new RADIUS attribute type from the IANA registry "Radius Attribute Types" [RADIUS-Types]:¶
Value | Description | Data Type | Reference |
---|---|---|---|
TBA1 | User-Access-Group-ID | string | This-Document |
Let's consider an organization that would like to restrict the access of R&D employees that bring personally owned devices (BYOD) into the workplace.¶
The access requirements are as follows:¶
Permit traffic from R&D BYOD of employees, destined to R&D employees' devices every work day from 8:00:00 to 18:00:00 UTC, starting in January 1st, 2025.¶
Deny traffic from R&D BYOD of employees, destined to finance servers located in the enterprise DC network starting at 8:30:00 of January 20, 2025 with an offset of -08:00 from UTC (Pacific Standard Time) and ending at 18:00:00 in Pacific Standard Time on December 31, 2025.¶
The example shown in Figure 3 illustrates the configuration of an SDN controller using the group-based ACL:¶
This section illustrates an example to configure a PEP using the group-based ACL.¶
The PEP which enforces group-based ACL may acquire group identities from the AAA server if working as a NAS authenticating both the source endpoint and the destination endpoint users. Another case for a PEP enforcing a group-based ACL is to obtain the group identity of the source endpoint directly from the packet field [I-D.smith-vxlan-group-policy]. This example does not intend to be exhaustive.¶
Assume the mapping between device group ID and IP addresses is predefined or acquired via device authentication. Figure 4 shows the ACL configuration delivered from the controller to the PEP. This example is consistent with the example presented in Appendix A.1.¶
The section describes an example of configuring a PEP using IP address based ACL. IP address based access control policies could be applied to the PEP that may not understand the group information, e.g., firewall.¶
Assume an employee in the R&D department accesses the network wirelessly from a non-corporate laptop using IP address 192.0.2.10. The SDN controller associates the user group to which the employee belongs with the user address according to step 1 to 4 in Section 4.1.¶
Assume the mapping between device group ID and IP addresses is predefined or acquired via device authentication. Figure 5 shows an IPv4 address based ACL configuration delivered from the controller to the PEP. This example is consistent with the example presented in Appendix A.1.¶
Figure 6 shows an example of the same policy but with a destination IPv6 prefix.¶
This work has benefited from the discussions of User-group-based Security Policy over the years. In particular, [I-D.you-i2nsf-user-group-based-policy] and [I-D.yizhou-anima-ip-to-access-control-groups] provide mechanisms to establish a mapping between the IP address/prefix of users and access control group IDs.¶
Jianjie You, Myo Zarny, Christian Jacquenet, Mohamed Boucadair, and Yizhou Li contributed to an earlier version of [I-D.you-i2nsf-user-group-based-policy]. We would like to thank the authors of that draft on modern network access control mechanisms for material that assisted in thinking about this document.¶
The authors would like to thank Joe Clarke, Bill Fenner, Benoit Claise, Rob Wilton, David Somers-Harris, Alan Dekok, and Heikki Vatiainen for their valuable comments and great input to this work.¶