RFC 8995
Bootstrapping Remote Secure Key Infrastructure (BRSKI)
RFC 8995: Bootstrapping Remote Secure Key Infrastructure (BRSKI) ✨
okay bestie, this is the title block 🎀
- Series and number: Internet Engineering Task Force (IETF), Request for Comments 8995 🌸
- Category: Standards Track 👑
- ISSN: 2070-1721
- Date: May 2021 📅
- Authors: M. Pritikin (Cisco), M. Richardson (Sandelman Software Works), T. Eckert (Futurewei USA), M. Behringer, and K. Watsen (Watsen Networks) 💜
- Obsoletes and updates: The title block doesn't list any Obsoletes or Updates entries. 🦋
Abstract 🌈
okay so this doc is all about automatically bootstrapping an Autonomic Control Plane 🤖✨ and the way it does that is by bootstrapping a Secure Key Infrastructure first. ngl that's the lowkey genius part. It uses X.509 certificates that the manufacturer installed on the device, and it works together with the manufacturer's authorizing service, both online and offline. We call the whole process the Bootstrapping Remote Secure Key Infrastructure (BRSKI) protocol 💅
Bootstrapping a new device can happen when you have a routable address and a cloud service, when there's only link-local connectivity, or when the network is limited or disconnected 🍓 Support for deployment models with less stringent security requirements is included too, so it's not one-size-fits-all. 🫶
Bootstrapping is complete when the cryptographic identity of the new key infrastructure gets successfully deployed to the device 🔥 and then the established secure connection can be used to deploy a locally issued certificate to the device as well. iconic, periodt 🧚
Status of This Memo 💖
This is an Internet Standards Track document.
This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Further information on Internet Standards is available in Section 2 of RFC 7841.
Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at https://www.rfc-editor.org/info/rfc8995. the official place to check if there's new drama 🙈
Copyright Notice 🎀
Copyright (c) 2021 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 Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. legal stuff, but it matters, so read it 💅
Table of Contents 🦋
1. Introduction
1.1. Prior Bootstrapping Approaches
1.2. Terminology
1.3. Scope of Solution
1.3.1. Support Environment
1.3.2. Constrained Environments
1.3.3. Network Access Controls
1.3.4. Bootstrapping is Not Booting
1.4. Leveraging the New Key Infrastructure / Next Steps
1.5. Requirements for Autonomic Networking Infrastructure (ANI)
Devices
2. Architectural Overview
2.1. Behavior of a Pledge
2.2. Secure Imprinting Using Vouchers
2.3. Initial Device Identifier
2.3.1. Identification of the Pledge
2.3.2. MASA URI Extension
2.4. Protocol Flow
2.5. Architectural Components
2.5.1. Pledge
2.5.2. Join Proxy
2.5.3. Domain Registrar
2.5.4. Manufacturer Service
2.5.5. Public Key Infrastructure (PKI)
2.6. Certificate Time Validation
2.6.1. Lack of Real-Time Clock
2.6.2. Infinite Lifetime of IDevID
2.7. Cloud Registrar
2.8. Determining the MASA to Contact
3. Voucher-Request Artifact
3.1. Nonceless Voucher-Requests
3.2. Tree Diagram
3.3. Examples
3.4. YANG Module
4. Proxying Details (Pledge -- Proxy -- Registrar)
4.1. Pledge Discovery of Proxy
4.1.1. Proxy GRASP Announcements
4.2. CoAP Connection to Registrar
4.3. Proxy Discovery and Communication of Registrar
5. Protocol Details (Pledge -- Registrar -- MASA)
5.1. BRSKI-EST TLS Establishment Details
5.2. Pledge Requests Voucher from the Registrar
5.3. Registrar Authorization of Pledge
5.4. BRSKI-MASA TLS Establishment Details
5.4.1. MASA Authentication of Customer Registrar
5.5. Registrar Requests Voucher from MASA
5.5.1. MASA Renewal of Expired Vouchers
5.5.2. MASA Pinning of Registrar
5.5.3. MASA Check of the Voucher-Request Signature
5.5.4. MASA Verification of the Domain Registrar
5.5.5. MASA Verification of the Pledge
'prior-signed-voucher-request'
5.5.6. MASA Nonce Handling
5.6. MASA and Registrar Voucher Response
5.6.1. Pledge Voucher Verification
5.6.2. Pledge Authentication of Provisional TLS Connection
5.7. Pledge BRSKI Status Telemetry
5.8. Registrar Audit-Log Request
5.8.1. MASA Audit-Log Response
5.8.2. Calculation of domainID
5.8.3. Registrar Audit-Log Verification
5.9. EST Integration for PKI Bootstrapping
5.9.1. EST Distribution of CA Certificates
5.9.2. EST CSR Attributes
5.9.3. EST Client Certificate Request
5.9.4. Enrollment Status Telemetry
5.9.5. Multiple Certificates
5.9.6. EST over CoAP
6. Clarification of Transfer-Encoding
7. Reduced Security Operational Modes
7.1. Trust Model
7.2. Pledge Security Reductions
7.3. Registrar Security Reductions
7.4. MASA Security Reductions
7.4.1. Issuing Nonceless Vouchers
7.4.2. Trusting Owners on First Use
7.4.3. Updating or Extending Voucher Trust Anchors
8. IANA Considerations
8.1. The IETF XML Registry
8.2. YANG Module Names Registry
8.3. BRSKI Well-Known Considerations
8.3.1. BRSKI .well-known Registration
8.3.2. BRSKI .well-known Registry
8.4. PKIX Registry
8.5. Pledge BRSKI Status Telemetry
8.6. DNS Service Names
8.7. GRASP Objective Names
9. Applicability to the Autonomic Control Plane (ACP)
9.1. Operational Requirements
9.1.1. MASA Operational Requirements
9.1.2. Domain Owner Operational Requirements
9.1.3. Device Operational Requirements
10. Privacy Considerations
10.1. MASA Audit-Log
10.2. What BRSKI-EST Reveals
10.3. What BRSKI-MASA Reveals to the Manufacturer
10.4. Manufacturers and Used or Stolen Equipment
10.5. Manufacturers and Grey Market Equipment
10.6. Some Mitigations for Meddling by Manufacturers
10.7. Death of a Manufacturer
11. Security Considerations
11.1. Denial of Service (DoS) against MASA
11.2. DomainID Must Be Resistant to Second-Preimage Attacks
11.3. Availability of Good Random Numbers
11.4. Freshness in Voucher-Requests
11.5. Trusting Manufacturers
11.6. Manufacturer Maintenance of Trust Anchors
11.6.1. Compromise of Manufacturer IDevID Signing Keys
11.6.2. Compromise of MASA Signing Keys
11.6.3. Compromise of MASA Web Service
11.7. YANG Module Security Considerations
12. References
12.1. Normative References
12.2. Informative References
Appendix A. IPv4 and Non-ANI Operations
A.1. IPv4 Link-Local Addresses
A.2. Use of DHCPv4
Appendix B. mDNS / DNS-SD Proxy Discovery Options
Appendix C. Example Vouchers
C.1. Keys Involved
C.1.1. Manufacturer Certification Authority for IDevID
Signatures
C.1.2. MASA Key Pair for Voucher Signatures
C.1.3. Registrar Certification Authority
C.1.4. Registrar Key Pair
C.1.5. Pledge Key Pair
C.2. Example Process
C.2.1. Pledge to Registrar
C.2.2. Registrar to MASA
C.2.3. MASA to Registrar
Acknowledgements
Authors' Addresses1. Introduction ✨
okay so the Bootstrapping Remote Secure Key Infrastructure (BRSKI) protocol is all about securely zero-touch (automated) bootstrapping of brand new, unconfigured devices. We call those devices "pledges" in this doc 🎀 Every pledge comes with an Initial Device Identifier (IDevID) baked in at the factory 🏭✨
"BRSKI" is pronounced like "brewski," which is a casual word for beer in Canada and parts of the Midwestern United States 🍺 [brewski] 😭
This document is mainly built for the ISP and enterprise crowd, specifically the Autonomic Networking Integrated Model and Approach (ANIMA) Autonomic Control Plane (ACP) [RFC8994] 💅 This bootstrap process meets the "make everything secure by default" requirement from Section 3.3 of [RFC7575] 🔒 Other folks who want to use BRSKI will need their own applicability statements, complete with the privacy and security considerations for their setup 🫶 Section 9 covers the detailed applicability for this ACP use case 🌸
Now, BRSKI needs a LOT of back-and-forth between the manufacturer and the owner 💬 In its default modes, it hands control over to the initial owner through a cryptographic transfer 🔑 In its strongest modes, it uses sales channel info to figure out who the owner is ahead of time 🧾 Reselling devices is totally possible, as long as the manufacturer is cool with authorizing the transfer 🛍️ Ownership transfers without manufacturer authorization are not part of this version, though it could be designed into future versions 🔮
This document describes how a pledge finds (or gets found by) the network element it's going to belong to, which will do its bootstrap 🧚 That element is called the "registrar" 👑 Before anything else happens, the pledge and the registrar have to establish mutual trust, and that breaks down into four questions:
- Registrar authenticating the pledge: "Who is this device? What is its identity?"
- Registrar authorizing the pledge: "Is it mine? Do I want it? What are the chances it has been compromised?"
- Pledge authenticating the registrar: "What is this registrar's identity?"
- Pledge authorizing the registrar: "Should I join this network?"
This doc details the protocols and messages that answer those questions 📜 It uses a TLS connection plus a PKIX-shaped (X.509v3) certificate from the pledge (an IEEE 802.1AR IDevID [IDevID]) to answer points 1 and 2 🔐 It also uses a brand new artifact called a "voucher," which the registrar gets from a Manufacturer Authorized Signing Authority (MASA) and then passes along to the pledge, answering points 3 and 4 🎟️✨
A proxy provides very limited connectivity between the pledge and the registrar 🦋
The syntax details of vouchers are covered in [RFC8366] 📖 This document covers the automated protocol mechanisms for getting vouchers, including the "voucher-request" message, which is a minor extension to the voucher format (see Section 3) as defined by [RFC8366] 💜
BRSKI ends with the pledge storing an X.509 root certificate that's enough to verify the registrar's identity 🪪 Along the way, a TLS connection gets set up that can be used directly for Enrollment over Secure Transport (EST) 🙌 Basically, BRSKI gives you an automated version of "Bootstrap Distribution of CA Certificates" from [RFC7030], Section 4.1.1, which says the pledge "MUST [...] engage a human user to authorize the CA certificate using out-of-band data" 🙈 With BRSKI, the pledge can skip that human step and use the voucher instead 🤖 Hooking it up to a full EST enrollment is optional but super easy 🍓
BRSKI is flexible enough to bootstrap other key infrastructures too, like a symmetric key solution, but this document doesn't describe any such system 🌈
1.1. Prior Bootstrapping Approaches
To literally "pull yourself up by the bootstraps" is impossible 💀 Same deal with securely setting up a key infrastructure without outside help, which is also impossible 🙃 Right now, it's commonly accepted that the first connections between nodes are insecure until key distribution is finished 😬 Or, domain-specific keying material (often pre-shared keys, including stuff like Subscriber Identification Module (SIM) cards) gets pre-loaded onto every new device, which is costly and doesn't scale 📉 The existing automated approaches are called non-secured "Trust on First Use (TOFU)" [RFC7435], "resurrecting duckling" [Stajano99theresurrecting], or "pre-staging" 🦆
Another older approach tried to cut down on user actions during bootstrapping without getting rid of them completely 🫠 The original EST protocol [RFC7030] does reduce user actions during bootstrapping, but it doesn't solve how these later steps can be made autonomic (no user actions involved) 👀
- using the Implicit Trust Anchor (TA) [RFC7030] database to authenticate an owner-specific service (not an autonomic solution because the URL must be securely distributed),
- engaging a human user to authorize the CA certificate using out-of-band data (not an autonomic solution because the human user is involved),
- using a configured Explicit TA database (not an autonomic solution because the distribution of an explicit TA database is not autonomic), and
- using a certificate-less TLS mutual authentication method (not an autonomic solution because the distribution of symmetric key material is not autonomic).
These "touch" methods don't meet the requirements for zero-touch 👎
Then there are "call home" technologies 📞 where the pledge first connects to a well-known manufacturer service using a common client-server authentication model. After mutual authentication, the credentials needed to authenticate the target domain get transferred to the pledge 🎁 This causes a few problems and limitations:
- the pledge needs real-time connectivity to the manufacturer service,
- the domain identity is exposed to the manufacturer service (this is a privacy concern), and
- the manufacturer is responsible for making the authorization decisions (this is a liability concern).
BRSKI fixes these issues by extending the EST protocol to automatically distribute vouchers 🦋✨
1.2. Terminology ✨
okay bestie, let's talk vocab 💅 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 read as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they show up in all capitals, just like here 🎀 Anything in lowercase is just regular English, no cap.
okay so here's the glossary, and honestly it's the whole vibe 🧚
ANI: The Autonomic Networking Infrastructure as defined by [RFC8993]. Section 9 spells out the specific requirements for pledges, proxies, and registrars when they're part of an ANI 🌸
Circuit Proxy: A stateful version of the Join Proxy. This is the assumed type of proxy 🦋
drop-ship: Physically shipping equipment that has the "factory default" configuration to its final destination. In zero-touch scenarios, nothing gets staged or preconfigured during the drop-ship 📦
Domain: The set of entities that share a common local trust anchor. That includes the proxy, the registrar, the domain CA, management components, and any existing entity that's already a member of the domain 👑
Domain CA: The domain Certification Authority (CA) provides certification functionalities to the domain. At a minimum, it provides them to a registrar and manages the private key that defines the domain. Optionally, it certifies all elements 🔐
domainID: The domain IDentity is a unique value based on the registrar's CA certificate. Section 5.8.2 explains how it's calculated 🔢
enrollment: The process where a device presents key material to a network and gets a network-specific identity back. For example, a certificate signing request gets presented to a CA, and a certificate comes back in response 📝
IDevID: An Initial Device Identifier X.509 certificate that the vendor installs on new equipment. This is a term from 802.1AR [IDevID] 🪪
imprint: The process where a device gets the cryptographic key material it needs to identify and trust future interactions with a network. This term comes from Konrad Lorenz's biology work with new ducklings 🦆 During a critical period, the duckling assumes that anything that looks like a mother duck is its mother. For a device, the equivalent is getting the fingerprint of the network's root CA certificate. A device that imprints on an attacker meets the same fate as a duckling that imprints on a hungry wolf 🐺 Securely imprinting is a primary focus of this document [imprinting]. The link to Lorenz's work was first noted in [Stajano99theresurrecting] 💜
IPIP Proxy: A stateless alternative to the proxy 🌈
Join Proxy: A domain entity that helps the pledge join the domain. A Join Proxy handles communication for devices that are in an environment where they don't get connectivity until after they've been validated as domain members. To keep things simple, this document sometimes just says "proxy" when it means the Join Proxy. The pledge has no idea it's talking to a proxy instead of directly to a registrar 🫣
Join Registrar (and Coordinator): A representative of the domain that's configured, maybe autonomically, to decide whether a new device is allowed to join. The domain administrator talks to a "Join Registrar (and Coordinator)" to control this process. A Join Registrar is typically "inside" its domain. To keep things simple, this document usually just calls it a "registrar". Within [RFC8993], it's called the "Join Registrar Autonomic Service Agent (ASA)". Other communities abbreviate it as "JRC" ✨
LDevID: A Local Device Identifier X.509 certificate that the equipment's owner installs. This is also a term from 802.1AR [IDevID] 🎀
manufacturer: This is the entity that created the device, and it's the word used all throughout this document. It's usually the original equipment manufacturer (OEM), but in more complicated situations it could be a value added retailer (VAR), or even a systems integrator 🏭 Generally, one of BRSKI's goals is to wipe out small differences between sales channels. The payoff is that a single device with one uniform firmware load can ship directly to every customer. That cuts costs for the manufacturer, and it also means fewer products to support in the field, which makes it more likely that firmware stays up to date 🙌
MASA Audit-Log: An anonymized list of previous owners that the MASA keeps on a per-device (per-pledge) basis, as described in Section 5.8.1 🕵️♀️
MASA Service: A third-party MASA service on the global Internet. The MASA signs vouchers. It also keeps a repository of audit-log information for privacy-protected bootstrapping events. It does not track ownership 🙅♀️
nonced: A voucher (or request) that contains a nonce. This is the normal case 🍓
nonceless: A voucher (or request) that has no nonce, and either relies on accurate clocks for expiration or never expires ⏰
offline: When an architectural component can't do real-time communication with a peer, either because of network connectivity or because the peer is turned off, the operation is said to be happening offline 🌙
Ownership Tracker: An Ownership Tracker service on the global Internet. It uses business processes to accurately track ownership of every device shipped against domains that bought them. It's optional, but this component lets vendors add extra value in cases where their sales and distribution channels make accurate ownership tracking possible 📊 Ownership tracking info shows up in vouchers, as described in [RFC8366] 🧾
Pledge: The prospective (unconfigured) device, which has an identity installed at the factory. Main character energy, honestly 👑
(Public) Key Infrastructure: The collection of systems and processes that keeps a public key system running. The registrar acts as a "Registration Authority"; see [RFC5280] and Section 7 of [RFC5272] 🔑
TOFU: Trust on First Use. It's used similarly to how [RFC7435] describes it. A pledge device makes no security decisions. It just trusts the first registrar that contacts it 🥹 This is also called the "resurrecting duckling" model 🦆💖
Voucher: A signed artifact from the MASA that says which registrar's cryptographic identity a pledge should trust. Trust can be asserted in different ways, so there are different types of vouchers. Multiple voucher types are defined in [RFC8366] 🎟️
1.3. Scope of Solution 🌸
1.3.1. Support Environment
okay so BRSKI can totally handle big router platforms with multi-gigabit interconnections, sitting in tightly controlled data centers 🏢 But it's not only for big gear. It's also meant to scale to thousands of devices in hostile environments, like ISP-provided Customer Premises Equipment (CPE) that gets drop-shipped to the end user 📬 The case where an order gets filled from a distributed warehouse out of common stock and shipped straight to the target location, because the domain owner asked for it, is explicitly supported 🚚 That stock ("SKU") could go to a bunch of possible domain owners, and the eventual owner won't know beforehand which device ends up where 🎁
The bootstrapping process can take minutes, depending on the network infrastructure and how fast the device processes things ⏳ The network communication itself isn't optimized for speed. For privacy reasons, the discovery process lets the pledge avoid announcing itself through broadcasting 🙈
Nomadic or mobile devices often need credentials to get onto the network at a new location. Mobile phone roaming between network operators, or even between cell towers, is one example. This is usually called "handoff" 📱 BRSKI does not provide a low-latency handoff, which is usually a requirement in those situations. For those solutions, BRSKI can be used to create a relationship (an LDevID) with the "home" domain owner. The resulting credentials are then used to get credentials that are better suited to a low-latency handoff 💫
1.3.2. Constrained Environments
People have asked whether this solution works for Internet of Things (IoT) networks in general 🤔 It depends on what the devices are capable of, and [RFC7228] is the best terminology for describing those boundaries 📏
The solution in this document is aimed at non-constrained (i.e., Class 2+ [RFC7228]) devices on a non-challenged network 💪 The whole solution, as described here, is not meant to be used as-is by constrained devices on challenged networks, like 802.15.4 Low-Power and Lossy Networks (LLNs) 🪫
Specifically, some protocol aspects described here might cause congestion collapse or energy exhaustion in intermediate battery-powered routers in an LLN. Those types of networks should not use this solution 😬 These limitations are mostly about the large credential and key sizes needed for device authentication. Defining symmetric key techniques that meet the operational requirements is out of scope. The underlying protocol operations (TLS handshake and signing structures) do have enough algorithm agility to support such techniques once they're defined 🔄
The imprint protocol described here could, however, be used by non-energy-constrained devices joining a non-constrained network. For instance, smart light bulbs are usually mains powered and use 802.11 wireless technology 💡 It could also be used by non-constrained devices across a non-energy-constrained but challenged network, like 802.15.4 🌿 The certificate contents, and the process for resolving the four questions above, do apply to constrained devices. Only the actual on-the-wire imprint protocol might be inappropriate 🫶
1.3.3. Network Access Controls
This document assumes that network access control has already happened, isn't required, or is integrated by the proxy and registrar in a way that the device itself doesn't need to know the details 🛡️ The X.509 IDevID is consistent with IEEE 802.1AR [IDevID] and lines up with 802.1X network access control methods, but here it's used for pledge authentication, not network access control 🔐 Integrating this protocol with network access control, maybe as an Extensible Authentication Protocol (EAP) method (see [RFC3748]), is out of scope for this document 🙅♀️
1.3.4. Bootstrapping is Not Booting
This document describes "bootstrapping" as the protocol used to obtain a local trust anchor. That trust anchor, along with any additional configuration installed later, is expected to persist on the device across system restarts ("booting") 🔁 Bootstrapping only happens infrequently, like when a device changes hands or gets reset to factory default settings 🧹✨
1.4. Leveraging the New Key Infrastructure / Next Steps ✨
Okay so, once the protocol in this document has done its thing, every bootstrapped device ends up with the domain CA trust anchor in common. 🔑 An end-entity (EE) certificate may also have been issued from the domain CA, optionally. This is the part where it gets super fun, because it means you can securely deploy stuff across the whole domain, like:
- Device management 🛠️
- Routing authentication 🛣️
- Service discovery 🔍
The biggest payoff is being able to use the credentials this protocol deploys to secure the Autonomic Control Plane (ACP) [RFC8994]. Main character energy, no cap 👑
1.5. Requirements for Autonomic Networking Infrastructure (ANI) Devices 🎀
BRSKI can be used in a bunch of different environments, and some of the options in this document exist because of requirements that sit outside the ANI scope. So this section lays out the base requirements for ANI devices. 💅
For devices that want to become part of an ANI [RFC8993] that includes an Autonomic Control Plane [RFC8994], the BRSKI protocol MUST be implemented. Non-negotiable, bestie 🫶
The pledge has to do discovery of the proxy as described in Section 4.1, using the Discovery Unsolicited Link-Local (DULL) [RFC8990] M_FLOOD announcements of the GeneRic Autonomic Signaling Protocol (GRASP). 📣
Once a voucher artifact has been successfully validated, a status telemetry MUST be returned; see Section 5.7. Always say thank you, even to a voucher 🥹
An ANIMA ANI pledge MUST implement the EST automation extensions described in Section 5.9. These add onto the EST [RFC7030] so automated devices, the ones with no end user, can keep up. 🤖
The ANI Join Registrar ASA MUST support all the BRSKI and above-listed EST operations. It has to be able to handle the whole list 💪
All ANI devices SHOULD support the BRSKI proxy function, using Circuit Proxies over the Autonomic Control Plane (ACP) (see Section 4.3). Ideally everyone shows up to this party 🎉
2. Architectural Overview 🦋
Okay, the logical pieces of the bootstrapping framework are next. Figure 1 gives you the simplified tour. 🗺️
+------------------------+
+--------------Drop-Ship----------------| Vendor Service |
| +------------------------+
| | M anufacturer| |
| | A uthorized |Ownership|
| | S igning |Tracker |
| | A uthority | |
| +--------------+---------+
| ^
| | BRSKI-
V | MASA
+-------+ ............................................|...
| | . | .
| | . +------------+ +-----------+ | .
| | . | | | | | .
|Pledge | . | Join | | Domain <-------+ .
| | . | Proxy | | Registrar | .
| <-------->............<-------> (PKI RA) | .
| | | BRSKI-EST | | .
| | . | | +-----+-----+ .
|IDevID | . +------------+ | e.g., RFC 7030 .
| | . +-----------------+----------+ .
| | . | Key Infrastructure | .
| | . | (e.g., PKI CA) | .
+-------+ . | | .
. +----------------------------+ .
. .
................................................
"Domain" ComponentsFigure 1: Architecture Overview
We're assuming a multivendor network, which is very real life. 🌈 In that kind of setup, there could be a manufacturer service for each manufacturer that supports devices following this document's spec. Or an integrator could offer a generic service that's authorized by multiple manufacturers. 🤝 It's unlikely an integrator could do ownership tracking for multiple manufacturers, though, because the sales channel integrations needed to track ownership are a whole thing 😭
The domain is the managed network infrastructure, plus the key infrastructure, that the pledge is joining. 🏰 The domain gives the device enough initial connectivity to bootstrap through a proxy. The domain registrar authenticates the pledge, makes the authorization calls, and hands out vouchers it got from the manufacturer service. 🎟️ Optionally, the registrar can also act as a PKI CA, so it's giving multitasking queen 👸
2.1. Behavior of a Pledge 🧚
The pledge goes through a series of steps. Here's the high-level version. ✨
------------
/ Factory \
\ default /
-----+------
|
+------v-------+
| (1) Discover |
+------------> |
| +------+-------+
| |
| +------v-------+
| | (2) Identify |
^------------+ |
| rejected +------+-------+
| |
| +------v-------+
| | (3) Request |
| | Join |
| +------+-------+
| |
| +------v-------+
| | (4) Imprint |
^------------+ |
| Bad MASA +------+-------+
| response | send Voucher Status Telemetry
| +------v-------+
| | (5) Enroll |<---+ (non-error HTTP codes)
^------------+ |\___/ (e.g., 202 "Retry-After")
| Enroll +------+-------+
| failure |
| -----v------
| / Enrolled \
^------------+ |
Factory \------------/
resetFigure 2: Pledge State Diagram
Here's what each state is doing 👀
- Discover a communication channel to a registrar. 🔎
- Identify itself. This happens by presenting an X.509 IDevID credential to the discovered registrar (via the proxy) in a TLS handshake. (The registrar credentials are only provisionally accepted at this time.) 🪪
- Request to join the discovered registrar. A unique nonce is included, which makes sure any responses can be tied to this particular bootstrapping attempt. 🎯
- Imprint on the registrar. This needs verification of the manufacturer-service-provided voucher. A voucher has enough info for the pledge to finish authenticating a registrar. This document goes deep on this step. 🧐
- Enroll. After imprint, an authenticated TLS (HTTPS) connection exists between the pledge and registrar. EST [RFC7030] can then be used to get a domain certificate from a registrar. 🥳
Once that's done, the pledge is officially a member of the domain and can be managed by it. It only repeats the discovery parts of bootstrapping if it gets returned to factory default settings. 🔄
This spec also covers integration with EST enrollment, so pledges can optionally get a locally issued certificate. Any Representational State Transfer (REST) (see [REST]) interface could get integrated in future work too, so this isn't the end of the story 🌸
2.2. Secure Imprinting Using Vouchers 🔐
A voucher is a cryptographically protected artifact (using a digital signature) given to the pledge device. It authorizes a zero-touch imprint on the registrar domain. 💌
The format and cryptographic mechanism of vouchers is described in detail in [RFC8366].
Vouchers give you a flexible way to secure imprinting. The pledge device only imprints when a voucher can be validated. 🙌 At the lowest security levels, the MASA can hand out vouchers pretty much indiscriminately and log claims of ownership by domains. At the highest security levels, issuing vouchers can be wired into complicated sales channel integrations, and those are beyond the scope of this document. Those integrations would verify the actual (legal) ownership of the pledge by the domain. 🧾 This gives a bunch of use cases one shared protocol mechanism on the pledge and registrar devices that are going to be widely deployed in the field. The MASA services can either use the claim mechanisms that are already defined or try out higher or lower security levels. Choices, choices 🍓
Vouchers give you a signed but non-encrypted communication channel among the pledge, the MASA, and the registrar. The registrar keeps control over the transport and policy decisions, so the domain's local security policy gets enforced. 🛡️
2.3. Initial Device Identifier 🪪
Pledge authentication and pledge voucher-request signing happen through a PKIX-shaped certificate that gets installed during manufacturing. This is the 802.1AR IDevID, and it's the basis for authenticating the pledge during the protocol exchanges described here. 🔐 There's no requirement for a common root PKI hierarchy, so each device manufacturer can generate its own root certificate. 🌳 Specifically, the IDevID enables:
- Uniquely identifying the pledge by the Distinguished Name (DN) and subjectAltName (SAN) parameters in the IDevID. The unique identification of a pledge in the voucher objects is derived from those parameters, as described below. Section 10.3 discusses privacy implications of the identifier. 🕵️
- Providing a cryptographic authentication of the pledge to the registrar (see Section 5.3).
- Securing auto-discovery of the pledge's MASA by the registrar (see Section 2.8).
- Signing of a voucher-request by the pledge's IDevID (see Section 3).
- Providing a cryptographic authentication of the pledge to the MASA (see Section 5.5.5).
Sections 7.2.13 (2009 edition) and 8.10.3 (2018 edition) of [IDevID] discuss keyUsage and extendedKeyUsage extensions in the IDevID certificate. [IDevID] acknowledges that adding restrictions in the certificate limits how applicable these long-lived certificates are. This spec really leans into that point, so it RECOMMENDS that no key usage restrictions be included. 🙅♀️ That's consistent with [RFC5280], Section 4.2.1.3, which doesn't require key usage restrictions for end-entity certificates. Keep it open and flexible, babe 💖
2.3.1. Identification of the Pledge ✨
okay so, bestie, a pledge has a 1:1 relationship with a "serial-number" 🎀 This serial-number shows up in the serial-number field of a voucher or voucher-requests (see Section 3), and registrars or MASAs use it in their local policies (see Section 5). Main character energy for this one 💅
There's a (certificate) serialNumber field defined in [RFC5280], Section 4.1.2.2. In ASN.1 it's called the CertificateSerialNumber. This field is NOT relevant to this specification 🙈 Do not mix it up with the serial-number defined by this document, or by [IDevID] and [RFC4519], Section 2.31. Different thing, same vibes, totally separate 💀
The device serial number is defined in Appendix A.1 of [RFC5280] as the X520SerialNumber, with the OID tag id-at-serialNumber 🧚
The device _serialNumber_ field (X520SerialNumber) gets used like this by the pledge to build the serial-number that goes in the voucher-request. To build it, the fields have to be converted into a serial-number of "type string" 🍓
An example of a printable form of the serialNumber field is WI-3005, from [RFC4519], Section 2.31. That section also gives equality and syntax attributes ✨
Real talk: because of how existing device identity provisioning works, some manufacturers have stored serial-numbers in other fields 😭 So registrars SHOULD be configurable, on a per-manufacturer basis, to look for serial-number equivalents in other fields 🫶
As explained in Section 5.5, the registrar MUST again extract the serialNumber itself from the pledge's TLS certificate 👑 If there's any possible confusion about where the serial-number came from, it can consult the serial-number in the pledge request.
2.3.2. MASA URI Extension 🦋
This document defines a new PKIX non-critical certificate extension to carry the MASA URI. This extension is meant to be used in the IDevID certificate. The URI is represented as described in Section 7.4 of [RFC5280].
The URI provides the authority information. The BRSKI "/.well-known" tree [RFC8615] is described in Section 5.
A complete URI MAY be in this extension, including the "scheme", "authority", and "path" 🌈 The complete URI will typically be used in diagnostic or experimental situations. Typically (and with constrained systems in mind), this SHOULD be reduced to only the "authority", in which case a scheme of "https://" (see [RFC7230], Section 2.7.3) and a "path" of "/.well-known/brski" are to be assumed 🌸
The registrar can assume that only the "authority" is present in the extension, if there are no slash ("/") characters in the extension.
Section 7.4 of [RFC5280] calls out various schemes that MUST be supported, including the Lightweight Directory Access Protocol (LDAP), HTTP, and FTP. However, the registrar MUST use HTTPS for the BRSKI-MASA connection 🔥 Lowkey the most important rule in this whole subsection.
The new extension is identified as follows:
MASAURLExtnModule-2016 { iso(1) identified-organization(3) dod(6)
internet(1) security(5) mechanisms(5) pkix(7)
id-mod(0) id-mod-MASAURLExtn2016(96) }
DEFINITIONS IMPLICIT TAGS ::= BEGIN
-- EXPORTS ALL --
IMPORTS
EXTENSION
FROM PKIX-CommonTypes-2009
{ iso(1) identified-organization(3) dod(6) internet(1)
security(5) mechanisms(5) pkix(7) id-mod(0)
id-mod-pkixCommon-02(57) }
id-pe FROM PKIX1Explicit-2009
{ iso(1) identified-organization(3) dod(6) internet(1)
security(5) mechanisms(5) pkix(7) id-mod(0)
id-mod-pkix1-explicit-02(51) } ;
MASACertExtensions EXTENSION ::= { ext-MASAURL, ... }
ext-MASAURL EXTENSION ::= { SYNTAX MASAURLSyntax
IDENTIFIED BY id-pe-masa-url }
id-pe-masa-url OBJECT IDENTIFIER ::= { id-pe 32 }
MASAURLSyntax ::= IA5String
ENDFigure 3: MASAURL ASN.1 Module
The choice of id-pe is based on guidance in Section 4.2.2 of [RFC5280]: "These extensions may be used to direct applications to on-line information about the issuer or the subject". The MASA URL is exactly that: online info about the particular subject 🧠✨
2.4. Protocol Flow 🌊
A representative flow is shown in Figure 4.
+--------+ +---------+ +------------+ +------------+
| Pledge | | Circuit | | Domain | | Vendor |
| | | Join | | Registrar | | Service |
| | | Proxy | | (JRC) | | (MASA) |
+--------+ +---------+ +------------+ +------------+
| | | Internet |
[discover] | | |
|<-RFC 4862 IPv6 addr | | |
|<-RFC 3927 IPv4 addr | Appendix A | Legend |
|-++++++++++++++++++->| | C - Circuit |
| optional: mDNS query| Appendix B | Join Proxy |
| RFCs 6763/6762 (+) | | P - Provisional TLS|
|<-++++++++++++++++++-| | Connection |
| GRASP M_FLOOD | | |
| periodic broadcast| | |
[identity] | | |
|<------------------->C<----------------->| |
| TLS via the Join Proxy | |
|<--Registrar TLS server authentication---| |
[PROVISIONAL accept of server cert] | |
P---X.509 client authentication---------->| |
[request join] | |
P---Voucher-Request(w/nonce for voucher)->| |
P /------------------- | |
P | [accept device?] |
P | [contact vendor] |
P | |--Pledge ID-------->|
P | |--Domain ID-------->|
P | |--optional:nonce--->|
P optional: | [extract DomainID]
P can occur in advance | [update audit-log]
P if nonceless | |
P | |<- voucher ---------|
P \------------------- | w/nonce if provided|
P<------voucher---------------------------| |
[imprint] | |
|-------voucher status telemetry--------->| |
| |<-device audit-log--|
| [verify audit-log and voucher] |
|<--------------------------------------->| |
[enroll] | |
| Continue with enrollment using now | |
| bidirectionally authenticated TLS | |
| session per RFC 7030. | |
[enrolled] | |Figure 4: Protocol Time Sequence Diagram
On initial bootstrap, a new device (the pledge) uses a local service auto-discovery (the GeneRic Autonomic Signaling Protocol (GRASP) or Multicast DNS (mDNS)) to find a Join Proxy 🔍 The Join Proxy connects the pledge to a local registrar (the JRC).
Once it has found a candidate registrar, the fledgling pledge sends some info about itself: its serial number in the form of a voucher-request, plus its IDevID certificate as part of the TLS session 🐣
The registrar can check whether it expected this device to show up, and then it locates a MASA. The MASA location is usually found in an extension in the IDevID. Once the MASA is judged suitable, the whole initial voucher-request (including the device's serial number) gets sent over the Internet in a TLS-protected channel to the manufacturer, along with info about the registrar/owner 🛰️
The manufacturer can then apply policy based on the info provided, plus other sources (like sales records), to decide whether to approve the registrar's claim to own the device. If the claim is accepted, a voucher is issued that tells the device to accept its new owner 💍
The voucher goes back to the registrar, but not straight to the device. The registrar gets a chance to look at the voucher, the MASA's audit-logs, and other sources to figure out whether the device has been tampered with and whether the bootstrap should go ahead 🕵️♀️
No filtering of info is possible in the signed voucher, so this is a binary yes-or-no decision 😤 After the registrar has applied any local policy to the voucher, if it accepts the voucher, the voucher is returned to the pledge for imprinting.
The voucher also includes a trust anchor that the pledge uses to represent the owner. This is what lets the device go from an environment where only the manufacturer is built-in trusted, to one where the owner now has a PKI footprint on the device 🌷
When BRSKI is followed by EST, this single footprint gets leveraged into the full owner's PKI and an LDevID for the device. Later reporting steps send info about whether the process succeeded or failed 📣
2.5. Architectural Components 🧩
2.5.1. Pledge 🌟
The pledge is the device trying to join. It's assumed that the pledge talks to the Join Proxy using link-local network connectivity. In most cases, the pledge has no other connectivity until it completes enrollment and receives some kind of network credential 🥺
2.5.2. Join Proxy 🌉
The Join Proxy provides HTTPS connectivity between the pledge and the registrar. A Circuit Proxy mechanism is described in Section 4. Additional mechanisms, including a Constrained Application Protocol (CoAP) mechanism and a stateless IP in IP (IPIP) mechanism, are the subject of future work 🔮
2.5.3. Domain Registrar 🏰
The domain's registrar operates as the BRSKI-MASA client when requesting vouchers from the MASA (see Section 5.4). The registrar operates as the BRSKI-EST server when pledges request vouchers (see Section 5.1). If the pledge requests an end-entity certificate over the BRSKI-EST connection, the registrar operates as the BRSKI-EST server "Registration Authority" (see Section 5.9).
The registrar uses an Implicit Trust Anchor database for authenticating the BRSKI-MASA connection's MASA TLS server certificate. Configuration or distribution of trust anchors is out of scope for this specification.
The registrar uses a different Implicit Trust Anchor database for authenticating the BRSKI-EST connection's pledge TLS Client Certificate. Configuration or distribution of the BRSKI-EST client trust anchors is out of scope of this specification. Note that the trust anchors in / excluded from the database will affect which manufacturers' devices are acceptable to the registrar as pledges, and they can also be used to limit the set of MASAs that are trusted for enrollment 🔐
2.5.4. Manufacturer Service 🏭
The manufacturer service provides two logically separate functions: the MASA as described in Sections 5.5 and 5.6, and an ownership tracking/auditing function as described in Sections 5.7 and 5.8 💜
2.5.5. Public Key Infrastructure (PKI) 🔐✨
The Public Key Infrastructure (PKI) is the thing that runs all the certificates for the domain, bestie. It's the trust anchor(s) for the whole domain and it's how pledges get enrolled with domain certificates. Iconic behavior. 💅
The voucher gives a way to hand out a single PKI trust anchor, called the "pinned-domain-cert" 🎀. If you want the full set of current trust anchors, that's possible too, using the optional EST integration 🌈.
The domain's registrar acts as a Registration Authority [RFC5272] 🫶. It requests certificates for pledges from the PKI, so it's basically the middle person making the ask. 🧚
The expectations of the PKI are unchanged from EST [RFC7030] 💖. This document does not place any additional architectural requirements on the PKI. No extra homework for the PKI, periodt 🙈.
2.6. Certificate Time Validation ⏰
2.6.1. Lack of Real-Time Clock 🕰️
So, when a device is bootstrapping, a lot of them don't actually know what time it is 💀. Mechanisms like Network Time Protocols can't be secured until bootstrapping is done, so the whole framework is built so it doesn't need the current time at all 🦋. A pledge MAY ignore all time stamps in the voucher and in the certificate validity periods if it doesn't know the current time. Totally valid choice, no judgment 💜.
The pledge is exposed to dates in five places: registrar certificate notBefore, registrar certificate notAfter, voucher created-on, and voucher expires-on. Plus, Cryptographic Message Syntax (CMS) signatures contain a signingTime. That's a lot of dates, bestie 😭.
A pledge with a real-time clock it actually has confidence in MUST check the above time fields in all certificates and signatures that it processes. No skipping this one, it's giving strict 🔥.
If the voucher contains a nonce, then the pledge MUST confirm the nonce matches the original pledge voucher-request. This is how it makes sure the voucher is fresh ✨. See Section 5.2.
2.6.2. Infinite Lifetime of IDevID ♾️
Long-lived pledge certificates "SHOULD be assigned the GeneralizedTime value of 99991231235959Z" for the notAfter field, as explained in [RFC5280] 🍓.
Some deployed IDevID management systems aren't compliant with the 802.1AR requirement for infinite lifetimes, and they end up with typical <= 3 year certificate lifetimes 🥲. Registrars SHOULD be configurable on a per-manufacturer basis to ignore pledge lifetimes when the pledge doesn't follow the recommendations in [RFC5280]. Flexible, lowkey iconic 👑.
2.7. Cloud Registrar ☁️
There are operationally open networks where devices get unauthenticated access to the Internet at large 🌸. In those use cases, the device's management domain has to be discovered within the bigger Internet. The case where a device can boot and reach a larger Internet is less likely within the ANIMA ACP scope, but it may matter more in the future 🔮. In the ANIMA ACP scope, new devices get quarantined behind a Join Proxy 🛡️.
There are also some greenfield situations, like an entirely new installation, where a device might have a management uplink it can use (over a 3G network, for instance) 📡. In that future situation, the device might use that management interface to learn that it should configure itself to become the local registrar 🧚♀️.
To support these scenarios, the pledge MAY contact a well-known URI of a cloud registrar if a local registrar can't be discovered, or if the pledge's target use cases don't include a local registrar 🦋.
If the pledge uses a well-known URI for contacting a cloud registrar, a manufacturer-assigned Implicit Trust Anchor database (see [RFC7030]) MUST be used to authenticate that service as described in [RFC6125] 🔐. Using a DNS-ID for validation is appropriate, and it may include wildcard components on the left-mode side. This matches the human-user configuration of an EST server URI in [RFC7030], which also depends on [RFC6125] 💖.
2.8. Determining the MASA to Contact 🎯
The registrar needs to be able to contact a MASA that the pledge trusts, so it can get vouchers 🎀.
The device's IDevID will normally contain the MASA URL as detailed in Section 2.3. This is the RECOMMENDED mechanism 👑.
Sometimes it's operationally hard to make sure the necessary X.509 extensions are in the pledge's IDevID, because lining up pledge manufacturing with software releases and development is tricky 😩. So, as a final fallback, the registrar MAY be manually configured or distributed with a MASA URL for each manufacturer 🌈. Note that the registrar can only select the configured MASA URL based on the trust anchor, so manufacturers can only use this approach if they make sure a single MASA URL works for all pledges associated with each trust anchor 🫶.
- Voucher-Request Artifact 📜✨
Voucher-requests are how vouchers get requested 💅. The semantics of the voucher-request are described below, in the YANG module.
A pledge forms the "pledge voucher-request", signs it with its IDevID, and submits it to the registrar 🦋.
Then the registrar forms the "registrar voucher-request", signs it with its registrar key pair, and submits it to the MASA 🧚.
The "proximity-registrar-cert" leaf is used in the pledge voucher-requests 🌸. It gives the pledge a way to assert the registrar's proximity.
This network proximity comes from the following properties in the ACP context: the pledge is connected to the Join Proxy (Section 4) over a link-local IPv6 connection 🔗. The Join Proxy doesn't really participate in the cryptography of the TLS connection (like via a Channel Binding), but the registrar can see that the connection comes through the private ACP (ULA) address of the Join Proxy, and it can't come from outside the ACP 👀. So the pledge has to be at most one IPv6 link-local hop away from an existing node on the ACP 🙌.
Other users of BRSKI will need to define other kinds of assertions if the network proximity described above doesn't match their needs 💭.
The "prior-signed-voucher-request" leaf is used in registrar voucher-requests 🎁. If it's present, it's the signed pledge voucher-request artifact. This gives the registrar a way to forward the pledge's signed request to the MASA 🚀. This completes transmission of the signed proximity-registrar-cert leaf 💜.
Unless it's otherwise signaled (outside the voucher-request artifact), the signing structure is as defined for vouchers; see [RFC8366] 📝.
3.1. Nonceless Voucher-Requests 🫥
A registrar MAY also retrieve nonceless vouchers by sending nonceless voucher-requests to the MASA, so it has vouchers ready for when the registrar doesn't have connectivity to the MASA 📦. No prior-signed-voucher-request leaf would be included. The registrar will also need to know the serial number of the pledge. This document doesn't provide a mechanism for the registrar to learn that automatically 🤷♀️. Typically, this happens by scanning a bar code or QR code on the packaging, or through some sales channel integration 📱.
3.2. Tree Diagram 🌳
The following tree diagram gives a high-level view of a voucher-request document 🖼️. The voucher-request builds on the voucher artifact described in [RFC8366]. The tree diagram is described in [RFC8340]. Each node in the diagram is fully described by the YANG module in Section 3.4. Please review the YANG module for a detailed description of the voucher-request format 📚.
module: ietf-voucher-request
grouping voucher-request-grouping
+-- voucher
+-- created-on? yang:date-and-time
+-- expires-on? yang:date-and-time
+-- assertion? enumeration
+-- serial-number string
+-- idevid-issuer? binary
+-- pinned-domain-cert? binary
+-- domain-cert-revocation-checks? boolean
+-- nonce? binary
+-- last-renewal-date? yang:date-and-time
+-- prior-signed-voucher-request? binary
+-- proximity-registrar-cert? binary
Figure 5: YANG Tree Diagram for a Voucher-Request3.3. Examples 🍭
This section has voucher-request examples for illustration purposes 🎨. These examples show JSON prior to CMS wrapping. JSON encoding rules say that any binary content gets base64 encoded ([RFC4648], Section 4). The contents of the (base64) encoded certificates have been elided to save space. For detailed examples, see Appendix C.2. These examples follow the encoding rules defined in [RFC7951] ✨.
Example (1): The following example shows a pledge voucher-request. The assertion leaf is set to "proximity", and the registrar's TLS server certificate is included in the proximity-registrar-cert leaf. See Section 5.2. 💖
{
"ietf-voucher-request:voucher": {
"assertion": "proximity",
"nonce": "62a2e7693d82fcda2624de58fb6722e5",
"serial-number" : "JADA123456789",
"created-on": "2017-01-01T00:00:00.000Z",
"proximity-registrar-cert": "base64encodedvalue=="
}
}
Figure 6: JSON Representation of an Example Voucher-RequestExample (2): The following example shows a registrar voucher-request. The prior-signed-voucher-request leaf is filled in with the pledge's voucher-request (like the prior example). The pledge's voucher-request is a binary CMS-signed object. In the JSON encoding used here, it must be base64 encoded. The nonce and assertion were carried forward from the pledge request to the registrar request. The serial-number is pulled from the pledge's Client Certificate on the TLS connection. See Section 5.5. 🧁
{
"ietf-voucher-request:voucher": {
"assertion" : "proximity",
"nonce": "62a2e7693d82fcda2624de58fb6722e5",
"created-on": "2017-01-01T00:00:02.000Z",
"idevid-issuer": "base64encodedvalue==",
"serial-number": "JADA123456789",
"prior-signed-voucher-request": "base64encodedvalue=="
}
}
Figure 7: JSON Representation of an Example Prior-Signed Voucher-
RequestExample (3): The following example shows a registrar voucher-request. The prior-signed-voucher-request leaf is not filled in with the pledge's voucher-request, and the nonce leaf isn't either. This form might be used by a registrar requesting a voucher when the pledge can't communicate with the registrar (like when it's powered down or still in packaging) and so can't submit a nonce. This scenario is most useful when the registrar knows it won't be able to reach the MASA during deployment. See Section 5.5. 📦🥺
{
"ietf-voucher-request:voucher": {
"created-on": "2017-01-01T00:00:02.000Z",
"idevid-issuer": "base64encodedvalue==",
"serial-number": "JADA123456789"
}
}
Figure 8: JSON Representation of an Offline Voucher-Request3.4. YANG Module 🎀
okay bestie, this next bit is a YANG module [RFC7950] that totally glams up a voucher [RFC8366] into a voucher-request 💅 It also references [ITU.X690] for the encoding stuff ✨
<CODE BEGINS> file "ietf-voucher-request@2021-05-20.yang"
module ietf-voucher-request {
yang-version 1.1;
namespace "urn:ietf:params:xml:ns:yang:ietf-voucher-request";
prefix vcr;
import ietf-restconf {
prefix rc;
description
"This import statement is only present to access
the yang-data extension defined in RFC 8040.";
reference
"RFC 8040: RESTCONF Protocol";
}
import ietf-voucher {
prefix vch;
description
"This module defines the format for a voucher,
which is produced by a pledge's manufacturer or
delegate (MASA) to securely assign a pledge to
an 'owner', so that the pledge may establish a secure
connection to the owner's network infrastructure.";
reference
"RFC 8366: A Voucher Artifact for
Bootstrapping Protocols";
}
organization
"IETF ANIMA Working Group";
contact
"WG Web: <https://datatracker.ietf.org/wg/anima/>
WG List: <mailto:anima@ietf.org>
Author: Kent Watsen
<mailto:kent+ietf@watsen.net>
Author: Michael H. Behringer
<mailto:Michael.H.Behringer@gmail.com>
Author: Toerless Eckert
<mailto:tte+ietf@cs.fau.de>
Author: Max Pritikin
<mailto:pritikin@cisco.com>
Author: Michael Richardson
<mailto:mcr+ietf@sandelman.ca>";
description
"This module defines the format for a voucher-request.
It is a superset of the voucher itself.
It provides content to the MASA for consideration
during a voucher-request.
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 (RFC 2119) (RFC 8174) when, and only when,
they appear in all capitals, as shown here.
Copyright (c) 2021 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 Simplified 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 8995; see the
RFC itself for full legal notices.";
revision 2021-05-20 {
description
"Initial version";
reference
"RFC 8995: Bootstrapping Remote Secure Key Infrastructure
(BRSKI)";
}
// Top-level statement
rc:yang-data voucher-request-artifact {
uses voucher-request-grouping;
}
// Grouping defined for future usage
grouping voucher-request-grouping {
description
"Grouping to allow reuse/extensions in future work.";
uses vch:voucher-artifact-grouping {
refine "voucher/created-on" {
mandatory false;
}
refine "voucher/pinned-domain-cert" {
mandatory false;
description
"A pinned-domain-cert field is not valid in a
voucher-request, and any occurrence MUST be ignored.";
}
refine "voucher/last-renewal-date" {
description
"A last-renewal-date field is not valid in a
voucher-request, and any occurrence MUST be ignored.";
}
refine "voucher/domain-cert-revocation-checks" {
description
"The domain-cert-revocation-checks field is not valid in a
voucher-request, and any occurrence MUST be ignored.";
}
refine "voucher/assertion" {
mandatory false;
description
"Any assertion included in registrar voucher-requests
SHOULD be ignored by the MASA.";
}
augment "voucher" {
description
"Adds leaf nodes appropriate for requesting vouchers.";
leaf prior-signed-voucher-request {
type binary;
description
"If it is necessary to change a voucher, or re-sign and
forward a voucher that was previously provided along a
protocol path, then the previously signed voucher SHOULD
be included in this field.
For example, a pledge might sign a voucher-request
with a proximity-registrar-cert, and the registrar
then includes it as the prior-signed-voucher-request
field. This is a simple mechanism for a chain of
trusted parties to change a voucher-request, while
maintaining the prior signature information.
The registrar and MASA MAY examine the prior-signed
voucher information for the
purposes of policy decisions. For example, this
information could be useful to a MASA to determine
that both the pledge and registrar agree on proximity
assertions. The MASA SHOULD remove all
prior-signed-voucher-request information when
signing a voucher for imprinting so as to minimize
the final voucher size.";
}
leaf proximity-registrar-cert {
type binary;
description
"An X.509 v3 certificate structure, as specified by
RFC 5280, Section 4, encoded using the ASN.1
distinguished encoding rules (DER), as specified
in ITU X.690.
The first certificate in the registrar TLS server
certificate_list sequence (the end-entity TLS
certificate; see RFC 8446) presented by the registrar
to the pledge. This MUST be populated in a pledge's
voucher-request when a proximity assertion is
requested.";
reference
"ITU X.690: Information Technology - ASN.1 encoding
rules: Specification of Basic Encoding Rules (BER),
Canonical Encoding Rules (CER) and Distinguished
Encoding Rules (DER)
RFC 5280: Internet X.509 Public Key Infrastructure
Certificate and Certificate Revocation List (CRL)
Profile
RFC 8446: The Transport Layer Security (TLS)
Protocol Version 1.3";
}
}
}
}
}
<CODE ENDS>Figure 9: YANG Module for Voucher-Request
4. Proxying Details (Pledge -- Proxy -- Registrar) ✨
okay so this section is normative for uses with an ANIMA ACP 🫶 The GRASP mechanism is part of the ACP. Other users of BRSKI will need to define an equivalent proxy mechanism and an equivalent mechanism to configure the proxy, so keep that in mind if you're not in ANIMA land 🌸
The role of the proxy is to facilitate communications. It forwards packets between the pledge and a registrar that has been provisioned to the proxy via full GRASP ACP discovery 🦋
This section defines a stateful proxy mechanism that is referred to as a "circuit" proxy. It's a form of Application Level Gateway (see [RFC2663], Section 2.9) and it's giving main character energy 👑
The proxy does not terminate the TLS handshake: it passes streams of bytes onward without examination. A proxy MUST NOT assume any specific TLS version. Please see [RFC8446], Section 9.3 for details on TLS invariants 💜
A registrar can directly provide the proxy announcements described below, in which case the announced port can point directly to the registrar itself. In this scenario, the pledge is unaware that there is no proxying occurring. This is useful for registrars that are servicing pledges on directly connected networks 🧚
As a result of the proxy discovery process in Section 4.1.1, the port number exposed by the proxy does not need to be well known or require an IANA allocation. No stress, bestie 🌈
During the discovery of the registrar by the Join Proxy, the Join Proxy will also learn which kinds of proxy mechanisms are available. This will allow the Join Proxy to use the lowest impact mechanism that the Join Proxy and registrar have in common. Literally the smart move 🧠✨
In order to permit the proxy functionality to be implemented on the maximum variety of devices, the chosen mechanism should use the minimum amount of state on the proxy device. While many devices in the ANIMA target space will be rather large routers, the proxy function is likely to be implemented in the control-plane CPU of such a device, with available capabilities for the proxy function similar to many class 2 IoT devices 🍓
The document [ANIMA-STATE] provides a more extensive analysis and background of the alternative proxy methods. If you want the deep dive, that's where to go 💅
4.1. Pledge Discovery of Proxy ✨
okay so this is where the pledge goes hunting for a proxy, and honestly the vibe is sneaky but polite 💅 The result of all this discovery is a logical communication with a registrar, but it goes through a proxy. The proxy is totally transparent to the pledge, like it's not even there 🫥 And the communication between the pledge and the Join Proxy happens over IPv6 link-local addresses 🌸
To find the proxy, the pledge does the following:
- MUST: Get a local address using the IPv6 methods in "IPv6 Stateless Address Autoconfiguration" [RFC4862]. Using temporary addresses [RFC8981] is encouraged 🦋 To limit pervasive monitoring [RFC7258], a new temporary address MAY use a short lifetime (aka set
TEMP_PREFERRED_LIFETIMEto be short) 🙈 Pledges will generally prefer IPv6 link-local addresses, and discovery of the proxy will be by link-local mechanisms, so it's giving very local, very private energy ✨ IPv4 methods are described in Appendix A. - MUST: Listen for GRASP M_FLOOD [RFC8990] announcements of the objective:
"AN_Proxy". See Section 4.1.1 for the details of the objective 💜 The pledge MAY listen concurrently for other sources of information; see Appendix B.
Once a proxy is discovered, the pledge talks to a registrar through that proxy using the bootstrapping protocol defined in Section 5 🎀
Here's the tea: the GRASP M_FLOOD mechanism is passive for the pledge, but the non-normative other methods (mDNS and IPv4 methods) in Appendix B are active 😭 The pledge SHOULD run those methods in parallel with listening for M_FLOOD. The active methods SHOULD back off by doubling, up to a maximum of one hour, so the network doesn't get flooded with discovery attempts 🫶 Detecting a physical link status change (Ethernet carrier, for instance) SHOULD reset the back-off timers 🌈
The pledge could discover more than one proxy on a given physical interface. And it can have a multitude of physical interfaces too, because a Layer 2/3 Ethernet switch may have hundreds of physical ports 🍓
Each possible proxy offer SHOULD be attempted until a valid voucher is received. Lowkey, there are many ways the attempt can fail, but it doesn't count as a success until the voucher has been validated ✨
The connection attempts via a single proxy SHOULD exponentially back off, up to a maximum of one hour, to avoid overloading the network infrastructure 💀 The back-off timer for each one MUST be independent of the other connection attempts 💅
Connection attempts SHOULD run in parallel to avoid head-of-queue problems. The reason is that an attacker running a fake proxy or registrar could intentionally do protocol actions slowly 🙈 Connection attempts to different proxies SHOULD be sent with an interval of 3 to 5s. The pledge SHOULD keep listening for additional GRASP M_FLOOD messages during those attempts 👑
Each connection attempt through a distinct Join Proxy MUST have a unique nonce in the voucher-request 🔥
Once a connection to a registrar is established (for example, a TLS session key is set up), there are expectations of more timely responses; see Section 5.2 🌸
Once all the discovered services have been attempted (assuming none succeeded), the device MUST go back to listening for GRASP M_FLOOD. It SHOULD periodically retry any manufacturer-specific mechanisms. The pledge MAY prioritize selection order as appropriate for the anticipated environment 🧚
4.1.1. Proxy GRASP Announcements 📣
A proxy uses the DULL GRASP M_FLOOD mechanism to announce itself. This announcement can ride in the same message as the ACP announcement detailed in [RFC8994] 🦋
The formal Concise Data Definition Language (CDDL) [RFC8610] definition is:
<CODE BEGINS> file "proxygrasp.cddl"
flood-message = [M_FLOOD, session-id, initiator, ttl,
+[objective, (locator-option / [])]]
objective = ["AN_Proxy", objective-flags, loop-count,
objective-value]
ttl = 180000 ; 180,000 ms (3 minutes)
initiator = ACP address to contact registrar
objective-flags = sync-only ; as in the GRASP spec
sync-only = 4 ; M_FLOOD only requires
; synchronization
loop-count = 1 ; one hop only
objective-value = any ; none
locator-option = [ O_IPv6_LOCATOR, ipv6-address,
transport-proto, port-number ]
ipv6-address = the v6 LL of the Proxy
$transport-proto /= IPPROTO_TCP ; note that this can be any value
; from the IANA protocol registry,
; as per RFC 8990, Section 2.9.5.1,
; Note 3.
port-number = selected by Proxy
<CODE ENDS>
Figure 10: CDDL Definition of Proxy Discovery MessageHere's an example M_FLOOD announcing a proxy at fe80::1, on TCP port 4443 🎀
[M_FLOOD, 12340815, h'fe800000000000000000000000000001', 180000,
[["AN_Proxy", 4, 1, ""],
[O_IPv6_LOCATOR,
h'fe800000000000000000000000000001', IPPROTO_TCP, 4443]]]
Figure 11: Example of Proxy Discovery MessageOn a small network, the registrar MAY include the GRASP M_FLOOD announcements to locally connected networks 🌈
The $transport-proto above indicates the method that the pledge-proxy-registrar will use. The TCP method described here is mandatory, and other proxy methods, such as CoAP methods not defined in this document, are optional. Other methods MUST NOT be enabled unless the Join Registrar ASA indicates support for them in its own announcement 💜
4.2. CoAP Connection to Registrar 🧚
The use of CoAP to connect from pledge to registrar is out of scope for this document and is described in future work 🌸 See [ANIMA-CONSTRAINED-VOUCHER].
4.3. Proxy Discovery and Communication of Registrar 🔥
The registrar SHOULD announce itself so that proxies can find it and determine what kinds of connections can be terminated 💅
The registrar announces itself using GRASP M_FLOOD messages, with the "AN_join_registrar" objective, within the ACP instance. A registrar may announce any convenient port number, including stock port 443 ✨ ANI proxies MUST support GRASP discovery of registrars 🫶
The M_FLOOD is formatted as follows:
[M_FLOOD, 51804321, h'fda379a6f6ee00000200000064000001', 180000,
[["AN_join_registrar", 4, 255, "EST-TLS"],
[O_IPv6_LOCATOR,
h'fda379a6f6ee00000200000064000001', IPPROTO_TCP, 8443]]]
Figure 12: An Example of a Registrar Announcement MessageThe formal CDDL definition is:
<CODE BEGINS> file "jrcgrasp.cddl"
flood-message = [M_FLOOD, session-id, initiator, ttl,
+[objective, (locator-option / [])]]
objective = ["AN_join_registrar", objective-flags, loop-count,
objective-value]
initiator = ACP address to contact registrar
objective-flags = sync-only ; as in the GRASP spec
sync-only = 4 ; M_FLOOD only requires
; synchronization
loop-count = 255 ; mandatory maximum
objective-value = text ; name of the (list of) supported
; protocols: "EST-TLS" for RFC 7030.
<CODE ENDS>
Figure 13: CDDL Definition for Registrar Announcement MessageThe M_FLOOD message MUST be sent periodically. The default period SHOULD be 60 seconds, and the value SHOULD be operator configurable, but SHOULD NOT be smaller than 60 seconds. The frequency of sending MUST be such that the aggregate amount of periodic M_FLOODs from all flooding sources causes only negligible traffic across the ACP 🦋
Here are some examples of locators, for illustrative purposes. Only the first one ($transport-protocol = 6, TCP) is defined in this document and is mandatory to implement 🌈
locator1 = [O_IPv6_LOCATOR, fd45:1345::6789, 6, 443]
locator2 = [O_IPv6_LOCATOR, fd45:1345::6789, 17, 5683]
locator3 = [O_IPv6_LOCATOR, fe80::1234, 41, nil]A protocol of 6 indicates that TCP proxying on the indicated port is desired 💖
Registrars MUST announce the set of protocols that they support, and they MUST support TCP traffic 👑
Registrars MUST accept HTTPS/EST traffic on the TCP ports indicated 🍓
Registrars MUST support the ANI TLS Circuit Proxy and therefore BRSKI across HTTPS/TLS native across the ACP 🫶
In the ANI, the ACP-secured instance of GRASP [RFC8990] MUST be used for discovery of ANI registrar ACP addresses and ports by ANI proxies. Therefore, the TCP leg of the proxy connection between the ANI proxy and ANI registrar also runs across the ACP 🌸
5. Protocol Details (Pledge -- Registrar -- MASA) ✨
okay so this is where the BRSKI flow actually starts 🦋 the pledge MUST initiate BRSKI after boot if it is unconfigured 🌸 but the pledge MUST NOT automatically initiate BRSKI if it has been configured or is in the process of being configured 🙅♀️ bestie, a device that already knows who it is does NOT need to restart onboarding 💅
BRSKI is described as extensions to EST [RFC7030] 🧁 the whole point is to cut down on the number of TLS connections and crypto operations the pledge has to do 🧠 the registrar implements the BRSKI REST interface inside the /.well-known/brski URI tree, and it also implements the existing EST URIs as described in EST [RFC7030], Section 3.2.2. the channel between the pledge and the registrar is called "BRSKI-EST" (see Figure 1) 💜
the channel between the registrar and the MASA is a brand new one, kinda like EST, living in the newly registered /.well-known/brski tree 🌈 for clarity we call this one "BRSKI-MASA" (see Figure 1) 🎀
the MASA URI is https:// authority /.well-known/brski 🔗
BRSKI uses existing CMS message formats for the existing EST operations 📦 and it uses JSON [RFC8259] for all the new operations defined here, plus the voucher formats 🍓 anywhere a binary value has to ride along inside a JSON string, use base64 ([RFC4648], Section 4), as per [RFC7951], Section 6.6 ✨
while EST ([RFC7030], Section 3.2) doesn't insist on HTTP persistent connections ([RFC7230], Section 6.3), BRSKI-EST connections SHOULD use persistent connections 🫶 the reason is lowkey genius: the provisional TLS state only happens once, and resolving that provisional state isn't open to a Man-in-the-Middle (MITM) attack during the critical phase 🔥
if non-persistent connections are used anyway, then BOTH the pledge and the registrar MUST remember the certificates they saw and sent on the first connection 🧠 they MUST check every later connection for those same certificates, and each end MUST use the same certificates as well 😭 ngl this makes rolling certificates on the registrar kinda painful 💀
summarized automation extensions for the BRSKI-EST flow:
- the pledge either attempts concurrent connections via each discovered proxy, or it times out quickly and tries connections in series, as explained at the end of Section 5.1 🌊
- the pledge provisionally accepts the registrar certificate during the TLS handshake, as detailed in Section 5.1 🤝
- the pledge requests a voucher using the new REST calls described below, and then that voucher gets validated ✅
- the pledge completes authentication of the server certificate, as detailed in Section 5.6.1 🔐 this is what moves the BRSKI-EST TLS connection out of the provisional state 🦋
- the mandatory bootstrap steps wrap up with voucher status telemetry (see Section 5.7) 📊
once all that is done, the BRSKI-EST TLS connection can be used for EST enrollment 🎉
the extensions for a registrar (the equivalent of an EST server) are:
- client authentication is automated using the IDevID, as per the EST certificate-based client authentication 🪪 the subject field's DN encoding MUST include the "serialNumber" attribute with the device's unique serial number, as explained in Section 2.3.1 🔢
- the registrar requests and validates the voucher from the MASA 🧾
- the registrar forwards the voucher to the pledge when the pledge asks for it 🎁
- the registrar performs log verifications (described in Section 5.8.3) on top of its local authorization checks before accepting optional pledge device enrollment requests 🕵️♀️
5.1. BRSKI-EST TLS Establishment Details
the pledge sets up the TLS connection with the registrar through the Circuit Proxy (see Section 4) 🚇 but the TLS handshake is actually with the registrar 💅 the BRSKI-EST pledge is the TLS client, and the BRSKI-EST registrar is the TLS server 🤓 every security association that gets established is between the pledge and the registrar, no matter what the proxy is doing in the middle 🫶
use of TLS 1.3 (or newer) is encouraged ✨ TLS 1.2 or newer is REQUIRED on the pledge side 🔒 TLS 1.3 (or newer) SHOULD be available on the registrar server interface and the registrar client interface, but TLS 1.2 MAY be used 🌸 TLS 1.3 (or newer) SHOULD be available on the MASA server interface too, but TLS 1.2 MAY be used there as well 🦋
setting up the BRSKI-EST TLS connection follows "Bootstrap Distribution of CA Certificates" (Section 4.1.1) of [RFC7030] 🧁 so the client gets authenticated with the IDevID certificate, and the EST server (the registrar) gets provisionally authenticated with an unverified server certificate 🤝 configuring or distributing the trust anchor database used to validate the IDevID certificate is out of scope for this spec 📚 but heads up: which trust anchors are in / excluded from that database decides which manufacturers' devices the registrar will accept as pledges 👀 it can also be used to limit which MASAs are trusted for enrollment 🚫
the signature in the certificate MUST be validated even if a signing key cannot (yet) be validated 🧐 and the certificate (or chain) MUST be retained for later validation 🗂️
a self-signed certificate for the registrar is acceptable, because the voucher can validate it once enrollment succeeds 🌈
⚠️ the pledge does input validation on ALL the data it receives until a voucher is verified (as specified in Section 5.6.1) and the TLS connection leaves the provisional state. until all of that is done, the pledge could be talking to an attacker 😱
the pledge code needs to be written assuming that ALL data is being sent at this point to an unauthenticated peer 🙈 so any data received inside a TLS connection MUST be treated as untrusted 🚩 this especially applies to HTTP headers and to the CMS structures that make up the voucher 🎀
a pledge that can connect to multiple registrars concurrently SHOULD do so 💜 some devices just can't, because of missing threading or resource limits 🥲 concurrent connections also defeat a malicious proxy trying to pull off a TCP Slowloris-like attack (see [slowloris]) 🐌🚫
a pledge that can't keep up with as many connections as there are eligible proxies will need to rotate through the choices, terminating connections that don't seem to be making progress 🔄 if no connection is making progress after 5 seconds, the pledge SHOULD drop the oldest connection and move on to a different proxy: the one it has talked to least recently 🕰️ if no other proxies were discovered, the pledge MAY keep waiting, as long as it is also listening for new proxy announcements 👂
5.2. Pledge Requests Voucher from the Registrar
when the pledge bootstraps, it asks a registrar for a voucher 🎟️
this is an HTTPS POST using the operation path value /.well-known/brski/requestvoucher 📮
the pledge voucher-request Content-Type is as follows:
application/voucher-cms+json: [RFC8366] defines a "YANG-defined JSON document that has been signed using a Cryptographic Message Syntax (CMS) structure" 🧠 and the voucher-request from Section 3 gets created in the same way ✨ this is the same media type defined in [RFC8366], and it's also used for the pledge voucher-request 💅 the pledge MUST sign the request using the credentials in Section 2.3 🔏
registrar implementations SHOULD anticipate future media types 🔮 but of course, they'll just fail the request if they don't know the type yet 🤷♀️
the pledge SHOULD include an "Accept" header field (see [RFC7231], Section 5.3.2) saying which media type it can take for the voucher response 🎀 application/voucher-cms+json is defined in [RFC8366], but constrained voucher formats are expected down the line 🌱 registrars and the MASA are expected to be flexible about what they accept 🫶
the pledge fills in the voucher-request fields like this:
created-on: pledges with a real-time clock are RECOMMENDED to fill this in with the current date and time in yang:date-and-time format ⏰ it gives the MASA extra info 📝 pledges without a real-time clock MAY omit this field 🕰️
nonce: the pledge voucher-request MUST contain a cryptographically strong random or pseudo-random number nonce (see [RFC4086], Section 6.2) 🎲 the nonce is usually generated super early in the boot sequence, so there's a worry that the same nonce could come out across multiple boots or after a factory reset 😬 different nonces MUST be generated for each bootstrapping attempt, whether they happen in series or concurrently 🔁 the freshness of this nonce helps make up for not having a real-time clock, as explained in Section 2.6.1 ⭐
assertion: the pledge shows it supports the mechanism in this document by putting the value proximity in the voucher-request, and it MUST include the proximity-registrar-cert field (below) 🧲
proximity-registrar-cert: in a pledge voucher-request, this is the first certificate in the TLS server "certificate_list" sequence (see [RFC8446], Section 4.4.2) that the registrar showed to the pledge 📜 so basically the end-entity certificate 🙌 this MUST be populated in a pledge voucher-request 💖
serial-number: the pledge's serial number goes in the voucher-request 🔢 it's only there as a sanity check, though, and the registrar must NOT forward it, as described in Section 5.5 🙅♀️
ALL other fields MAY be omitted in the pledge voucher-request 🌷
see an example JSON payload of a pledge voucher-request in Section 3.3, Example 1 📖
the registrar confirms that the assertion is proximity and that the pinned proximity-registrar-cert is the registrar's certificate 🕵️♀️ if that validation fails, then there's an on-path attacker (MITM) 🚨 and the connection MUST be closed after an HTTP 401 error code gets returned 🚪
😤 this is the anti-sneaky-attacker check. if the certs don't match, the registrar bounces with a 401 and the connection is closed. no vibes, just rules 💀
5.3. Registrar Authorization of Pledge
in a fully automated network, every device has to be securely identified and authorized to join the domain 🌐 all devices must be securely identified, no exceptions 🫡
a registrar accepts or declines a request to join the domain based on the authenticated identity presented 🛂 for different networks, examples of automated acceptance can include allowing:
- any device of a specific type (as determined by the X.509 IDevID) 🧩
- any device from a specific vendor (as determined by the X.509 IDevID) 🏭
- a specific device from a vendor (as determined by the X.509 IDevID) against a domain acceptlist 📋 (the mechanism for checking a shared acceptlist that multiple registrars might use is out of scope 🙃)
if validation fails, the registrar SHOULD respond with the HTTP 404 error code 🙈 if the voucher-request is in an unknown format, an HTTP 406 error code is more appropriate 🙅♀️ a situation that admin action could fix (like adding a vendor to an acceptlist) MAY be answered with a 403 HTTP error code 🛠️
if authorization succeeds, the registrar gets a voucher from the MASA service (see Section 5.5) and hands that MASA-signed voucher back to the pledge, as described in Section 5.6 🎀👑
5.4. BRSKI-MASA TLS Establishment Details ✨
okay so the BRSKI-MASA TLS connection is a "normal" TLS connection 💖, the same kind you'd use for HTTPS REST interfaces 🌸. The registrar starts it and uses the MASA URL we got from Section 2.8. The mechanisms in [RFC6125] SHOULD be used when authenticating the MASA, with a DNS-ID that matches the one in the IDevID 🎀. Registrars MAY include a mechanism to override the MASA URL on a manufacturer-by-manufacturer basis, and inside that override it's appropriate to provide alternate anchors. Some vendors typically do this to set up explicit (or private) trust anchors for validating their MASA as part of a sales channel integration 🧚.
Using TLS 1.3 (or newer) is encouraged, bestie 🫶. TLS 1.2 or newer is REQUIRED, no exceptions 💅. TLS 1.3 (or newer) SHOULD be available ✨.
As described in [RFC7030], the MASA and the registrars SHOULD be prepared to support TLS Client Certificate authentication and/or HTTP Basic, Digest, or Salted Challenge Response Authentication Mechanism (SCRAM) authentication 🔐. This connection MAY also have no client authentication at all 🦋.
Registrars SHOULD permit trust anchors to be preconfigured on a per-vendor (MASA) basis 🌈. Registrars SHOULD include the ability to configure a TLS Client Certificate on a per-MASA basis, or to use no Client Certificate. Registrars SHOULD also permit HTTP Basic and Digest authentication to be configured 💜.
The authentication of the BRSKI-MASA connection does not change the voucher-request process, because voucher-requests are already signed by the registrar 🙈. Instead, this authentication gives access control to the audit-log, as described in Section 5.8 📒.
Implementers, heads up: contacting the MASA establishes a secured API connection with a web service, and there are a bunch of authentication models being explored across the industry 👀. Registrars are RECOMMENDED to fail gracefully and generate useful administrative notifications or logs when they get unexpected HTTP 401 (Unauthorized) responses from the MASA 😭.
5.4.1. MASA Authentication of Customer Registrar
Per-customer options need the customer's registrar to be uniquely identified 🎯. You can do this with any stateless method HTTPS supports, like HTTP Basic or Digest authentication (so, using a password), but TLS Client Certificate authentication is RECOMMENDED 🔑.
Stateful methods involving API tokens, or HTTP Cookies, are not recommended 🙅♀️.
The setup and configuration of per-customer Client Certificates is expected to happen as part of a sales ordering process 🛒.
Using public PKI (that is, WebPKI) end-entity certificates to identify the registrar is reasonable. If everyone did this universally, a MASA could identify a customer's registrar just by a Fully Qualified Domain Name (FQDN) 🌐.
Using DANE records in DNSSEC-signed zones would also let you use a FQDN to identify customer registrars 🧁.
A third option, the simplest but least flexible, is for the MASA to just store the registrar's certificate pinned in a database 📌.
A MASA without any supply-chain integration can simply accept registrars without any authentication, or on a blind TOFU basis as described in Section 7.4.2 🤷♀️.
This document does not make a specific recommendation on how the MASA authenticates the registrar, because the tradeoffs are likely different in different environments and product values 🌸. Even within the ANIMA ACP applicability, there's a huge difference between the supply-chain logistics for $100 CPE devices and $100,000 core routers 💀.
5.5. Registrar Requests Voucher from MASA
When a registrar gets a pledge voucher-request, it in turn submits a registrar voucher-request to the MASA service over an HTTPS interface [RFC7231] 📮.
This is done with an HTTP POST using the operation path value of /.well-known/brski/requestvoucher 🚀.
The voucher media type application/voucher-cms+json is defined in [RFC8366], and it's also used for the registrar voucher-request 🎀. It's a JSON document that has been signed using a CMS structure. The registrar MUST sign the registrar voucher-request ✍️.
MASA implementations SHOULD anticipate future media ntypes, but of course they'll simply fail the request if those types aren't known yet 🔮.
The voucher-request CMS object includes some number of certificates that the MASA takes in as it populates the pinned-domain-cert 📜. Since [RFC8366] is pretty flexible about what can go into pinned-domain-cert, the MASA needs a signal about which certificate should be used to fill it in 🧩. It could be anything from the end-entity certificate the registrar uses all the way up to the entire private Enterprise CA certificate. More-specific certificates mean a tighter binding of the voucher to the domain, while less-specific certificates give more flexibility in how the domain is represented by certificates 💫.
A registrar seeking a nonceless voucher for later offline use benefits from a less-specific certificate, since it lets the actual key pair used by a future registrar be determined by the pinned CA 🌙.
Heads up, bestie: in some cases a less-specific certificate, like a public WebPKI CA, could be too open. It could let any entity issued a certificate by that authority take ownership of a device that has a voucher pinned 😱. Future work may provide a solution to pin both a certificate and a name, which would reduce that risk of malicious ownership assertions 🛡️.
The registrar SHOULD request a voucher with the most specificity consistent with the mode it's operating in 🎯. To do this, when the registrar prepares the CMS structure for the signed voucher-request, it SHOULD include only certificates that are part of the chain it wants the MASA to pin 📌. This MAY be as small as only the end-entity certificate (with id-kp-cmcRA set) that it uses as its TLS server certificate, or it MAY be the entire chain, including the domain CA 🔗.
The registrar SHOULD include an Accept header field (see [RFC7231], Section 5.3.2) listing the response media types it finds acceptable 📋. This list SHOULD be the entire list presented to the registrar in the pledge's original request (see Section 5.2), but it MAY be a subset. The MASA is expected to be flexible in what it accepts 💖.
The registrar fills in the voucher-request fields like this 📝:
created-on: The registrar SHOULD populate this field with the current date and time when the voucher-request is formed. This field gives the MASA additional information ⏰.nonce: This value, if present, is copied from the pledge voucher-request. The registrar voucher-request MAY omit the nonce as per Section 3.1 🙈.serial-number: The serial number of the pledge the registrar would like a voucher for. The registrar works this out by parsing the authenticated pledge IDevID certificate; see Section 2.3 🔎. The registrar MUST verify that theserial-numberfield it parsed matches theserial-numberfield the pledge provided in its voucher-request. This is a sanity check that's useful for catching error conditions and for logging 🧐. The registrar MUST NOT simply copy theserial-numberfield from a pledge voucher-request, because that field is claimed but not certified. 🚫idevid-issuer: The Issuer value from the pledge IDevID certificate is included so the serial-number gets a unique interpretation 🏷️. In the case of a nonceless (offline) voucher-request, an appropriate value needs to be configured from the same out-of-band source as the serial-number 📦.prior-signed-voucher-request: The signed pledge voucher-request SHOULD be included in the registrar voucher-request 🎁. The entire CMS-signed structure is included and base64 encoded for transport in the JSON structure 🧃.
A nonceless registrar voucher-request MAY be submitted to the MASA 🌙. Doing so lets the registrar request a voucher when the pledge is offline, or when the registrar expects it won't be able to connect to the MASA while the pledge is being deployed 🚚. Some use cases require the registrar to learn the right IDevID serialNumber field and the right Accept header field values from the physical device labeling or from the sales channel (which is out of scope for this document) 🏷️.
All other fields MAY be omitted in the registrar voucher-request 🫧.
Important!! The proximity-registrar-cert field MUST NOT be present in the registrar voucher-request 🚫.
See the example JSON payloads of registrar voucher-requests in Section 3.3, Examples 2 through 4 📚.
The MASA verifies that the registrar voucher-request is internally consistent, but it doesn't necessarily authenticate the registrar certificate, since the registrar MAY be unknown to the MASA in advance 🕵️♀️. The MASA performs the actions and validation checks described in the following subsections before issuing a voucher ✅.
5.5.1. MASA Renewal of Expired Vouchers
As described in [RFC8366], vouchers are normally short-lived to avoid revocation issues ⏳. If the request is for a previous (expired) voucher using the same registrar (that is, a registrar with the same domain CA), then the request for a renewed voucher SHOULD be automatically authorized 💅. The MASA has enough information to figure this out by examining the request, the registrar authentication, and the existing audit-log 🧠. Issuing a renewed voucher is logged as detailed in Section 5.6 📝.
If you want the MASA to stop renewing existing vouchers, you can update or revoke the registrar credentials used to authorize the request (see Sections 5.5.4 and 5.5.3) 🔐. More flexible methods will likely involve sales channel integration and authorizations, which are out of scope for this document 🛍️.
5.5.2. MASA Pinning of Registrar
A certificate chain gets extracted from the registrar's signed CMS container 📦. This chain may be as short as a single end-entity certificate, or go all the way up to the entire registrar certificate chain, including the domain CA certificate, as specified in Section 5.5 🌳.
If the domain's CA is unknown to the MASA, then it's treated as a temporary trust anchor for the rest of the steps in this section 🫧. The point isn't to authenticate the message as coming from a fully validated origin. The point is to establish that the domain PKI is consistent 🧩.
The MASA MAY use the certificate in the chain that is farthest from the registrar's end-entity certificate, as determined by MASA policy 🧭. A MASA MAY have a local policy where it only pins the end-entity certificate. This is consistent with [RFC8366] 💜. The details of the policy will typically depend on the degree of supply-chain integration and the mechanism the registrar uses to authenticate. That policy also determines how the MASA responds to a request for a nonceless voucher 🌸.
5.5.3. MASA Check of the Voucher-Request Signature 🔍
Okay bestie, remember how the MASA already pulled out the registrar's domain CA back in Section 5.5.2? 💅 Now it uses that CA to check the CMS signature [RFC5652] on the voucher-request. This is lowkey the big check 🔐
Normal PKIX revocation checking is just assumed to happen during voucher-request signature validation, no cap. This CA certificate MAY have Certificate Revocation List (CRL) distribution points or Online Certificate Status Protocol (OCSP) information [RFC6960]. If they're there, the MASA MUST be able to reach the relevant servers belonging to the registrar's domain CA so it can do the revocation checks. 🌸
The use of OCSP Stapling is preferred. ✨
5.5.4. MASA Verification of the Domain Registrar 🕵️♀️
The MASA MUST verify that the registrar voucher-request is signed by a registrar. ✅ You confirm this by checking that the id-kp-cmcRA extended key usage extension field (as detailed in EST [RFC7030], Section 3.6.1) exists in the certificate of the entity that signed the registrar voucher-request.
This check is only a consistency check, so the MASA can make sure the unauthenticated domain CA meant for the voucher-request signer to be a registrar. It's giving reassurance 🫶 because it helps the domain PKI prove to the domain admin that the MASA will only listen to claims from authorized registration authorities of the domain.
Even when a domain CA is authenticated to the MASA, and there's strong sales channel integration to know who the legit owner is, the above id-kp-cmcRA check stops arbitrary end-entity certificates (such as an LDevID certificate) from getting vouchers issued against them. Not on my watch 💀
Other cases of inappropriate voucher issuance get caught by examining the audit-log. 📓
If a nonceless voucher-request is submitted, the MASA MUST authenticate the registrar. It can do that either as described in EST (see Sections 3.2.3 and 3.3.2 of [RFC7030]) or by validating the registrar's certificate used to sign the registrar voucher-request using a configured trust anchor. Any of these methods reduce the risk of DDoS attacks and give you an authenticated identity to feed into sales channel integration and authorizations (details are out of scope of this document). 🔥
In the nonced case, validating the registrar's identity (via TLS Client Certificate or HTTP authentication) MAY be omitted if the MASA knows the device policy is to accept audit-only vouchers. 🙈
5.5.5. MASA Verification of the Pledge 'prior-signed-voucher-request' 📜
The MASA MAY verify that the registrar voucher-request includes the prior-signed-voucher-request field. 🧐 If it does, the prior-signed-voucher-request MUST include a proximity-registrar-cert that's consistent with the certificate used to sign the registrar voucher-request. On top of that, the voucher-request serial-number leaf MUST match the pledge serial-number that the MASA pulls out of the signing certificate of the prior-signed-voucher-request. Main character energy 👑
The consistency check means checking that the proximity-registrar-cert Subject Public Key Info (SPKI) Fingerprint exists within the registrar voucher-request CMS signature's certificate chain. This is substantially the same as the pin validation described in [RFC7469], Section 2.6. 🦋
If these checks succeed, the MASA updates the voucher and audit-log assertion leafs with the "proximity" assertion, as defined by [RFC8366], Section 5.3. 💖
5.5.6. MASA Nonce Handling 🎟️
The MASA does not verify the nonce itself. 🙅♀️ But if the registrar voucher-request contains a nonce, and the prior-signed-voucher-request exists, then the MASA MUST verify that the nonce is consistent. (Recall from above that the voucher-request might not contain a nonce; see Sections 5.5 and 5.5.4.) 🌈
The MASA fills the audit-log with the nonce that got verified. If a nonceless voucher is issued, the audit-log gets populated with the JSON value "null". 📝
5.6. MASA and Registrar Voucher Response 🎀
The MASA voucher response to the registrar gets forwarded to the pledge without changes, so this section applies to both the MASA and the registrar. The HTTP signaling described applies to both the MASA and registrar responses. 💜
When a voucher-request arrives at the registrar, if it already has a cached response from the MASA for the matching registrar voucher-request, that cached response can be used according to local policy. Otherwise, the registrar builds a new registrar voucher-request and sends it to the MASA. 🧚
Registrar evaluation of the voucher itself is purely for transparency and audit purposes, to help inform log verification (see Section 5.8.3). So a registrar could accept future voucher formats that are opaque to it. 🔍
If the voucher-request is successful, the server response MUST contain an HTTP 200 response code. (The server here is a MASA responding to a registrar, or a registrar responding to a pledge.) The server MUST answer with a suitable 4xx or 5xx HTTP [RFC7230] error code when a problem occurs. In that case, the response data from the MASA MUST be a plain text human-readable (UTF-8) error message that explains why the request was rejected. 🙋♀️
The registrar MAY respond with an HTTP 202 ("the request has been accepted for processing, but the processing has not been completed") as described in EST [RFC7030], Section 4.2.3. When it does, the client "MUST wait at least the specified "retry-after" time before repeating the same request" (also see [RFC7231], Section 6.6.4). ⏳
The pledge is RECOMMENDED to give local feedback during this wait cycle (blinked LED, etc.) if the mechanisms for it are available. 💡 To stop an attacker registrar from significantly delaying bootstrapping, the pledge MUST limit the Retry-After time to 60 seconds. ⏱️
Ideally, the pledge would track the right Retry-After header field values for any number of outstanding registrars, but that would mean a state table on the pledge, and nobody wants that. So instead, the pledge MAY ignore the exact Retry-After value and use a single hard-coded value. (A registrar that can't finish the transaction after the first 60 seconds just gets another chance a minute later.) 🔁
A pledge SHOULD be willing to maintain a 202 retry-state for up to 4 days, which is longer than a long weekend 🏖️. After that, the enrollment attempt fails and the pledge goes back to Discovery state. This gives an alert time to get from the registrar to a human operator, who can decide whether or not to go ahead with the enrollment. 🧑💻
A pledge that retries a request after getting a 202 message MUST resend the same voucher-request. It MUST NOT sign a new voucher-request each time, and especially, it MUST NOT change the nonce value. 🚫
To avoid infinite redirect loops (a malicious registrar could try this to keep the pledge from finding the correct registrar), the pledge MUST NOT follow more than one redirection (3xx code) to another web origin. EST supports redirection but needs user input. This change lets the pledge follow a single redirection without any user interaction. 🌀
A 403 (Forbidden) response is appropriate if the voucher-request isn't signed correctly, is stale, or if the pledge has another outstanding voucher that can't be overridden. 🙅♀️
A 404 (Not Found) response is appropriate when the request is for a device the MASA doesn't know about. 🤷♀️
A 406 (Not Acceptable) response is appropriate if a voucher of the desired type, or one that uses the desired algorithms (as indicated by the "Accept" header fields and the algorithms used in the signature), can't be issued that way because the MASA knows the pledge can't process that type. The registrar SHOULD use this response if it decides the pledge is unacceptable because of inventory control, MASA audit-logs, or any other reason. 🍓
A 415 (Unsupported Media Type) response is appropriate for a request whose voucher-request or "Accept" value isn't understood. 🙃
The voucher response format is whatever's indicated in the submitted "Accept" header fields, or based on the MASA's prior understanding of the proper format for this pledge. Only the "application/voucher-cms+json" media type [RFC8366] is defined at this time. The syntactic details of vouchers are described in detail in [RFC8366]. Figure 14 shows a sample of what's in a voucher. 📦
{
"ietf-voucher:voucher": {
"nonce": "62a2e7693d82fcda2624de58fb6722e5",
"assertion": "logged",
"pinned-domain-cert": "base64encodedvalue==",
"serial-number": "JADA123456789"
}
}
Figure 14: An Example VoucherThe MASA fills in the voucher fields like this:
nonce: The nonce from the pledge if available. See Section 5.5.6. 🔑assertion: The method used to verify the relationship between the pledge and registrar. See Section 5.5.5. 🤝pinned-domain-cert: A certificate; see Section 5.5.2. This figure is illustrative. For an example, see Appendix C.2, where an end-entity certificate is used. 📜serial-number: The serial-number as provided in the voucher-request. Also see Section 5.5.5. 🔢domain-cert-revocation-checks: Set as appropriate for the pledge's capabilities and as documented in [RFC8366]. The MASA MAY set this field to"false", since setting it to"true"would require that revocation information be available to the pledge. This document doesn't make normative requirements for [RFC6961], Section 4.4.2.1 of [RFC8446], or equivalent integrations. 🧩expires-on: This is set for nonceless vouchers. The MASA makes sure the voucher lifetime is consistent with any revocation or pinned-domain-cert consistency checks the pledge might perform. See Section 2.6.1. There are three times to think about: (a) a configured voucher lifetime in the MASA, (b) the expiry time for the registrar's certificate, and (c) any CRL lifetime. Theexpires-onfield SHOULD be before the earliest of these three values. Typically (b) will be some significant time in the future, but (c) will typically be short (on the order of a week or less). The RECOMMENDED period for (a) is on the order of 20 minutes, so it will usually decide how long the resulting voucher lasts. 20 minutes is enough time to reach the post-provisional state in the pledge, which is where there's an established trust relationship between the pledge and registrar. Operations after that can take as long as they need. The lifetime of the voucher has no impact on the life span of the ownership relationship. ⏰
Whenever a voucher is issued, the MASA MUST update the audit-log enough to generate the response described in Section 5.8.1. The internal state requirements for maintaining the audit-log are out of scope. 📓✨
5.6.1. Pledge Voucher Verification ✨
okay bestie, the pledge MUST verify the voucher signature 🔐 using the manufacturer-installed trust anchor(s) tied to the manufacturer's MASA (this is likely baked into the pledge's firmware 💾). managing those trust anchors is out of scope for this doc, and this protocol does not update them. not our problem 🙅♀️
the pledge MUST verify that the serial-number field of the signed voucher matches its own serial-number 🔢
the pledge MUST verify the nonce info in the voucher. if a nonce is present, it has to match the nonce the pledge sent to the registrar 🎟️ but vouchers with no nonce can also be accepted (depending on local policy, see Section 7.2) 🤷♀️
the pledge MUST be prepared to parse and fail gracefully 🫠 from a voucher response that doesn't have a pinned-domain-cert field. that just means it failed to enroll in this domain, so the pledge MUST attempt joining with other available Join Proxies 🏃♀️💨
and the pledge MUST be prepared to ignore any additional fields it doesn't recognize 🙈
5.6.2. Pledge Authentication of Provisional TLS Connection 🦋
following the process in [RFC8366], the pledge should consider the public key from the pinned-domain-cert as the sole temporary trust anchor 🔑
the pledge then checks the TLS server certificate chain it got when the TLS connection was formed, using that trust anchor. it's possible that the public key in the pinned-domain-cert directly matches the public key in the end-entity certificate from the TLS server 👀
if a registrar's credentials can't be verified using the pinned-domain-cert trust anchor from the voucher, the TLS connection is discarded and the pledge gives up on bootstrapping with this discovered registrar 🚫 the pledge SHOULD send voucher status telemetry (described below) before closing the TLS connection. the pledge MUST attempt to enroll using any other proxies it has found. it SHOULD return to the same proxy again after unsuccessful attempts with other proxies. attempts should be made at repeated intervals according to the back-off timer described earlier. attempts SHOULD be repeated, because the failure might just be a temporary inconsistency (an inconsistently rolled registrar key, or some other misconfiguration) 🔄 the inconsistency could also be the result of an active MITM attack on the EST connection 😱
the registrar MUST use a certificate that chains to the pinned-domain-cert as its TLS server certificate 📜
the pledge's PKIX path validation of a registrar certificate's validity period information is as described in Section 2.6.1. once the PKIX path validation is successful, the TLS connection is no longer provisional ✨
the pinned-domain-cert MAY be installed as a trust anchor for future operations such as enrollment (e.g., as recommended per [RFC7030]) or trust anchor management or raw protocols that don't need full PKI-based key management 💅 it can be used to authenticate any dynamically discovered EST server that contains the id-kp-cmcRA extended key usage extension, as detailed in EST (see [RFC7030], Section 3.6.1). but to keep system complexity low, the pledge SHOULD avoid additional discovery operations. instead, the pledge SHOULD communicate directly with the registrar as the EST server 🫶 the pinned-domain-cert is not a complete distribution of the CA certificate response, as described in [RFC7030], Section 4.1.3, which is an extra reason to go ahead with EST key management operations. once a full CA certificate response is obtained, it is more authoritative for the domain than the limited pinned-domain-cert response 👑
5.7. Pledge BRSKI Status Telemetry 📊
the domain is expected to give system administrators indications about device life-cycle status. to make that happen, it needs telemetry info about the device's status 🧚
the pledge MUST indicate its pledge status regarding the voucher. it does this by sending a status message to the registrar 💌
The posted data media type: application/json
the client sends an HTTP POST to the server at the URI .well-known/brski/voucher_status 📮
the format and semantics described below are for version 1. a version field is included so that significant changes to this feedback can happen in the future 🔮 a registrar that receives a status message with a version larger than it knows about SHOULD log the contents and alert a human 🚨
the status field tells you if the voucher was acceptable. boolean values are acceptable, where "true" means the voucher was acceptable ✅
if the voucher was not acceptable, the Reason string says why. in a failure case, this message may be sent to an unauthenticated, potentially malicious registrar, so the Reason string SHOULD NOT provide information beneficial to an attacker 🕵️♀️ the operational benefit of this telemetry info is balanced against the operational costs of not recording that a voucher was ignored by a client the registrar expected would keep joining the domain 🧐
the reason-context attribute is an arbitrary JSON object (a literal value or a hash of values) that gives extra info specific to this pledge 🍓 the contents of this field are not subject to standardization 🤐
the version and status fields MUST be present. the Reason field SHOULD be present whenever the status field is false. the Reason-Context field is optional. in the case of a SUCCESS, the Reason string MAY be omitted 💫
the keys to this JSON object are case sensitive and MUST be lowercase 🔤 Figure 16 shows an example JSON 👇
<CODE BEGINS> file "voucherstatus.cddl"
voucherstatus-post = {
"version": uint,
"status": bool,
? "reason": text,
? "reason-context" : { $$arbitrary-map }
}
}
<CODE ENDS>Figure 15: CDDL for Voucher Status POST
{
"version": 1,
"status":false,
"reason":"Informative human-readable message",
"reason-context": { "additional" : "JSON" }
}Figure 16: Example Status Telemetry
the server SHOULD respond with an HTTP 200 but MAY simply fail with an HTTP 404 error. the client ignores any response. the server SHOULD capture this telemetry info within the server logs 📚
additional standard JSON fields in this POST MAY be added; see Section 8.5. a server that sees unknown fields should log them, but otherwise ignore them 🙃
5.8. Registrar Audit-Log Request 🪵
after receiving the pledge status telemetry (see Section 5.7), the registrar SHOULD request the MASA audit-log from the MASA service 📬
this is done with an HTTP POST using the operation path value of /.well-known/brski/requestauditlog 🎯
the registrar SHOULD HTTP POST the same registrar voucher-request it sent when requesting a voucher (using the same Content-Type). it gets posted to the /requestauditlog URI instead 🔁 the idevid-issuer and serial-number tell the MASA which log is requested, so the right log can be prepared for the response. using the same media type and message keeps the cryptographic and message operations to a minimum, even though it means extra network traffic 🌐 the relying MASA implementation MAY leverage internal state to link this request to the original voucher-request, which was already validated, so it can skip an extra crypto validation 🧠
a registrar MAY request logs at future times. if the registrar generates a new request, the MASA is forced to do the extra cryptographic operations to verify the new request 🥲
a MASA that receives a request for a device that doesn't exist, or for which the requesting owner was never an owner, returns an HTTP 404 ("Not found") code 🙅♀️
it's reasonable for a registrar that the MASA doesn't believe to be the current owner to request the audit-log. there are probably reasons for this that are hard to predict in advance 🤔 for instance, that registrar may not know the device has been resold. or the device may have been resold inappropriately, and this is how the original owner will find out 😤 it's also possible that the device legitimately spends time in two different networks 🌍
rather than returning the audit-log as a response to the POST (with a return code 200), the MASA MAY instead return a 201 ("Created") response ([RFC7231], Sections 6.3.2 and 7.1), with the URL to the prepared (and idempotent, therefore cachable) audit response in the "Location" header field 🎁
in order to avoid enumeration of device audit-logs, a MASA that returns URLs SHOULD take care to make the returned URL unguessable 🕶️ [W3C.capability-urls] provides very good additional guidance. for example, instead of returning URLs with a database number like https://example.com/auditlog/1234, or with the Extended Unique Identifier (EUI) of the device like https://example.com/auditlog/10-00-00-11-22-33, the MASA SHOULD return a randomly generated value (a "slug" in web parlance). the value is used to find the relevant database entry 🔍
a MASA that returns a code 200 MAY also include a "Location" header for future reference by the registrar 📌
5.8.1. MASA Audit-Log Response ✨
okay so bestie, this is the part where the MASA hands over its audit-log like a little diary 📖💅 it returns every log entry tied to the device that showed up with that IDevID in the request 🦋 and the log is allowed to get a little abridged (old or repeated stuff can get trimmed, more on that in a sec) 🙈 the data comes back as JSON [RFC8259] and the Content-Type SHOULD be "application/json" 💖
here's the CDDL [RFC8610] that describes the shape of the audit-log response, so you know exactly what's going on 🎀
<CODE BEGINS> file "auditlog.cddl"
audit-log-response = {
"version": uint,
"events": [ + event ]
"truncation": {
? "nonced duplicates": uint,
? "nonceless duplicates": uint,
? "arbitrary": uint,
}
}
event = {
"date": text,
"domainID": text,
"nonce": text / null,
"assertion": "verified" / "logged" / "proximity",
? "truncated": uint,
}
<CODE ENDS>Figure 17: CDDL for Audit-Log Response
and here's an example, iykyk 🌸
{
"version":"1",
"events":[
{
"date":"2019-05-15T17:25:55.644-04:00",
"domainID":"BduJhdHPpfhQLyponf48JzXSGZ8=",
"nonce":"VOUFT-WwrEv0NuAQEHoV7Q",
"assertion":"proximity",
"truncated":"0"
},
{
"date":"2017-05-15T17:25:55.644-04:00",
"domainID":"BduJhdHPpfhQLyponf48JzXSGZ8=",
"nonce":"f4G6Vi1t8nKo/FieCVgpBg==",
"assertion":"proximity"
}
],
"truncation": {
"nonced duplicates": "0",
"nonceless duplicates": "1",
"arbitrary": "2"
}
}Figure 18: Example of an Audit-Log Response
the domainID is a binary SubjectKeyIdentifier value calculated according to Section 5.8.2 🧚 then it gets base64 encoded once so it can ride along inside this JSON container 🍓
the date is formatted per [RFC3339], which is super consistent with typical JavaScript usage of JSON 💜
the truncation structure MAY be left out if every value is zero 🙈 and any counter missing from it is assumed to be zero ✨
the nonce is a string, the same one that was in the voucher-request, and it's the one used in the voucher 💅 if no nonce ended up in the voucher, then null SHOULD be used instead of just leaving the entry out 🫶 people often make the nonce a base64 random blob, but you should NOT assume that 😭
okay now the bit about big logs 🌈 distribution of a large log is lowkey not ideal, so the structure can be optimized: nonced or nonceless entries for the same domainID MAY be abridged, leaving only the single most recent nonced or nonceless entry for that domainID 💖 when truncation happens, the "event" truncation value SHOULD hold a count of how many events for that domainID got omitted 🎀 the log SHOULD NOT be reduced any further, but sometimes an operational situation means keeping the full log just isn't possible 🥺 in that case the log MAY be arbitrarily abridged for length, and the number of removed entries gets listed as "arbitrary" 🦋
if the truncation count goes over 1024, the MASA MAY use that value without incrementing it any further 💀
a log where duplicates were dropped ("nonced duplicates" and/or "nonceless duplicates") could still be acceptable for informed decisions ✨ a log with "arbitrary" truncations is less acceptable, but manufacturer transparency is still better than hidden truncations 💅
a registrar that sees a version value greater than 1 knows the audit-log got enhanced with extra info 🌈 no information will be removed in future versions 🫶 and if an incompatible change is ever wanted, a new HTTP endpoint will be used 🦋
this document only specifies a simple log format that the MASA gives the registrar 🌸 it could be improved by distributed consensus tech that ties vouchers to things like block-chain, hash trees, or optimized logging approaches 🧚 doing that is out of scope here, but it's an anticipated improvement for future work 💜 so the registrar SHOULD anticipate new kinds of responses, and SHOULD provide operator controls that say how to process unknown responses 💖
5.8.2. Calculation of domainID 🎀
the domainID is a binary value (a BIT STRING) that uniquely identifies a registrar by the pinned-domain-cert 🦋
if the pinned-domain-cert certificate includes the SubjectKeyIdentifier ([RFC5280], Section 4.2.1.2), then that becomes the domainID ✨ if not, the SPKI Fingerprint from [RFC7469], Section 2.4 gets used 💅 both the MASA (to fill in the audit-log) and the registrar (to recognize itself in the audit-log) need to calculate this value 🫶
[RFC5280], Section 4.2.1.2 does not mandate that the SubjectKeyIdentifier extension be present in non-CA certificates 😭 so it is RECOMMENDED that registrar certificates (even self-signed ones) always include the SubjectKeyIdentifier to be used as a domainID 🌈
the domainID gets determined from the certificate chain tied to the pinned-domain-cert, and that's what's used to update the audit-log 🧚
5.8.3. Registrar Audit-Log Verification 🌸
each time the MASA issues a voucher, it adds the details of the assignment to an internal audit-log for that device 💖 the internal audit-log gets processed when responding to requests for details, as described in Section 5.8 🦋 the contents of the audit-log can express a variety of trust levels, and this section explains what kind of trust a registrar can actually derive from the entries 💅
while the audit-log lists the vouchers the MASA issued, those vouchers came from voucher-requests, and it's the content of those voucher-requests that decides how meaningful the audit-log entries are 🎀
a registrar SHOULD use the log information to make an informed decision about continuing to bootstrap the pledge 🫶 the exact policy is out of scope of this document, since it depends on the security requirements inside the registrar domain 🙈 equipment that's bought preowned can be expected to have a pretty extensive history 🍓 the following discussion helps explain the value of each log element 👑
date: the date field gives the registrar a chance to split up the log around known events, like the purchase date 📅 depending on what the registrar or admin knows, events before or after certain dates can matter more or less 💜 for example, equipment that's expected to be brand new and has no history would be a total surprise if it had older entries 😭
domainID: if the log has an unexpected domainID, the pledge could have imprinted on a domain nobody planned for 💀 the registrar can use a bunch of techniques to decide what "unexpected" means, from acceptlists of prior domains all the way to anomaly detection (like "this device was previously bound to a different domain than any other device deployed") ✨ log entries can also be compared against local history logs to hunt for discrepancies (like "this device was re-deployed some number of times internally, but the external audit-log shows extra re-deployments our internal logs don't know about") 🕵️♀️
nonce: nonceless entries mean the logged domainID could theoretically trigger a reset of the pledge and then take over management by using the existing nonceless voucher 😱
assertion: the assertion leaf in the voucher and audit-log tells you why the MASA issued the voucher 🌈 a "verified" entry means the MASA issued the voucher after positively verifying ownership 💅 however, this entry does NOT indicate whether or not the pledge was actually deployed in the prior domain 🙈 a "logged" assertion informs the registrar that the prior vouchers were issued with minimal verification 🥺 a "proximity" assertion assures the registrar that the pledge was truly communicating with the prior domain, and so it gives assurance that the prior domain really did deploy the pledge 🦋
a relatively simple policy is to acceptlist known (internal or external) domainIDs and require all vouchers to have a nonce 🎀 an alternative is to require that all nonceless vouchers come from a subset of domainIDs (for example, only internal ones) 🌸 if the policy gets violated, a simple action is to revoke any locally issued credentials for the pledge in question, or to refuse to forward the voucher 💖 the registrar MUST then refuse any EST actions and SHOULD inform a human via a log 🚨 a registrar MAY be configured to ignore the device's history (overriding the policy above), but it is RECOMMENDED that this only be configured if hardware-assisted (Transport Performance Metrics (TPM) anchored) Network Endpoint Assessment (NEA) [RFC5209] is supported 🔥
5.9. EST Integration for PKI Bootstrapping 🧚
the pledge SHOULD follow the BRSKI operations with EST enrollment operations, including "CA Certificates Request", "CSR Attributes Request", and "Client Certificate Request" or "Server-Side Key Generation", etc. 💅 this is a pretty seamless integration, because BRSKI API calls give an automated alternative to the manual bootstrapping method described in [RFC7030] 🫶 as noted above, using HTTP-persistent connections simplifies the pledge state machine 🌈
although EST lets clients get multiple certificates by sending multiple Certificate Signing Requests (CSRs), BRSKI doesn't support this mechanism directly 🙈 that's because BRSKI pledges MUST use the CSR Attributes request ([RFC7030], Section 4.5) 💖 the registrar MUST validate the CSR against the expected attributes 🎀 this means client requests will "look the same", so even if the client makes multiple requests, they end up producing a single logical certificate 🦋 registrars MAY contain more complex logic, but doing that is out of scope of this specification 🌸 BRSKI doesn't signal any enhancement or restriction to this capability 💜
5.9.1. EST Distribution of CA Certificates 💅
the pledge SHOULD request the full EST Distribution of CA certificate messages; see [RFC7030], Section 4.1 ✨
this makes sure the pledge has the complete set of current CA certificates beyond the pinned-domain-cert (see Section 5.6.2 for a discussion of the limitations of having a single certificate instead of a full CA certificate response) 🫶 although these limitations are fine during initial bootstrapping, they are NOT appropriate for ongoing PKIX end-entity certificate validation 😭🔥
5.9.2. EST CSR Attributes ✨
okay so bestie, automated bootstrapping is supposed to work without anybody hand-configuring the pledge 🫶 which means the pledge might generate a certificate request that only has the identity info it already knows (basically the X.509 IDevID info 🎀), and then it gets back a certificate with the domain-specific identity stuff filled in. In theory the CA gets total control over every field in the end-entity certificate 👑. In reality? Lowkey a mess 💀. PKI CAs are usually hooked up through a bunch of non-standard protocols, and even when everything is standard, the CA is often run by a group that has zero clue what the network services need. The CA could even be outsourced 😭. Not ideal, periodt.
So to fix this, the pledge MUST request the EST "CSR Attributes" from the EST server, and the EST server needs to be able to reply with the attributes the certificate needs for its intended protocols and services ✨. This keeps CA integrations minimal, because the local infrastructure (the EST server) tells the pledge which fields to put in its CSR (like rfc822Name 💅). This is the approach that works best across the widest range of environments, which is so iconic 🌈.
In networks that use the BRSKI enrolled certificate to authenticate the ACP, the EST CSR Attributes MUST include the ACP domain information fields defined in [RFC8994], Section 6.2.2. 🦋
The registrar MUST also confirm that the resulting CSR is formatted as indicated before it forwards the request to a CA 🔥. If the registrar talks to the CA over a protocol like full Certificate Management over CMS (CMC), which has ways to override the CSR Attributes, then those mechanisms MAY be used even if the client ignores the guidance for the CSR Attributes 💜.
5.9.3. EST Client Certificate Request 🌸
The pledge MUST request a new Client Certificate; see [RFC7030], Section 4.2. 🧚
5.9.4. Enrollment Status Telemetry 💖
okay so the admin side that handles bootstrapping also gives system admins info about device life-cycle status 📋. That can include info about bootstrapping messages the client has seen. The MASA keeps logs and the status of credential enrollment 🍓. Since [RFC7030] assumes an end user, there's no final success indication sent back to the server, and honestly that's not enough for automated use cases 😭.
The client MUST send an indicator to the registrar about its enrollment status ✨. It does this with an HTTP POST of a JSON dictionary (with the attributes described below) to the new EST endpoint at "/.well-known/brski/enrollstatus" 🦋.
When it's a successful enrollment, the client SHOULD first re-establish the EST TLS session using the newly obtained credentials 🫶. TLS 1.3 can do this in-band, but TLS 1.2 can't. So the client SHOULD always close the existing TLS connection and start a new one, using the same Join Proxy 🎀.
If enrollment fails, the client MUST send the telemetry over the same TLS connection that was used for the enrollment attempt, with a Reason string saying why the most recent enrollment failed 💀. (For failed attempts, that TLS connection is the most reliable way to match server-side info with what the client sends. Main character correlation energy 👑.)
The version and status fields MUST be present 💅. The Reason field SHOULD be present whenever the status field is false 🙈. In the case of a SUCCESS, the Reason string MAY be omitted ⭐.
The reason-context attribute is an arbitrary JSON object (a literal value or a hash of values) that gives extra info specific to the failure to enroll this pledge 🌈. The contents of this field are not subject to standardization, and it's represented by the group-socket "$$arbitrary-map" in the CDDL 🧚.
<CODE BEGINS> file "enrollstatus.cddl"
enrollstatus-post = {
"version": uint,
"status": bool,
? "reason": text,
? "reason-context" : { $$arbitrary-map }
}
}
<CODE ENDS>Figure 19: CDDL for Enrollment Status POST
Here's an example status report 💜. It's sent with the media type application/json:
{
"version": 1,
"status":true,
"reason":"Informative human readable message",
"reason-context": { "additional" : "JSON" }
}Figure 20: Example of Enrollment Status POST
The server SHOULD respond with an HTTP 200 but MAY simply fail with an HTTP 404 error 🦋.
On the server side, the server MUST capture in its logs whether this message was received over a TLS session with a matching Client Certificate 🔥.
5.9.5. Multiple Certificates 🎀
Pledges that need multiple certificates could set up direct EST connections to the registrar 💖.
5.9.6. EST over CoAP 🌈
This document extends EST so it can bootstrap remote key infrastructures ✨. Bootstrapping matters for CoAP enrollment discussions too 🧚. The definition of EST and BRSKI over CoAP isn't covered here beyond making sure proxies support CoAP operations 💅. Instead, CoAP mappings are expected to be defined in later documents like [ACE-COAP-EST], and CoAP mappings for BRSKI will be discussed there or in future work 🫶.
6. Clarification of Transfer-Encoding 🦋
[RFC7030] defines endpoints to include a "Content-Transfer-Encoding" heading and payloads to be base64-encoded DER [RFC4648] 🌸.
When used within BRSKI, the original EST endpoints stay base64 encoded [RFC7030] (as clarified by [RFC8951]) 💜. But the new BRSKI endpoints that send and receive binary artifacts (specifically, "/.well-known/brski/requestvoucher") are binary, meaning no encoding is used 🔥.
In the BRSKI context, the EST "Content-Transfer-Encoding" header field SHOULD be ignored if it's present 🙈. This header field doesn't need to be included 💅.
7. Reduced Security Operational Modes 🌈
A common bootstrapping requirement is supporting less secure operational modes for support-specific use cases ✨. This section suggests a range of mechanisms that would change the security assurance of BRSKI so it can handle alternative deployment architectures and mitigate the life-cycle management issues identified in Section 10 🧚. They're presented as informative (non-normative) design guidance for future standardization work 🎀. Section 9 gives standardization applicability statements for the ANIMA ACP 👑. Other users would expect that subsets of these mechanisms could be profiled with applicability statements similar to the one in Section 9 💖.
This section is non-normative in the generality of the protocol. Using the mechanisms suggested here MUST be detailed in specific profiles of BRSKI, such as the one in Section 9 🔥.
7.1. Trust Model 🦋
This section explains the trust relationships detailed in Section 2.4 💅:
+--------+ +---------+ +------------+ +------------+
| Pledge | | Join | | Domain | |Manufacturer|
| | | Proxy | | Registrar | | Service |
| | | | | | | (Internet) |
+--------+ +---------+ +------------+ +------------+
Figure 21: Elements of BRSKI Trust ModelPledge: the pledge could be compromised and become an attack vector for malware 💀. The entity is trusted to only imprint using the secure methods described in this document 🫶. Additional endpoint assessment techniques are RECOMMENDED 🌸 but are out of scope for this document 🙈.
Join Proxy: provides proxy functionality but isn't involved in security considerations 🧚.
Registrar: when it's talking to a MASA, the registrar makes all the decisions 👑. For Ownership Audit Vouchers (see [RFC8366]), the registrar gets a chance to accept MASA decisions 💜.
Vendor Service, MASA: this form of manufacturer service is trusted to accurately log all claim attempts and to give authoritative log info to registrars 📋. The MASA doesn't know which devices belong to which domains 🎀. These claims could be strengthened with cryptographic log techniques that give append-only, cryptographically assured, publicly auditable logs 🌈.
Vendor Service, Ownership Validation: this form of manufacturer service is trusted to accurately know which device is owned by which domain 🍓.
7.2. Pledge Security Reductions 💖
Here's a list of alternative behaviors the pledge can be programmed to implement ✨. These behaviors aren't mutually exclusive, and they don't depend on each other 🦋. Some of them enable offline and emergency (touch-based) deployment use cases 🔥. Normative language is used because these behaviors are referenced in later sections in a normative way 💅.
- The pledge MUST accept nonceless vouchers 🫶. This allows a use case where the registrar can't connect to the MASA at deployment time 🌸. Logging and validity periods address the security considerations of supporting these use cases 💜.
- Lots of devices already support "trust on first use" for physical interfaces like console ports 🧚. This document doesn't change that reality. Devices supporting this protocol MUST NOT support "trust on first use" on network interfaces 💀. That's because "trust on first use" over network interfaces would undermine the logging-based security protections this specification provides 😭.
- The pledge MAY have an operational mode where it skips voucher validation one time, for example if a physical button is pressed during bootstrapping 🎀. This can be useful if the manufacturer service is unavailable 🌈. This behavior SHOULD be available via local configuration or physical presence methods (such as a serial/craft console) so new entities can always be deployed even when autonomic methods fail 👑. This allows for unsecured imprint 🙈.
- A craft/serial console could include a command like "est-enroll [2001:db8:0:1]:443" that starts the EST process from the point after the voucher is validated ✨. This process SHOULD include server certificate verification using an on-screen fingerprint 🦋.
It is RECOMMENDED that "trust on first use" or any method of skipping voucher validation (including use of a craft serial console) only be available if hardware-assisted Network Endpoint Assessment (NEA) [RFC5209] is supported 💖. This recommendation makes sure domain network monitoring can detect inappropriate use of offline or emergency deployment procedures when voucher-based bootstrapping isn't used 🔥.
7.3. Registrar Security Reductions ✨
okay so a registrar can choose to accept devices using less secure methods 💅 but they MUST NOT be the default behavior 🙅♀️. These methods might be fine when the threat model says low security is good enough 🌸. That includes situations where the local administrator is the one making the security calls:
- A registrar MAY choose to accept all devices, or all devices of a particular type 🧚. An administrator might pick this when it's operationally a pain to configure the registrar with the unique identifier of every new device they're expecting 😭.
- A registrar MAY choose to accept devices that claim a unique identity without authenticating that claimed identity 👀. This could happen when the pledge doesn't include an X.509 IDevID factory-installed credential. New entities without an X.509 IDevID credential MAY form the request per Section 5.2, using the format per Section 5.5, so the pledge's serial number information gets to the registrar 🎀 (this includes the IDevID AuthorityKeyIdentifier value, which would be statically configured on the pledge). The pledge MAY refuse to provide a TLS Client Certificate, since there isn't one available 🙃. The pledge SHOULD support HTTP-based or certificate-less TLS authentication as described in EST [RFC7030], Section 3.3.2 🫶. A registrar MUST NOT accept unauthenticated new entities unless it has been configured to do so by an administrator who has verified that only expected new entities can communicate with a registrar (presumably via a physically secured perimeter.) 🔐
- A registrar MAY submit a nonceless voucher-request to the MASA service (by not including a nonce in the voucher-request) 🦋. The resulting vouchers can then be stored by the registrar until they're needed during bootstrapping operations. This is for use cases where the target network is protected by an air gap and therefore cannot contact the MASA service during pledge deployment 🧱.
- A registrar MAY ignore unrecognized nonceless log entries 🙈. This could happen when used equipment is purchased with a valid history of being deployed in air gap networks that required offline vouchers.
- A registrar MAY accept voucher formats of future types that cannot be parsed by the registrar 🔮. This reduces the registrar's visibility into the exact voucher contents, but it does not change the protocol operations.
7.4. MASA Security Reductions 🌈
Lower security modes chosen by the MASA service affect all device deployments unless the lower security behavior is tied to specific device identities. The modes described below can be applied to specific devices via knowledge of what devices were sold. They can also be bound to specific customers (independent of the device identity) by authenticating the customer's registrar 🍓.
7.4.1. Issuing Nonceless Vouchers 🍓
A MASA has the option of not including a nonce in the voucher and/or not requiring one to be present in the voucher-request. This results in distribution of a voucher that may never expire and, in effect, makes the specified domain an always trusted entity to the pledge during any subsequent bootstrapping attempts 👑. The log information captures when a nonceless voucher is issued, so the registrar can make appropriate security decisions when a pledge joins the domain. Nonceless vouchers are useful to support use cases where registrars might not be online during actual device deployment.
While a nonceless voucher may include an expiry date, a typical use for a nonceless voucher is for it to be long lived. If the device can be trusted to have an accurate clock (the MASA will know), then a nonceless voucher CAN be issued with a limited lifetime ⏳.
A more typical case for a nonceless voucher is for use with offline onboarding scenarios where it is not possible to pass a fresh voucher-request to the MASA. The use of a long-lived voucher also eliminates concern about the availability of the MASA many years in the future. Thus, many nonceless vouchers will have no expiry dates 🌙.
Thus, the long-lived nonceless voucher does not require proof that the device is online. Issuing such a thing is only accepted when the registrar is authenticated by the MASA and the MASA is authorized to provide this functionality to this customer. The MASA is RECOMMENDED to use this functionality only in concert with an enhanced level of ownership tracking, the details of which are out of scope for this document 🔍.
If the pledge device is known to have a real-time clock that is set from the factory, use of a voucher validity period is RECOMMENDED ⏰.
7.4.2. Trusting Owners on First Use 🕵️
A MASA has the option of not verifying ownership before responding with a voucher. This is expected to be a common operational model, because doing so relieves the manufacturer providing MASA services from having to track ownership during shipping and throughout the supply chain, and it allows for a very low overhead MASA service 📦. A registrar uses the audit-log information as an in-depth defense strategy to ensure that this does not occur unexpectedly (for example, when purchasing new equipment, the registrar would throw an error if any audit-log information is reported) 🚨. The MASA SHOULD verify the prior-signed-voucher-request information for pledges that support that functionality. This provides a proof-of-proximity check that reduces the need for ownership verification. The proof-of-proximity comes from the assumption that the pledge and Join Proxy are on the same link-local connection 🤝.
A MASA that practices TOFU for registrar identity may wish to annotate the origin of the connection by IP address or netblock and restrict future use of that identity from other locations. A MASA that does this SHOULD take care to not create nuisance situations for itself when a customer has multiple registrars or uses outgoing IPv4-to-IPv4 NAT (NAT44) connections that change frequently 😤.
7.4.3. Updating or Extending Voucher Trust Anchors 🔑
This section deals with two problems: a MASA that is no longer available due to a failed business, and a MASA that is uncooperative to a secondary sale 💔.
A manufacturer could offer a management mechanism that allows the list of voucher verification trust anchors to be extended. [YANG-KEYSTORE] describes one such interface that could be implemented using YANG. Pretty much any configuration mechanism used today could be extended to provide the needed additional update. A manufacturer could even decide to install the domain CA trust anchors received during the EST "cacerts" step as voucher verification anchors. Some additional signals will be needed to clearly identify which keys have voucher validation authority from among those signed by the domain CA. This is future work 🛠️.
With the above change to the list of anchors, vouchers can be issued by an alternate MASA. This could be the previous owner (the seller) or some other trusted third party who is mediating the sale. If it is a third party, the seller would need to take steps to introduce the third-party configuration to the device prior to disconnection. The third party (e.g., a wholesaler of used equipment) could, however, use a mechanism described in Section 7.2 to take control of the device after receiving it physically. This would permit the third party to act as the MASA for future onboarding actions. As the IDevID certificate probably cannot be replaced, the new owner's registrar would have to support an override of the MASA URL 🔁.
To be useful for resale or other transfers of ownership, one of two situations will need to occur. The simplest is that the device is not put through any kind of factory default/reset before going through onboarding again. Some other secure, physical signal would be needed to initiate it. This is most suitable for redeploying a device within the same enterprise. This would entail having previous configuration in the system until entirely replaced by the new owner, and it represents some level of risk 😬.
For the second scenario, there would need to be two levels of factory reset. One would take the system back entirely to manufacturer state, including removing any added trust anchors. The other (more commonly used) one would just restore the configuration back to a known default without erasing trust anchors. This weaker factory reset might leave valuable credentials on the device, and this may be unacceptable to some owners 🥺.
As a third option, the manufacturer's trust anchors could be entirely overwritten with local trust anchors. A factory default would never restore those anchors. This option comes with a lot of power, but it's also a lot of responsibility: if access to the private part of the new anchors is lost, the manufacturer may be unable to help 👑⚠️.
8. IANA Considerations 🏷️
Per this document, IANA has completed the following actions.
8.1. The IETF XML Registry
This document registers a URI in the "IETF XML Registry" [RFC3688]. IANA has registered the following:
URI: urn:ietf:params:xml:ns:yang:ietf-voucher-request
Registrant Contact: The ANIMA WG of the IETF.
XML: N/A; the requested URI is an XML namespace.8.2. YANG Module Names Registry
This document registers a YANG module in the "YANG Module Names" registry [RFC6020]. IANA has registered the following:
Name: ietf-voucher-request
Namespace: urn:ietf:params:xml:ns:yang:ietf-voucher-request
Prefix: vch
Reference: RFC 89958.3. BRSKI Well-Known Considerations
8.3.1. BRSKI .well-known Registration
To the "Well-Known URIs" registry at https://www.iana.org/assignments/well-known-uris/, this document registers the well-known name "brski" with the following filled-in template from [RFC8615]:
URI Suffix: brski
Change Controller: IETFIANA has changed the registration of "est" to now only include [RFC7030] and no longer this document. Earlier draft versions of this document used /.well-known/est rather than /.well-known/brski.
8.3.2. BRSKI .well-known Registry
IANA has created a new registry entitled: "BRSKI Well-Known URIs". The registry has three columns: URI, Description, and Reference. New items can be added using the Specification Required [RFC8126] process. The initial contents of this registry are:
+=================+==========================+===========+
| URI | Description | Reference |
+=================+==========================+===========+
| requestvoucher | pledge to registrar, and | RFC 8995 |
| | from registrar to MASA | |
+-----------------+--------------------------+-----------+
| voucher_status | pledge to registrar | RFC 8995 |
+-----------------+--------------------------+-----------+
| requestauditlog | registrar to MASA | RFC 8995 |
+-----------------+--------------------------+-----------+
| enrollstatus | pledge to registrar | RFC 8995 |
+-----------------+--------------------------+-----------+Table 1: BRSKI Well-Known URIs
8.4. PKIX Registry 🎀
okay so IANA did two things here, bestie 💅
First, they registered a number for id-mod-MASAURLExtn2016(96) from the pkix(7) id-mod(0) Registry. Main character energy ✨
Second, they grabbed a number from the id-pe registry (that's the Structure of Management Information (SMI) Security for PKIX Certificate Extension one 🙈) for id-pe-masa-url. The value is 32, so the OID ends up being 1.3.6.1.5.5.7.1.32 🦋
1.3.6.1.5.5.7.1.328.5. Pledge BRSKI Status Telemetry 📡
so IANA made a brand new registry called "BRSKI Parameters" 🌸 and inside it there's a table called "Pledge BRSKI Status Telemetry Attributes".
New items can be added using the Specification Required process. Obsessed with how organized this is 😭
The initial items in that table come from this document (see Section 5.7) and they are:
versionStatusReasonreason-context
8.6. DNS Service Names 🧚
IANA registered two DNS service names, and they're giving matching energy 💜
Service Name: brski-proxy Transport Protocol(s): tcp Assignee: IESG <iesg@ietf.org> Contact: IESG <iesg@ietf.org> Description: The Bootstrapping Remote Secure Key Infrastructure Proxy Reference: RFC 8995
Service Name: brski-registrar Transport Protocol(s): tcp Assignee: IESG <iesg@ietf.org> Contact: IESG <iesg@ietf.org> Description: The Bootstrapping Remote Secure Key Infrastructure Registrar Reference: RFC 8995
8.7. GRASP Objective Names 🌈
okay so IANA registered some GRASP Objective Names 🔥
First up, the value "AN_Proxy" (no quotes needed, bestie) went into the "GRASP Objective Names" table in the GRASP Parameter registry. Its spec is Section 4.1.1 of this document. ✨
Then the value "AN_join_registrar" (also without quotes) got added to that same table. Its spec is Section 4.3 of this document. 👑
9. Applicability to the Autonomic Control Plane (ACP) 🦋
okay this section is about where BRSKI fits, and it's giving "solution to the problem" vibes fr 💖
This document is an answer to the requirements for secure bootstrapping defined in "Using an Autonomic Control Plane for Stable Connectivity of Network Operations, Administration, and Maintenance (OAM)" [RFC8368], "A Reference Model for Autonomic Networking" [RFC8993], and specifically "An Autonomic Control Plane (ACP)" [RFC8994]. Go look at Sections 3.2 ("Secure Bootstrap over an Unconfigured Network") and 6.2 ("ACP Domain, Certificate, and Network") 🧚
The protocol here is appealing for a bunch of non-ANIMA use cases too 🌸 But those will be deployed into environments with different tradeoffs around privacy, security, reliability, and autonomy from manufacturers. So those use cases need their own applicability statements and have to deal with their own privacy and security considerations 🫶
The ACP that BRSKI bootstraps is typically used in medium to large Internet service provider organizations. Enterprises with a lot of Layer 3 router connectivity get major benefits too, especially if they have lots of sites 💅 (A network that's mostly Layer 2 isn't excluded, but the adjacencies the ACP creates and maintains won't match the topology until every device is in the ACP 🙈)
In the ACP, the Join Proxy counts as proximal because pledge-to-Join-Proxy communication happens exclusively over IPv6 link-local addresses. The registrar checks proximity using ANI ACP IPv6 ULAs. ULAs aren't routable over the Internet, so as long as the Join Proxy is operating correctly, the proximity assertion holds 🔥 If other uses of BRSKI use proximity assertions, they'll need similar analysis 💜
As the ANIMA charter puts it, this work "focuses on professionally-managed networks." That kind of network has an operator who can install, configure, and operate the registrar function. The operator makes the purchasing decisions and knows which manufacturers they expect to see on the network 👑
That operator can also bootstrap a device over a serial console (craft console) 🌸 The zero-touch mechanism here and in the ACP document [RFC8994] is a huge efficiency win. In particular, it cuts down on sending senior experts on airplanes to configure devices in person ✈️😭
Technology keeps evolving, but not every situation works out, and sometimes a human still has to visit 🙈 So Section 7.2 presents some mechanisms for that. The manufacturer MUST provide at least one of the one-touch mechanisms described that let enrollment proceed without any manufacturer server (such as the MASA) being available 🫶
BRSKI is heading into environments that already have plenty of vendor proprietary management systems 🌈 Those aren't expected to disappear quickly. Instead, they'll leverage the secure credentials that BRSKI provisions. The connectivity those management systems need is provided by the ACP ✨
9.1. Operational Requirements 📋
This section collects operational requirements for the three roles in BRSKI: the MASA, the (domain) owner, and the device 💖 It's worth knowing that the manufacturer can be involved in two roles, since it creates the software/firmware for the device and may also operate the MASA 🧚
The requirements in this section use BCP 14 language [RFC2119] [RFC8174]. They don't add any new normative statements, they just review a few existing ones in one place, organized by role 💅 They also apply specifically to the ANIMA ACP use case. Other use cases likely have similar requirements, but they MAY have different ones 🦋
9.1.1. MASA Operational Requirements 🏭
The manufacturer MUST arrange for an online service called the MASA to be available. It MUST be available at the URL encoded in the IDevID certificate extensions described in Section 2.3.2 🌸
The online service MUST have access to a private key for signing voucher artifacts [RFC8366]. The public key, certificate, or certificate chain MUST be built into the device as part of the firmware 🔐
It is RECOMMENDED that the manufacturer arrange for this signing key (or keys) to be escrowed according to typical software source code escrow practices [softwareescrow] 🫶
The MASA accepts voucher-requests from domain owners according to an operational practice that fits the device. This can range from any domain owner (first-come first-served, on a TOFU-like basis) to full sales channel integration, where domain owners need to be positively identified by TLS pinned Client Certificates or an HTTP authentication process 👑 The MASA creates signed voucher artifacts according to its internally defined policies 💅
The MASA MUST operate an audit-log for devices that is accessible. The audit-log is designed to be easily cacheable, and the MASA MAY find it useful to put this content on a Content Delivery Network (CDN) 🌐✨
9.1.2. Domain Owner Operational Requirements 🏠
The domain owner MUST operate an EST [RFC7030] server with the extensions described in this document. This is the JRC or registrar 🎀 This JRC/EST server MUST announce itself using GRASP within the ACP 📣 This EST server will typically live in the Network Operations Center for the organization 🖥️
The domain owner MAY operate an internal CA that is separate from the EST server, or it MAY combine everything into a single device. The architecture depends on the scale and resiliency requirements of the organization 🧠 Multiple JRC instances MAY be announced into the ACP from multiple locations to get an appropriate level of redundancy 🌈
To recognize which devices and manufacturers are welcome on the domain owner's network, the domain owner SHOULD maintain an acceptlist of manufacturers. This MAY extend to integration with purchasing departments so they know the serial numbers of devices 📝
The domain owner SHOULD use the resulting overlay ACP network to manage devices, replacing legacy out-of-band mechanisms 🦋
The domain owner SHOULD operate one or more EST servers that can renew the domain certificates (LDevIDs) deployed to devices. These servers MAY be the same as the JRC or MAY be a distinct set of devices, depending on what's best for resiliency 🛡️
The organization MUST take appropriate precautions against loss of access to the CA private key. Hardware security modules and/or secret splitting are appropriate 🔑😭
9.1.3. Device Operational Requirements 🔌
Devices MUST come with built-in trust anchors that let the device validate vouchers from the MASA 💜
Devices MUST come with (unique, per-device) IDevID certificates that include their serial numbers and the MASA URL extension 🪪
Devices are expected to find Join Proxies using GRASP, and then connect to the JRC using the protocol described in this document 🔍
Once a domain owner has been validated with the voucher, devices are expected to enroll into the domain using EST 🎉 Then they're expected to form ACPs using IPsec over IPv6 link-local addresses, as described in [RFC8994] 🧚
Once a device has been enrolled, it SHOULD listen for the address of the JRC using GRASP, and it SHOULD enable itself as a Join Proxy and announce itself on all links/interfaces using GRASP DULL 📢
Devices are expected to renew their certificates before they expire ⏰
10. Privacy Considerations 🙈
10.1. MASA Audit-Log 🕵️♀️
okay privacy time, and this part is important bestie 🫣
The MASA audit-log includes the domainID for each domain a voucher has been issued to. This info is closely related to the actual domain identity 😬 A MASA may need extra defenses against Denial-of-Service attacks (Section 11.1), and that may involve collecting additional information that isn't specified here. That could give the MASA service enough detail to build a pretty detailed picture of the devices provisioned within a domain 💀
There are a few design choices that reduce this risk 💖 The domain can keep some privacy since it hasn't necessarily been authenticated and isn't authoritatively bound to the supply chain 🫶
Also, the domainID only captures the unauthenticated subject key identifier of the domain. A privacy-sensitive domain could theoretically generate a new domainID for each device it deploys 🦋 Similarly, a privacy-sensitive domain would likely buy devices that support proximity assertions from a manufacturer that doesn't require sales channel integration 🛍️ That would give a significant level of privacy while still keeping the security characteristics provided by the registrar-based audit-log inspection 🔐✨
okay bestie, this part is about privacy and the sneaky stuff that leaks out 🦋✨
10.2. What BRSKI-EST Reveals 💅
okay so during the provisional phase of the BRSKI-EST connection between the pledge and the registrar, each party reveals its certificates to the other 🫶 for the pledge, that includes the serialNumber attribute, the MASA URL, and the identity that signed the IDevID certificate.
TLS 1.2 reveals the certificate identities to on-path observers, including the Join Proxy 😭 so anyone sitting on the path can see them. not it.
TLS 1.3 is better, bc it reveals the certificate identities only to the end parties 💜 but the connection is still provisional, so an on-path attacker (MITM) can see the certificates anyway 🙈 and that includes not just malicious attackers but also registrars that are visible to the pledge but are not part of the intended domain. main character villain energy, ngl 💀
the certificate of the registrar is rather arbitrary from the point of view of the BRSKI protocol 🌸 no validations [RFC6125] are expected to be done, so the contents could be easily pseudonymized. any device that can see a Join Proxy would be able to connect to the registrar and learn the identity of the network in question 👀 and even if the contents are pseudonymized, it would still be possible to correlate different connections in different locations that belong to the same entity 🔥 this is unlikely to be a significant privacy concern for ANIMA ACP uses of BRSKI, but it may be a concern to other users of BRSKI 🎀
the certificate of the pledge could be revealed by a malicious Join Proxy that performed a MITM attack on the provisional TLS connection 😱 and that attacker would be able to reveal the identity of the pledge to third parties if it chose to do so. weak. 💅
research into a mechanism to do multistep, multiparty authenticated key agreement, incorporating some kind of zero-knowledge proof, would be valuable ✨ ideally it would avoid disclosing identities until the pledge, registrar, and MASA agree to the transaction. it would also need to discover the location of the MASA without knowing the identity of the pledge or the identity of the MASA 🧚 this part of the problem may be unsolvable 😭 but we can dream, periodt.
10.3. What BRSKI-MASA Reveals to the Manufacturer 🍓
okay so with consumer-oriented devices, the "call-home" mechanism in IoT devices raises significant privacy concerns 🙈 see [livingwithIoT] and [IoTstrangeThings] for exemplars, those are the receipts. the ACP usage of BRSKI is not targeted at individual usage of IoT devices, though. it's aimed at enterprise and ISP creation of networks in a zero-touch fashion, where the "call-home" is a different class of privacy and life-cycle management concern 💜
it needs to be reiterated that the BRSKI-MASA mechanism only occurs once during the commissioning of the device 🎀 it is well defined, and although it's encrypted with TLS, it could in theory be made auditable because the contents are well defined. this connection does not occur when the device powers on or is restarted for normal routines. (it is conceivable, but remarkably unusual, that a device could be forced to go through a full factory reset during an exceptional firmware update situation, after which enrollment would have to be repeated, and a new connection would occur 🧐)
the BRSKI call-home mechanism is mediated via the owner's registrar, and the information that is transmitted is directly auditable by the device owner 🦋 this is in stark contrast to many "call-home" protocols where the device autonomously calls home and uses an undocumented protocol. 😤
while the contents of the signed part of the pledge voucher-request cannot be changed, they are not encrypted at the registrar 👀 the ability to audit the messages by the owner of the network is a mechanism to defend against exfiltration of data by a nefarious pledge 🛡️ both are, to reiterate, encrypted by TLS while in transit 🔒
the BRSKI-MASA exchange reveals the following information to the manufacturer:
- the identity of the device being enrolled. this is revealed by transmission of a signed voucher-request containing the serial-number. the manufacturer can usually link the serial number to a device model 🔍
- an identity of the domain owner in the form of the domain trust anchor. however, this is not a global PKI-anchored name within the WebPKI, so this identity could be pseudonymous. if there is sales channel integration, then the MASA will have authenticated the domain owner, via either a pinned certificate or perhaps another HTTP authentication method, as per Section 5.5.4. 🤝
- the time the device is activated. ⏰
- the IP address of the domain owner's registrar. for ISPs and enterprises, the IP address provides very clear geolocation of the owner 📍 no amount of IP address privacy extensions [RFC8981] can do anything about this, as a simple whois lookup likely identifies the ISP or enterprise from the upper bits anyway. a passive attacker who observes the connection definitely may conclude that the given enterprise/ISP is a customer of the particular equipment vendor 🧾 the precise model that is being enrolled will remain private 🤫
based on the above, the manufacturer is able to track a specific device from pseudonymous domain identity to the next pseudonymous domain identity 🕵️♀️ if there is sales-channel integration, then the identities are not pseudonymous. 💀
the manufacturer knows the IP address of the registrar, but it cannot see the IP address of the device itself 🙌 so it cannot track the device to a detailed physical or network location, only to the location of the registrar. that is likely to be at the enterprise or ISP's headquarters 🏢
the above situation is to be distinguished from a residential/individual person who registers a device from a manufacturer 🏠 individuals do not tend to have multiple offices, and their registrar is likely on the same network as the device. a manufacturer that sells switching/routing products to enterprises should hardly be surprised if additional purchases of switching/routing products are made 🛒 deviations from a historical trend or an established baseline would, however, be notable 📈
the situation is not improved by the enterprise/ISP using anonymization services such as Tor [Dingledine] 🧅 as a TLS 1.2 connection will reveal the ClientCertificate used, clearly identifying the enterprise/ISP involved 😬 TLS 1.3 is better in this regard, but an active attacker can still discover the parties involved by performing a MITM attack on the first attempt (breaking/killing it with a TCP reset (RST)), and then letting subsequent connections pass through 🥷
a manufacturer could attempt to mix the BRSKI-MASA traffic in with general traffic on their site by hosting the MASA behind the same (set) of load balancers that the company's normal marketing site is hosted behind 🧺 this makes a lot of sense from a straight capacity planning point of view, as the same set of services (and the same set of Distributed Denial-of-Service mitigations) may be used 💪 unfortunately, as the BRSKI-MASA connections include TLS ClientCertificate exchanges, this may easily be observed in TLS 1.2, and a traffic analysis may reveal it even in TLS 1.3 😩 this does not make such a plan irrelevant, though. there may be other organizational reasons to keep the marketing site (which is often subject to frequent redesigns, outsourcing, etc.) separate from the MASA, which may need to operate reliably for decades 🏛️
10.4. Manufacturers and Used or Stolen Equipment 🕵️
okay so as explained above, the manufacturer receives information each time a device that is in factory-default mode does a zero-touch bootstrap and attempts to enroll into a domain owner's registrar 📬
the manufacturer is therefore in a position to decline to issue a voucher if it detects that the new owner is not the same as the previous owner 🚫
- this can be seen as a feature if the equipment is believed to have been stolen. if the legitimate owner notifies the manufacturer of the theft, then when the new owner brings the device up, if they use the zero-touch mechanism, the new (illegitimate) owner reveals their location and identity 🚨
- in the case of used equipment, the initial owner could inform the manufacturer of the sale, or the manufacturer may just permit resales unless told otherwise. in which case, the transfer of ownership simply occurs 🤝
- a manufacturer could, however, decide not to issue a new voucher in response to a transfer of ownership. this is essentially the same as the stolen case, with the manufacturer having decided that the sale was not legitimate 🙅♀️
- there is a fourth case, if the manufacturer is providing protection against stolen devices. the manufacturer then has a responsibility to protect the legitimate owner against fraudulent claims that the equipment was stolen. in the absence of such manufacturer protection, such a claim would cause the manufacturer to refuse to issue a new voucher. should the device go through a deep factory reset (for instance, replacement of a damaged main board component), the device would not bootstrap 🥲
- finally, there is a fifth case: the manufacturer has decided to end-of-line the device, or the owner has not paid a yearly support amount, and the manufacturer refuses to issue new vouchers at that point. this last case is not new to the industry: many license systems are already deployed that have a significantly worse effect 😤
this section has outlined five situations in which a manufacturer could use the voucher system to enforce what are clearly license terms 💸 a manufacturer that attempted to enforce license terms via vouchers would find it rather ineffective, as the terms would only be enforced when the device is enrolled, and this is not (to repeat) a daily or even monthly occurrence 🗓️✨
that's the end of this chunk, so keep your eyes peeled for the next one 🌈💖
10.5. Manufacturers and Grey Market Equipment 🛍️
okay so manufacturers sell different products into different regional markets, and it's giving a lot of reasons. Which product shows up where can be driven by price differences, support issues (some markets may need manuals and tech support in the local language 📚), and government export rules, like whether strong crypto is allowed to be exported or used in a particular market 🌍. When a domain owner gets a device from a different market (it can be new!) and moves it somewhere else, that's called a Grey Market 🕵️♀️.
A manufacturer could decide not to issue a voucher to an enterprise/ISP based on where they are, and there are a few ways they could figure that out. They could use the geolocation of the registrar, sales channel knowledge about the customer, or what products are available or unavailable in that market. If the device has a GPS, the coordinates could even get tucked into an extension of the voucher 📍.
The above actions are not illegal, and they're not new either. Many manufacturers have shipped crypto-weak (exportable) versions of firmware as the default on equipment for decades 💅. The first task of an enterprise/ISP has always been to log in to a manufacturer system, show one's "entitlement" (country information, proof that support payments have been made), and then receive either a new updated firmware or a license key that activates the correct firmware.
BRSKI lets all of that happen automatically (in an autonomic fashion), so it perhaps encourages this kind of differentiation by making it cheaper to do 💸.
Here's the tricky part: manufacturers will have to deal with a device shipped to one country under one set of rules (or laws or entitlements), while the domain registry is in another one. Which rules apply is something that will have to be worked out. The manufacturer could believe they're dealing with Grey Market equipment when they're simply dealing with a global enterprise 🌐.
10.6. Some Mitigations for Meddling by Manufacturers 🛡️
The most obvious mitigation is to not buy the product 🙅♀️. Pick manufacturers that are up front about their policies and who don't change them gratuitously 💜.
Section 7.4.3 describes some ways a manufacturer could provide a mechanism to manage the trust anchors and built-in certificates (IDevID) as an extension. There are a variety of mechanisms, and some may take a substantial amount of work to get exactly correct 🧩. These mechanisms don't change the flow of the protocol described here. They just let the starting trust assumptions be changed. This is an area for future standardization work 🔮.
Replacing the voucher validation anchors (usually pointing to the original manufacturer's MASA) with those of the new owner lets the new owner issue vouchers to subsequent owners 🔁. This would be done by having the selling (old) owner run a MASA.
The BRSKI protocol depends on a trust anchor and an identity on the device. Managing these entities makes a few new operational modes possible without any changes to the BRSKI protocol 🎀. Those modes include: offline modes, where the domain owner operates an internal MASA for all devices; resell modes, where the first domain owner becomes the MASA for the next (resold-to) domain owner; and services, where an aggregator acquires a large variety of devices and then acts as a pseudonymized MASA for devices from a variety of manufacturers 🦋.
Although replacing the IDevID isn't required for all the modes described above, a manufacturer could support it. Some may wish to treat replacement of the IDevID as a sign that the device's warranty is terminated. For others, the privacy requirements of some deployments might make this standard operating practice 🙈.
As discussed at the end of Section 5.8.1, new work could be done to use a distributed consensus technology for the audit-log ⛓️. That would keep the audit-log useful even when there's a chain of MASA because of changes of ownership 👑.
10.7. Death of a Manufacturer 💔
A common worry is that a manufacturer could go out of business, leaving device owners unable to get new vouchers for existing products 😭. Those products might already be deployed but need to be reinitialized, used, or kept in a warehouse as long-term spares 📦.
The MASA was named the Manufacturer Authorized Signing Authority to make it clear that the manufacturer itself doesn't have to be the one doing this ✨. Specialist service providers are expected to pop up and handle voucher creation, the same way lots of companies outsource email, advertising, and janitorial services 🧹.
It's also expected that, as part of any service agreement, the manufacturer would arrange to escrow appropriate private keys, so a third party could provide a MASA service. This has routinely been done for source code for decades 🗝️.
11. Security Considerations 🔐
Okay, this part is serious, bestie 🥹. This document describes a bootstrapping protocol that balances operational concerns against security concerns. As detailed in the introduction, and touched on again in Section 7, the protocol allows for reduced security modes. These give more control to the local administrator and owner in cases where less security brings operational benefits. This section goes into more detail about a bunch of specific considerations 🧐.
To make logging and administrative oversight easier, the pledge reports on the voucher parsing status to the registrar, on top of triggering registrar verification of MASA logs 📋. If there's a failure, that information is informative to a potentially malicious registrar 😬. This is mandated anyway, because an informed administrator is worth a lot when a failure points to a real problem. The registrar is RECOMMENDED to verify MASA logs if voucher status telemetry is not received 🕵️.
To support truly limited clients, EST requires that the client MUST support a client authentication model (see [RFC7030], Section 3.3.2). Section 7 updates these requirements by stating that the registrar MAY choose to accept devices that fail cryptographic authentication. This reflects current (poor) practices of shipping devices without a cryptographic identity, which is NOT RECOMMENDED 🙃.
During the provisional period of the connection, the pledge MUST treat all HTTP header and content data as untrusted data 🚫. HTTP libraries are regularly exposed to non-secured HTTP traffic, so mature libraries should not have any problems 🧼.
Pledges might choose to engage in protocol operations with multiple discovered registrars in parallel. As noted above, they will only do so with distinct nonce values. The end result could still be multiple vouchers issued from the MASA if all the registrars try to claim the device. That's not a failure! The pledge picks whichever voucher to accept based on internal logic 💅. The registrars verifying log information will see multiple entries and take that into account for their analytics 📊.
11.1. Denial of Service (DoS) against MASA 🌩️
There are use cases where the MASA could be unavailable or uncooperative to the registrar. These include active DoS attacks, planned and unplanned network partitions, changes to MASA policy, and other situations where MASA policy rejects a claim 🚧. They introduce an operational risk for the registrar owner, because MASA behavior might limit the ability to bootstrap a pledge device. For example, this might be an issue during disaster recovery 🆘.
This risk can be mitigated by registrars that request and maintain long-term copies of "nonceless" vouchers. That way, they're guaranteed to be able to bootstrap their devices 🛟.
Issuing nonceless vouchers creates a security concern of its own 😱. If the registrar of a previous domain can intercept protocol communications, it can use a previously issued nonceless voucher to take management control of a pledge device, even after having sold it. This is mitigated by recording the issuance of such vouchers in the MASA audit-log, which the subsequent registrar verifies, and by pledges only bootstrapping when they're in a factory default state 🔒. It's a balance between enabling MASA independence during future bootstrapping and keeping bootstrapping itself secure ⚖️. Registrar control over requesting and auditing nonceless vouchers lets device owners pick the balance that fits them 🎛️.
The MASA is exposed to DoS attacks where attackers claim an unbounded number of devices. Making sure a registrar is representative of a valid manufacturer customer, even without validating ownership of specific pledge devices, helps mitigate this 🧱. Pledge signatures on the pledge voucher-request, as forwarded by the registrar in the prior-signed-voucher-request field of the registrar voucher-request, significantly reduce this risk. They let the MASA confirm proximity between the pledge and the registrar making the request 📍. Supply-chain integration ("know your customer") is an additional step that MASA providers and device vendors can explore 🤝.
11.2. DomainID Must Be Resistant to Second-Preimage Attacks 🧬
The domainID is used as the reference in the audit-log to the domain 📖. The domainID is expected to be calculated by a hash that is resistant to a second-preimage attack. Such an attack would let a second registrar create fake audit-log entries 🥸.
11.3. Availability of Good Random Numbers 🎲
The nonce used by the pledge in the voucher-request SHOULD be generated by a Strong Cryptographic Sequence ([RFC4086], Section 6.2). TLS has a similar requirement 🔑.
Implementers should pay particular attention to the advice in [RFC4086]; see Sections 3 and, in particular, 3.4 📚. The random seed used by a device at boot MUST be unique across all devices and all bootstraps. Resetting a device to factory default state does not get rid of this requirement 🚨.
11.4. Freshness in Voucher-Requests 💅🎀
Okay so bestie, someone raised a concern that the pledge voucher-request should include some content, a nonce, provided by the registrar and/or MASA, so those two can check that the request is fresh 🧚 fr fr, that's a fair question.
Here's the tea: there are a bunch of operational problems with getting a nonce from the MASA to the pledge 😭. It's kinda easier to grab a random value from the registrar, BUT the registrar isn't vouched for yet, so a nonce from it doesn't really mean much 🙈. Fixing these ops problems also comes with privacy and logistics headaches, so if anyone wants to do it, it has to provide some clear value ✨. This section looks at what happens when the pledge voucher-request isn't fresh.
Because the registrar authenticates the pledge, a full MITM attack isn't possible, even though the pledge does provisional TLS authentication (see Section 5.) 🔒 Instead, we look at a fake registrar (Rm) that talks to the pledge at the same time as, or super close in time to, the intended registrar 👀. (This scenario is intentionally supported as described in Section 4.1.)
Rm can get a voucher signed by the MASA either directly or through arbitrary intermediaries. If the MASA accepts the registrar voucher-request (because either Rm is collaborating with a legit registrar according to supply-chain information, or the MASA is in audit-log only mode), then a voucher linking the pledge to Rm gets issued 💀.
That voucher, when passed back to the pledge, links the pledge to Rm and lets it end the provisional state. Now it trusts Rm, and if it has any security vulnerabilities that an Rm with full administrative control could leverage, it can be treated as a threat to the intended registrar 😱.
This flow is mitigated by the intended registrar verifying the audit-logs available from the MASA, as described in Section 5.8. Rm might chose to collect a voucher-request but wait until after the intended registrar completes the authorization process before submitting it. That pledge voucher-request would be "stale," because it has a nonce that no longer matches the internal state of the pledge 🥀. To successfully use any resulting voucher, Rm would need to remove the stale nonce or anticipate the pledge's future nonce state. Reducing the possibility of this is exactly why the pledge is mandated to generate a strong random or pseudo-random number nonce 💪.
Additionally, to successfully use the resulting voucher, Rm would have to attack the pledge and return it to a bootstrapping-enabled state. That means wiping the pledge of its current configuration and triggering a rebootstrapping of the pledge. This is no more likely than simply taking control of the pledge directly 🤷♀️, but if this is a consideration for you, it is RECOMMENDED that the target network take the following steps:
- Ongoing network monitoring for unexpected bootstrapping attempts by pledges.
- Retrieval and examination of MASA log information upon the occurrence of any such unexpected events. The Rm will be listed in the logs along with nonce information for analysis 🔍.
11.5. Trusting Manufacturers 🏭
The BRSKI extensions to EST permit a new pledge to be completely configured with domain-specific trust anchors ✨. The link from built-in manufacturer-provided trust anchors to domain-specific trust anchors is mediated by the signed voucher artifact 🎟️.
If the manufacturer's IDevID signing key is not properly validated, then there is a risk that the network will accept a pledge that should not be a member of the network 😬. As the address of the manufacturer's MASA is provided in the IDevID using the extension from Section 2.3, the malicious pledge will have no problem collaborating with its MASA to produce a completely valid voucher 💅.
BRSKI does not, however, fundamentally change the trust model from domain owner to manufacturer 🧐. Assuming that the pledge used its IDevID with EST [RFC7030] and BRSKI, the domain (registrar) still needs to trust the manufacturer.
Establishing this trust between domain and manufacturer is outside the scope of BRSKI 🙃. There are a number of mechanisms that can be adopted, including:
- Manually configuring each manufacturer's trust anchor.
- A TOFU mechanism. A human would be queried upon seeing a manufacturer's trust anchor for the first time, and then the trust anchor would be installed to the trusted store. There are risks with this; even if the key to name mapping is validated using something like the WebPKI, there remains the possibility that the name is a look alike: e.g:
dem0.example.vs.demO.example.👀 - Scanning the trust anchor from a QR code that came with the packaging (this is really a manual TOFU mechanism) 📦.
- Some sales integration processing where trust anchors are provided as part of the sales process, probably included in a digital packing "slip", or a sales invoice 🧾.
- Consortium membership, where all manufacturers of a particular device category (e.g, a light bulb or a cable modem) are signed by a CA specifically for this. This is done by CableLabs today. It is used for authentication and authorization as part of [docsisroot] and [TR069] 🤝.
The existing WebPKI provides a reasonable anchor between manufacturer name and public key 🔑. It authenticates the key. It does not provide a reasonable authorization for the manufacturer, so it is not directly usable on its own 🙅♀️.
11.6. Manufacturer Maintenance of Trust Anchors 🗝️
BRSKI depends upon the manufacturer building in trust anchors to the pledge device 🧱. The voucher artifact that is signed by the MASA will be validated by the pledge using that anchor. This implies that the manufacturer needs to maintain access to a signing key that the pledge can validate 🔐.
The manufacturer will need to maintain the ability to make signatures that can be validated for the lifetime that the device could be onboarded ⏳. Whether this onboarding lifetime is less than the device lifetime depends upon how the device is used. An inventory of devices kept in a warehouse as spares might not be onboarded for many decades 📦💤.
There are good cryptographic hygiene reasons why a manufacturer would not want to maintain access to a private key for many decades 🙅. A manufacturer in that situation can leverage a long-term CA anchor, built in to the pledge, and then a certificate chain may be incorporated using the normal CMS certificate set 🪪. This may increase the size of the voucher artifacts, but that is not a significant issue in non-constrained environments 🌈.
There are a few other operational variations that manufacturers could consider ✨. For instance, there is no reason that every device need have the same set of trust anchors preinstalled. Devices built in different factories, or on different days, or in any other consideration, could have different trust anchors built in, and the record of which batch the device is in would be recorded in the asset database. The manufacturer would then know which anchor to sign an artifact against 🧠.
Aside from the concern about long-term access to private keys, a major limiting factor for the shelf life of many devices will be the age of the cryptographic algorithms included 🕰️. A device produced in 2019 will have hardware and software capable of validating algorithms common in 2019 and will have no defense against attacks (both quantum and von Neumann brute-force attacks) that have not yet been invented 💀. This concern is orthogonal to the concern about access to private keys, but this concern likely dominates and limits the life span of a device in a warehouse 📉. If any update to the firmware to support new cryptographic mechanisms were possible (while the device was in a warehouse), updates to trust anchors would also be done at the same time 🔄.
The set of standard operating procedures for maintaining high-value private keys is well documented 📚. For instance, the WebPKI provides a number of options for audits in [cabforumaudit], and the DNSSEC root operations are well documented in [dnssecroot] 🌐.
It is not clear if manufacturers will take this level of precaution, or how strong the economic incentives are to maintain an appropriate level of security 🤔.
The next section examines the risk due to a compromised manufacturer IDevID signing key 🎭. This is followed by examination of the risk due to a compromised MASA key 🔓. The third section below examines the situation where a MASA web server itself is under attacker control, but the MASA signing key itself is safe in a not-directly connected hardware module 🛡️.
11.6.1. Compromise of Manufacturer IDevID Signing Keys 🥷
An attacker that has access to the key that the manufacturer uses to sign IDevID certificates can create counterfeit devices 👻. Such devices can claim to be from a particular manufacturer but can be entirely different devices: Trojan horses in effect 🐴.
As the attacker controls the MASA URL in the certificate, the registrar can be convinced to talk to the attacker's MASA 😱. The registrar does not need to be in any kind of promiscuous mode to be vulnerable 📡.
In addition to creating fake devices, the attacker may also be able to issue revocations for existing certificates if the IDevID certificate process relies upon CRL lists that are distributed 🗑️.
There does not otherwise seem to be any risk from this compromise to devices that are already deployed or that are sitting locally in boxes waiting for deployment (local spares) 📦. The issue is that operators will be unable to trust devices that have been in an uncontrolled warehouse, as they do not know if those are real devices 🫥.
11.6.2. Compromise of MASA Signing Keys 🔑💥
There are two periods of time in which to consider: when the MASA key has fallen into the hands of an attacker and after the MASA recognizes that the key has been compromised ⏰.
11.6.2.1. Attacker Opportunities with a Compromised MASA Key
okay so bestie, an attacker who gets their hands on the MASA signing key 🔑 could create vouchers. These could be for devices that are already deployed in the field, or for devices still chilling in a warehouse 📦. But to actually pull this off, two things need to happen: the device has to go through a factory default boot cycle, and the registrar has to be convinced to contact the attacker's MASA. 😤
If the attacker controls a registrar that is visible to the device, then getting the false voucher delivered is no problem at all 💀. A possible practical example is a data center or an ISP peering point (whether that's a public IX or a private peering point) 🌐. In that kind of setup, there are already cables plugged into the equipment that lead to other devices (the peers at the IX), and the false voucher could come through those links 🔌. The hard part is putting the device through a factory reset. That might happen through social engineering of data center staff 🙈. Most locked cages have ventilation holes, and honestly a long "paperclip" could reach through and press a factory reset button 📎. Once that piece of ISP equipment is compromised, it could be used to compromise the equipment it was connected to (even over long haul links), assuming those pieces of equipment could also be forced through a factory reset ✨.
The scenario above seems pretty unlikely since it needs some element of physical access 🙄. But if there was a remote exploit that didn't cause a direct breach, and instead caused a fault that resulted in a factory reset, that could give the attacker a reasonable path 😬.
Everything above is about ANI uses of BRSKI 🏠. For cases where IEEE 802.11 or 802.15.4 is involved, the need to connect directly to the device goes away, but the need to do a factory reset does not 📡. Physical possession of the device isn't required like before, as long as there's some way to force a factory reset 🔁. With some consumer devices that have low overall implementation quality, end users might already be familiar with having to reset the device regularly 🫠.
The authors are unable to come up with an attack scenario where a compromised voucher signature lets an attacker introduce a compromised pledge into an existing operator's network 🙅♀️. That's because the operator controls the communication between registrar and MASA, and there's no opportunity to slip the fake voucher through that conduit 🔒.
11.6.2.2. Risks after Key Compromise is Known
Once the operator of the MASA realizes the voucher signing key has been compromised, it has to do a few things 🚨.
First, it MUST issue a firmware update to all devices that had that key as a trust anchor, so they will no longer trust vouchers from that key 🛠️. This will affect devices in the field that are operating, but since those devices are in operation and not doing onboarding, this isn't a critical patch 😌.
Devices in boxes (in warehouses) are vulnerable and stay vulnerable until patched 😭. An operator would be prudent to unbox the devices, onboard them in a safe environment, and then perform firmware updates 📦✨. This doesn't have to be done by the end-operator; a distributor that stores the spares could do it 🧚. A recommended practice for high-value devices (which typically have a <4hr service window) may be to validate the device operation on a regular basis anyway 💅.
If the onboarding process includes attestations about firmware versions, then through that process the operator would be advised to upgrade the firmware before going into production 🌸. Unfortunately, this doesn't help against situations where the attacker operates their own registrar (as listed above) 💔.
The need for short-lived vouchers is explained in [RFC8366], Section 6.1 🎀. The nonce guarantees freshness, and the short-lived nature of the voucher means the window to deliver a fake voucher is very short ⏱️. A nonceless, long-lived voucher would be the only option for the attacker, and devices in the warehouse would be vulnerable to that 😰.
A key operational recommendation is for manufacturers to sign nonceless, long-lived vouchers with a different key than the one used to sign short-lived vouchers 🔑💖. That key needs significantly better protection 🛡️. If both keys come from a common trust anchor (the manufacturer's CA), then a compromise of the manufacturer's CA would compromise both keys 😱. Such a compromise of the manufacturer's CA likely compromises all keys outlined in this section 💀.
11.6.3. Compromise of MASA Web Service
An attacker that takes over the MASA web service can cause a bunch of attacks 🕷️. The most obvious one is to take the database listing of customers and devices and sell the data to other attackers, who will now know where to find potentially vulnerable devices 💸.
The second most obvious thing the attacker can do is kill the service, or make it operate unreliably, which leaves customers frustrated 😤. This could seriously affect the ability of customers to deploy new services and would be a significant issue during disaster recovery 🚑.
While compromising the MASA web service may lead to compromising the MASA voucher signing key, if the signing happens offboard (such as in a hardware signing module (HSM)), then the key may well be safe, but control over it resides with the attacker 🔐👀.
Such an attacker can issue vouchers for any device presently in service 📲. That device still needs to be convinced to go through a factory reset process before the attack works 🙃.
If the attacker has access to a key that is trusted for long-lived nonceless vouchers, then they could issue vouchers for devices that aren't in service yet 📦😈. This attack may be very hard to verify, since it would involve doing firmware updates on every device in warehouses (a potentially ruinously expensive process) 💸💸. A manufacturer might be reluctant to admit this possibility 🙈.
11.7. YANG Module Security Considerations
As described in Section 7.4 (Security Considerations) of [RFC8366], the YANG module specified in this document defines the schema for data that is subsequently encapsulated by a CMS signed-data content type, as described in Section 5 of [RFC5652]. So all of the YANG-modeled data is protected from modification 🛡️💜.
The use of YANG to define data structures, via the "yang-data" statement, is relatively new and distinct from the traditional use of YANG to define an API accessed by network management protocols such as NETCONF [RFC6241] and RESTCONF [RFC8040]. For this reason, these guidelines do not follow the template described by Section 3.7 of [RFC8407] 🧚✨.
12. References 📚
12.1. Normative References 📚
okay bestie, these are the docs the spec leans on SO hard that you can't skip them 💅 here's the whole squad, in the same order as the original 🌸
[IDevID]IEEE, "IEEE Standard for Local and metropolitan area networks - Secure Device Identity", IEEE 802.1AR,<https://1.ieee802.org/security/802-1ar>.[ITU.X690]ITU-T, "Information Technology - ASN.1 encoding rules: Specification of Basic Encoding Rules (BER), Canonical Encoding Rules (CER) and Distinguished Encoding Rules (DER)", ITU-T Recommendation X.690, ISO/IEC 8825-1:2015, August 2015,<https://www.itu.int/rec/T-REC-X.690>.[REST]Fielding, R.F., "Architectural Styles and the Design of Network-based Software Architectures", 2000,<http://www.ics.uci.edu/~fielding/pubs/dissertation/fielding_dissertation.pdf>.[RFC2119]Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997,<https://www.rfc-editor.org/info/rfc2119>.[RFC3339]Klyne, G. and C. Newman, "Date and Time on the Internet: Timestamps", RFC 3339, DOI 10.17487/RFC3339, July 2002,<https://www.rfc-editor.org/info/rfc3339>.[RFC3688]Mealling, M., "The IETF XML Registry", BCP 81, RFC 3688, DOI 10.17487/RFC3688, January 2004,<https://www.rfc-editor.org/info/rfc3688>.[RFC3748]Aboba, B., Blunk, L., Vollbrecht, J., Carlson, J., and H. Levkowetz, Ed., "Extensible Authentication Protocol (EAP)", RFC 3748, DOI 10.17487/RFC3748, June 2004,<https://www.rfc-editor.org/info/rfc3748>.[RFC3927]Cheshire, S., Aboba, B., and E. Guttman, "Dynamic Configuration of IPv4 Link-Local Addresses", RFC 3927, DOI 10.17487/RFC3927, May 2005,<https://www.rfc-editor.org/info/rfc3927>.[RFC4086]Eastlake 3rd, D., Schiller, J., and S. Crocker, "Randomness Requirements for Security", BCP 106, RFC 4086, DOI 10.17487/RFC4086, June 2005,<https://www.rfc-editor.org/info/rfc4086>.[RFC4519]Sciberras, A., Ed., "Lightweight Directory Access Protocol (LDAP): Schema for User Applications", RFC 4519, DOI 10.17487/RFC4519, June 2006,<https://www.rfc-editor.org/info/rfc4519>.[RFC4648]Josefsson, S., "The Base16, Base32, and Base64 Data Encodings", RFC 4648, DOI 10.17487/RFC4648, October 2006,<https://www.rfc-editor.org/info/rfc4648>.[RFC4862]Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless Address Autoconfiguration", RFC 4862, DOI 10.17487/RFC4862, September 2007,<https://www.rfc-editor.org/info/rfc4862>.[RFC5272]Schaad, J. and M. Myers, "Certificate Management over CMS (CMC)", RFC 5272, DOI 10.17487/RFC5272, June 2008,<https://www.rfc-editor.org/info/rfc5272>.[RFC5280]Cooper, D., Santesson, S., Farrell, S., Boeyen, S., Housley, R., and W. Polk, "Internet X.509 Public Key Infrastructure Certificate and Certificate Revocation List (CRL) Profile", RFC 5280, DOI 10.17487/RFC5280, May 2008,<https://www.rfc-editor.org/info/rfc5280>.[RFC5652]Housley, R., "Cryptographic Message Syntax (CMS)", STD 70, RFC 5652, DOI 10.17487/RFC5652, September 2009,<https://www.rfc-editor.org/info/rfc5652>.[RFC6020]Bjorklund, M., Ed., "YANG - A Data Modeling Language for the Network Configuration Protocol (NETCONF)", RFC 6020, DOI 10.17487/RFC6020, October 2010,<https://www.rfc-editor.org/info/rfc6020>.[RFC6125]Saint-Andre, P. and J. Hodges, "Representation and Verification of Domain-Based Application Service Identity within Internet Public Key Infrastructure Using X.509 (PKIX) Certificates in the Context of Transport Layer Security (TLS)", RFC 6125, DOI 10.17487/RFC6125, March 2011,<https://www.rfc-editor.org/info/rfc6125>.[RFC6241]Enns, R., Ed., Bjorklund, M., Ed., Schoenwaelder, J., Ed., and A. Bierman, Ed., "Network Configuration Protocol (NETCONF)", RFC 6241, DOI 10.17487/RFC6241, June 2011,<https://www.rfc-editor.org/info/rfc6241>.[RFC6762]Cheshire, S. and M. Krochmal, "Multicast DNS", RFC 6762, DOI 10.17487/RFC6762, February 2013,<https://www.rfc-editor.org/info/rfc6762>.[RFC6763]Cheshire, S. and M. Krochmal, "DNS-Based Service Discovery", RFC 6763, DOI 10.17487/RFC6763, February 2013,<https://www.rfc-editor.org/info/rfc6763>.[RFC7030]Pritikin, M., Ed., Yee, P., Ed., and D. Harkins, Ed., "Enrollment over Secure Transport", RFC 7030, DOI 10.17487/RFC7030, October 2013,<https://www.rfc-editor.org/info/rfc7030>.[RFC7230]Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer Protocol (HTTP/1.1): Message Syntax and Routing", RFC 7230, DOI 10.17487/RFC7230, June 2014,<https://www.rfc-editor.org/info/rfc7230>.[RFC7231]Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content", RFC 7231, DOI 10.17487/RFC7231, June 2014,<https://www.rfc-editor.org/info/rfc7231>.[RFC7469]Evans, C., Palmer, C., and R. Sleevi, "Public Key Pinning Extension for HTTP", RFC 7469, DOI 10.17487/RFC7469, April 2015,<https://www.rfc-editor.org/info/rfc7469>.[RFC7950]Bjorklund, M., Ed., "The YANG 1.1 Data Modeling Language", RFC 7950, DOI 10.17487/RFC7950, August 2016,<https://www.rfc-editor.org/info/rfc7950>.[RFC7951]Lhotka, L., "JSON Encoding of Data Modeled with YANG", RFC 7951, DOI 10.17487/RFC7951, August 2016,<https://www.rfc-editor.org/info/rfc7951>.[RFC8040]Bierman, A., Bjorklund, M., and K. Watsen, "RESTCONF Protocol", RFC 8040, DOI 10.17487/RFC8040, January 2017,<https://www.rfc-editor.org/info/rfc8040>.[RFC8174]Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May 2017,<https://www.rfc-editor.org/info/rfc8174>.[RFC8259]Bray, T., Ed., "The JavaScript Object Notation (JSON) Data Interchange Format", STD 90, RFC 8259, DOI 10.17487/RFC8259, December 2017,<https://www.rfc-editor.org/info/rfc8259>.[RFC8366]Watsen, K., Richardson, M., Pritikin, M., and T. Eckert, "A Voucher Artifact for Bootstrapping Protocols", RFC 8366, DOI 10.17487/RFC8366, May 2018,<https://www.rfc-editor.org/info/rfc8366>.[RFC8368]Eckert, T., Ed. and M. Behringer, "Using an Autonomic Control Plane for Stable Connectivity of Network Operations, Administration, and Maintenance (OAM)", RFC 8368, DOI 10.17487/RFC8368, May 2018,<https://www.rfc-editor.org/info/rfc8368>.[RFC8407]Bierman, A., "Guidelines for Authors and Reviewers of Documents Containing YANG Data Models", BCP 216, RFC 8407, DOI 10.17487/RFC8407, October 2018,<https://www.rfc-editor.org/info/rfc8407>.[RFC8446]Rescorla, E., "The Transport Layer Security (TLS) Protocol Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018,<https://www.rfc-editor.org/info/rfc8446>.[RFC8610]Birkholz, H., Vigano, C., and C. Bormann, "Concise Data Definition Language (CDDL): A Notational Convention to Express Concise Binary Object Representation (CBOR) and JSON Data Structures", RFC 8610, DOI 10.17487/RFC8610, June 2019,<https://www.rfc-editor.org/info/rfc8610>.[RFC8951]Richardson, M., Werner, T., and W. Pan, "Clarification of Enrollment over Secure Transport (EST): Transfer Encodings and ASN.1", RFC 8951, DOI 10.17487/RFC8951, November 2020,<https://www.rfc-editor.org/info/rfc8951>.[RFC8981]Gont, F., Krishnan, S., Narten, T., and R. Draves, "Temporary Address Extensions for Stateless Address Autoconfiguration in IPv6", RFC 8981, DOI 10.17487/RFC8981, February 2021,<https://www.rfc-editor.org/info/rfc8981>.[RFC8990]Bormann, C., Carpenter, B., Ed., and B. Liu, Ed., "GeneRic Autonomic Signaling Protocol (GRASP)", RFC 8990, DOI 10.17487/RFC8990, May 2021,<https://www.rfc-editor.org/rfc/rfc8990>.[RFC8994]Eckert, T., Ed., Behringer, M., Ed., and S. Bjarnason, "An Autonomic Control Plane (ACP)", RFC 8994, DOI 10.17487/RFC8994, May 2021,<https://www.rfc-editor.org/rfc/rfc8994>.
12.2. Informative References 📚✨
Okay, these are the bonus reading picks, the stuff that adds context. Here's the full list, bestie 🫶
- [ACE-COAP-EST] van der Stok, P., Kampanakis, P., Richardson, M., and S. Raza, "EST over secure CoAP (EST-coaps)", Work in Progress, Internet-Draft, draft-ietf-ace-coap-est-18, 6 January 2020, <https://tools.ietf.org/html/draft-ietf-ace-coap-est-18>.
- [ANIMA-CONSTRAINED-VOUCHER] Richardson, M., van der Stok, P., Kampanakis, P., and E. Dijk, "Constrained Voucher Artifacts for Bootstrapping Protocols", Work in Progress, Internet-Draft, draft-ietf-anima-constrained-voucher-10, 21 February 2021, <https://tools.ietf.org/html/draft-ietf-anima-constrained-voucher-10>.
- [ANIMA-STATE] Richardson, M., "Considerations for stateful vs stateless join router in ANIMA bootstrap", Work in Progress, Internet-Draft, draft-richardson-anima-state-for-joinrouter-03, 22 September 2020, <https://tools.ietf.org/html/draft-richardson-anima-state-for-joinrouter-03>.
- [brewski] Urban Dictionary, "brewski", March 2003, <https://www.urbandictionary.com/define.php?term=brewski>.
- [cabforumaudit] CA/Browser Forum, "Information for Auditors and Assessors", August 2019, <https://cabforum.org/information-for-auditors-and-assessors/>.
- [Dingledine] Dingledine, R., Mathewson, N., and P. Syverson, "Tor: The Second-Generation Onion Router", August 2004, <https://svn-archive.torproject.org/svn/projects/design-paper/tor-design.pdf>.
- [dnssecroot] "DNSSEC Practice Statement for the Root Zone ZSK Operator", December 2017, <https://www.iana.org/dnssec/procedures/zsk-operator/dps-zsk-operator-v2.1.pdf>.
- [docsisroot] "CableLabs Digital Certificate Issuance Service", February 2018, <https://www.cablelabs.com/resources/digital-certificate-issuance-service/>.
- [imprinting] Wikipedia, "Imprinting (psychology)", January 2021, <https://en.wikipedia.org/w/index.php?title=Imprinting_(psychology)&=999211441>.
- [IoTstrangeThings] ESET, "IoT of toys stranger than fiction: Cybersecurity and data privacy update", March 2017, <https://www.welivesecurity.com/2017/03/03/internet-of-things-security-privacy-iot-update/>.
- [livingwithIoT] Silicon Republic, "What is it actually like to live in a house filled with IoT devices?", February 2018, <https://www.siliconrepublic.com/machines/iot-smart-devices-reality>.
- [minerva] Richardson, M., "Minerva reference implementation for BRSKI", 2020, <https://minerva.sandelman.ca/>.
- [minervagithub] "ANIMA Minerva toolkit", <https://github.com/ANIMAgus-minerva>.
- [openssl] OpenSSL, "OpenSSL X509 Utility", September 2019, <https://www.openssl.org/docs/man1.1.1/man1/openssl-x509.html/>.
- [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", RFC 2131, DOI 10.17487/RFC2131, March 1997, <https://www.rfc-editor.org/info/rfc2131>.
- [RFC2663] Srisuresh, P. and M. Holdrege, "IP Network Address Translator (NAT) Terminology and Considerations", RFC 2663, DOI 10.17487/RFC2663, August 1999, <https://www.rfc-editor.org/info/rfc2663>.
- [RFC5209] Sangster, P., Khosravi, H., Mani, M., Narayan, K., and J. Tardo, "Network Endpoint Assessment (NEA): Overview and Requirements", RFC 5209, DOI 10.17487/RFC5209, June 2008, <https://www.rfc-editor.org/info/rfc5209>.
- [RFC6960] Santesson, S., Myers, M., Ankney, R., Malpani, A., Galperin, S., and C. Adams, "X.509 Internet Public Key Infrastructure Online Certificate Status Protocol - OCSP", RFC 6960, DOI 10.17487/RFC6960, June 2013, <https://www.rfc-editor.org/info/rfc6960>.
- [RFC6961] Pettersen, Y., "The Transport Layer Security (TLS) Multiple Certificate Status Request Extension", RFC 6961, DOI 10.17487/RFC6961, June 2013, <https://www.rfc-editor.org/info/rfc6961>.
- [RFC7228] Bormann, C., Ersue, M., and A. Keranen, "Terminology for Constrained-Node Networks", RFC 7228, DOI 10.17487/RFC7228, May 2014, <https://www.rfc-editor.org/info/rfc7228>.
- [RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May 2014, <https://www.rfc-editor.org/info/rfc7258>.
- [RFC7435] Dukhovni, V., "Opportunistic Security: Some Protection Most of the Time", RFC 7435, DOI 10.17487/RFC7435, December 2014, <https://www.rfc-editor.org/info/rfc7435>.
- [RFC7575] Behringer, M., Pritikin, M., Bjarnason, S., Clemm, A., Carpenter, B., Jiang, S., and L. Ciavaglia, "Autonomic Networking: Definitions and Design Goals", RFC 7575, DOI 10.17487/RFC7575, June 2015, <https://www.rfc-editor.org/info/rfc7575>.
- [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for Writing an IANA Considerations Section in RFCs", BCP 26, RFC 8126, DOI 10.17487/RFC8126, June 2017, <https://www.rfc-editor.org/info/rfc8126>.
- [RFC8340] Bjorklund, M. and L. Berger, Ed., "YANG Tree Diagrams", BCP 215, RFC 8340, DOI 10.17487/RFC8340, March 2018, <https://www.rfc-editor.org/info/rfc8340>.
- [RFC8615] Nottingham, M., "Well-Known Uniform Resource Identifiers (URIs)", RFC 8615, DOI 10.17487/RFC8615, May 2019, <https://www.rfc-editor.org/info/rfc8615>.
- [RFC8993] Behringer, M., Ed., Carpenter, B., Eckert, T., Ciavaglia, L., and J. Nobre, "A Reference Model for Autonomic Networking", RFC 8993, DOI 10.17487/RFC8993, May 2021, <https://www.rfc-editor.org/info/rfc8993>.
- [slowloris] Wikipedia, "Slowloris (computer security)", January 2021, <https://en.wikipedia.org/w/index.php?title=Slowloris_(computer_security)&oldid=1001473290/>.
- [softwareescrow] Wikipedia, "Source code escrow", March 2020, <https://en.wikipedia.org/w/index.php?title=Source_code_escrow&oldid=948073074>.
- [Stajano99theresurrecting] Stajano, F. and R. Anderson, "The Resurrecting Duckling: Security Issues for Ad-hoc Wireless Networks", 1999, <https://www.cl.cam.ac.uk/~fms27/papers/1999-StajanoAnd-duckling.pdf>.
- [TR069] Broadband Forum, "CPE WAN Management Protocol", TR-069, Issue 1, Amendment 6, March 2018, <https://www.broadband-forum.org/download/TR-069_Amendment-6.pdf>.
- [W3C.capability-urls] Tennison, J., "Good Practices for Capability URLs", W3C First Public Working Draft, World Wide Web Consortium WD WD-capability-urls-20140218, February 2014, <https://www.w3.org/TR/2014/WD-capability-urls>.
- [YANG-KEYSTORE] Watsen, K., "A YANG Data Model for a Keystore", Work in Progress, Internet-Draft, draft-ietf-netconf-keystore-22, 18 May 2021, <https://tools.ietf.org/html/draft-ietf-netconf-keystore-22>.
Appendix A. IPv4 and Non-ANI Operations 🦋
The BRSKI spec in Section 4 is intentionally only about the mechanisms for an IPv6 pledge using link-local addresses, so this appendix is the bonus-track for other environments 🎀 Everything in here is non-normative, so think of it as extensions you can use, not rules.
A.1. IPv4 Link-Local Addresses 🌈
Instead of an IPv6 link-local address, an IPv4 address may be generated using "Dynamic Configuration of IPv4 Link-Local Addresses" [RFC3927].
In the case where an IPv4 link-local address is formed, the bootstrap process would continue, as in an IPv6 case, by looking for a (circuit) proxy. 🔥
A.2. Use of DHCPv4 🧚
The pledge MAY obtain an IP address via DHCP ([RFC2131]. The DHCP-provided parameters for the Domain Name System can be used to perform DNS operations if all local discovery attempts fail. Yes, the MAY is in capitals, and yes, the bracket is left open like that in the original 🙈
Appendix B. mDNS / DNS-SD Proxy Discovery Options 💜
Okay, this is where the discovery vibes get serious 💅
Pledge discovery of the proxy (Section 4.1) MAY be performed with DNS-based Service Discovery [RFC6763] over Multicast DNS [RFC6762] to discover the proxy at _brski-proxy._tcp.local..
Proxy discovery of the registrar (Section 4.3) MAY be performed with DNS-based Service Discovery over Multicast DNS to discover registrars by searching for the service _brski-registrar._tcp.local..
To prevent unacceptable levels of network traffic, when using mDNS, the congestion avoidance mechanisms specified in [RFC6762], Section 7 MUST be followed. The pledge SHOULD listen for an unsolicited broadcast response as described in [RFC6762]. This allows devices to avoid announcing their presence via mDNS broadcasts and instead silently join a network by watching for periodic unsolicited broadcast responses. The quiet-listener energy is so real 🤫🫶
Discovery of the registrar MAY also be performed with DNS-based Service Discovery by searching for the service _brski-registrar._tcp.example.com. In this case, the domain "example.com" is discovered as described in [RFC6763], Section 11 (Appendix A.2 of this document suggests the use of DHCP parameters).
If no local proxy or registrar service is located using the GRASP mechanisms or the above-mentioned DNS-based Service Discovery methods, the pledge MAY contact a well-known manufacturer-provided bootstrapping server by performing a DNS lookup using a well-known URI such as brski-registrar.manufacturer.example.com. The details of the URI are manufacturer specific. Manufacturers that leverage this method on the pledge are responsible for providing the registrar service. Also see Section 2.7. 🍓
The current DNS services returned during each query are maintained until bootstrapping is completed. If bootstrapping fails and the pledge returns to the Discovery state, it picks up where it left off and continues attempting bootstrapping. For example, if the first Multicast DNS _bootstrapks._tcp.local response doesn't work, then the second and third responses are tried. If these fail, the pledge moves on to normal DNS-based Service Discovery. Main character pledge, persistent through every retry 👑✨
Appendix C. Example Vouchers 🎀
Three entities are involved in a voucher: the MASA issues (signs) it, the registrar's public key is mentioned in it, and the pledge validates it. To give you reproducible examples, the public and private keys for an example MASA and registrar are listed first 🔑
The keys come from an open source reference implementation of BRSKI, called "Minerva" [minerva]. It's on GitHub [minervagithub]. The keys presented here are used in the unit and integration tests. The MASA code is called "highway", the registrar code is called "fountain", and the example client is called "reach". Iconic naming, no notes 🌸
The public key components of each are presented as base64 certificates and are decoded by openssl's x509 utility so that the extensions can be seen. This was version 1.1.1c of the library and utility of [openssl]. 🧚💖
C.1. Keys Involved 🔑
okay so the manufacturer has a CA that signs the pledge's IDevID 🎀. Then there's the MASA (Manufacturer Authorized Signing Authority), which signs the vouchers. That MASA certificate has to be distributed to the devices at manufacturing time so vouchers can actually be validated 💅
C.1.1. Manufacturer Certification Authority for IDevID Signatures ✨
this private key is the CA that signs IDevID certificates 🔐
file "vendor.key"
-----BEGIN EC PRIVATE KEY-----
MIGkAgEBBDCAYkoLW1IEA5SKKhMMdkTK7sJxk5ybKqYq9Yr5aR7tNwqXyLGS7z8G
8S4w/UJ58BqgBwYFK4EEACKhZANiAAQu5/yktJbFLjMC87h7b+yTreFuF8GwewKH
L4mS0r0dVAQubqDUQcTrjvpXrXCpTojiLCzgp8fzkcUDkZ9LD/M90LDipiLNIOkP
juF8QkoAbT8pMrY83MS8y76wZ7AalNQ=
-----END EC PRIVATE KEY-----and this public key validates IDevID certificates 🫶
file: examples/vendor.key
file "vendor.cert"
Certificate:
Data:
Version: 3 (0x2)
Serial Number: 1216069925 (0x487bc125)
Signature Algorithm: ecdsa-with-SHA256
Issuer: CN = highway-test.example.com CA
Validity
Not Before: Apr 13 20:34:24 2021 GMT
Not After : Apr 13 20:34:24 2023 GMT
Subject: CN = highway-test.example.com CA
Subject Public Key Info:
Public Key Algorithm: id-ecPublicKey
Public-Key: (384 bit)
pub:
04:2e:e7:fc:a4:b4:96:c5:2e:33:02:f3:b8:7b:6f:
ec:93:ad:e1:6e:17:c1:b0:7b:02:87:2f:89:92:d2:
bd:1d:54:04:2e:6e:a0:d4:41:c4:eb:8e:fa:57:ad:
70:a9:4e:88:e2:2c:2c:e0:a7:c7:f3:91:c5:03:91:
9f:4b:0f:f3:3d:d0:b0:e2:a6:22:cd:20:e9:0f:8e:
e1:7c:42:4a:00:6d:3f:29:32:b6:3c:dc:c4:bc:cb:
be:b0:67:b0:1a:94:d4
ASN1 OID: secp384r1
NIST CURVE: P-384
X509v3 extensions:
X509v3 Basic Constraints: critical
CA:TRUE
X509v3 Key Usage: critical
Certificate Sign, CRL Sign
X509v3 Subject Key Identifier:
5E:0C:A9:52:5A:8C:DF:A9:0F:03:14:E9:96:F1:80:76:
8C:53:8A:08
X509v3 Authority Key Identifier:
keyid:5E:0C:A9:52:5A:8C:DF:A9:0F:03:14:E9:96:F1:
80:76:8C:53:8A:08
Signature Algorithm: ecdsa-with-SHA256
30:64:02:30:60:37:a0:66:89:80:27:e1:0d:e5:43:9a:62:f1:
02:bc:0f:72:6d:a9:e9:cb:84:a5:c6:44:d3:41:9e:5d:ce:7d:
46:16:6e:15:de:f7:cc:e8:3e:61:f9:03:7c:20:c4:b7:02:30:
7f:e9:f3:12:bb:06:c6:24:00:2b:41:aa:21:6b:d8:25:ed:81:
07:11:ef:66:8f:06:bf:c8:be:f0:58:74:24:45:39:4d:04:fc:
31:69:6f:cf:db:fe:61:7b:c3:24:31:ff
-----BEGIN CERTIFICATE-----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-----END CERTIFICATE-----C.1.2. MASA Key Pair for Voucher Signatures 💜
the MASA is the Manufacturer Authorized Signing Authority, and this key pair is what signs vouchers 📜. An example TLS certificate for HTTP authentication isn't included here (see Section 5.4), since it's a super common form 🙈
this private key signs the vouchers that show up below:
file "masa.key"
-----BEGIN EC PRIVATE KEY-----
MHcCAQEEIFhdd0eDdzip67kXx72K+KHGJQYJHNy8pkiLJ6CcvxMGoAoGCCqGSM49
AwEHoUQDQgAEqgQVo0S54kT4yfkbBxumdHOcHrpsqbOpMKmiMln3oB1HAW25MJV+
gqi4tMFfSJ0iEwt8kszfWXK4rLgJS2mnpQ==
-----END EC PRIVATE KEY-----this public key validates vouchers, and the CA above signed it 🔏
file: examples/masa.key
file "masa.cert"
Certificate:
Data:
Version: 3 (0x2)
Serial Number: 193399345 (0xb870a31)
Signature Algorithm: ecdsa-with-SHA256
Issuer: CN = highway-test.example.com CA
Validity
Not Before: Apr 13 21:40:16 2021 GMT
Not After : Apr 13 21:40:16 2023 GMT
Subject: CN = highway-test.example.com MASA
Subject Public Key Info:
Public Key Algorithm: id-ecPublicKey
Public-Key: (256 bit)
pub:
04:aa:04:15:a3:44:b9:e2:44:f8:c9:f9:1b:07:1b:
a6:74:73:9c:1e:ba:6c:a9:b3:a9:30:a9:a2:32:59:
f7:a0:1d:47:01:6d:b9:30:95:7e:82:a8:b8:b4:c1:
5f:48:9d:22:13:0b:7c:92:cc:df:59:72:b8:ac:b8:
09:4b:69:a7:a5
ASN1 OID: prime256v1
NIST CURVE: P-256
X509v3 extensions:
X509v3 Basic Constraints: critical
CA:FALSE
Signature Algorithm: ecdsa-with-SHA256
30:66:02:31:00:ae:cb:61:2d:d4:5c:8d:6e:86:aa:0b:06:1d:
c6:d3:60:ba:32:73:36:25:d3:23:85:49:87:1c:ce:94:23:79:
1a:9e:41:55:24:1d:15:22:a1:48:bb:0a:c0:ab:3c:13:73:02:
31:00:86:3c:67:b3:95:a2:e2:e5:f9:ad:f9:1d:9c:c1:34:32:
78:f5:cf:ea:d5:47:03:9f:00:bf:d0:59:cb:51:c2:98:04:81:
24:8a:51:13:50:b1:75:b2:2f:9d:a8:b4:f4:b9
-----BEGIN CERTIFICATE-----
MIIBcDCB9qADAgECAgQLhwoxMAoGCCqGSM49BAMCMCYxJDAiBgNVBAMMG2hpZ2h3
YXktdGVzdC5leGFtcGxlLmNvbSBDQTAeFw0yMTA0MTMyMTQwMTZaFw0yMzA0MTMy
MTQwMTZaMCgxJjAkBgNVBAMMHWhpZ2h3YXktdGVzdC5leGFtcGxlLmNvbSBNQVNB
MFkwEwYHKoZIzj0CAQYIKoZIzj0DAQcDQgAEqgQVo0S54kT4yfkbBxumdHOcHrps
qbOpMKmiMln3oB1HAW25MJV+gqi4tMFfSJ0iEwt8kszfWXK4rLgJS2mnpaMQMA4w
DAYDVR0TAQH/BAIwADAKBggqhkjOPQQDAgNpADBmAjEArsthLdRcjW6GqgsGHcbT
YLoyczYl0yOFSYcczpQjeRqeQVUkHRUioUi7CsCrPBNzAjEAhjxns5Wi4uX5rfkd
nME0Mnj1z+rVRwOfAL/QWctRwpgEgSSKURNQsXWyL52otPS5
-----END CERTIFICATE-----C.1.3. Registrar Certification Authority 🌸
this CA enrolls the pledge once it's authorized, and it also signs the registrar's certificate 🧚
file "ownerca_secp384r1.key"
-----BEGIN EC PRIVATE KEY-----
MIGkAgEBBDCHnLI0MSOLf8XndiZqoZdqblcPR5YSoPGhPOuFxWy1gFi9HbWv8b/R
EGdRgGEVSjKgBwYFK4EEACKhZANiAAQbf1m6F8MavGaNjGzgw/oxcQ9l9iKRvbdW
gAfb37h6pUVNeYpGlxlZljGxj2l9Mr48yD5bY7VG9qjVb5v5wPPTuRQ/ckdRpHbd
0vC/9cqPMAF/+MJf0/UgA0SLi/IHbLQ=
-----END EC PRIVATE KEY-----the public key gets indicated in a pledge voucher-request to show proximity 👀
file: examples/ownerca_secp384r1.key
file "ownerca_secp384r1.cert"
Certificate:
Data:
Version: 3 (0x2)
Serial Number: 694879833 (0x296b0659)
Signature Algorithm: ecdsa-with-SHA256
Issuer: DC = ca, DC = sandelman,
CN = fountain-test.example.com Unstrung Fountain Root CA
Validity
Not Before: Feb 25 21:31:45 2020 GMT
Not After : Feb 24 21:31:45 2022 GMT
Subject: DC = ca, DC = sandelman,
CN = fountain-test.example.com Unstrung Fountain Root CA
Subject Public Key Info:
Public Key Algorithm: id-ecPublicKey
Public-Key: (384 bit)
pub:
04:1b:7f:59:ba:17:c3:1a:bc:66:8d:8c:6c:e0:c3:
fa:31:71:0f:65:f6:22:91:bd:b7:56:80:07:db:df:
b8:7a:a5:45:4d:79:8a:46:97:19:59:96:31:b1:8f:
69:7d:32:be:3c:c8:3e:5b:63:b5:46:f6:a8:d5:6f:
9b:f9:c0:f3:d3:b9:14:3f:72:47:51:a4:76:dd:d2:
f0:bf:f5:ca:8f:30:01:7f:f8:c2:5f:d3:f5:20:03:
44:8b:8b:f2:07:6c:b4
ASN1 OID: secp384r1
NIST CURVE: P-384
X509v3 extensions:
X509v3 Basic Constraints: critical
CA:TRUE
X509v3 Key Usage: critical
Certificate Sign, CRL Sign
X509v3 Subject Key Identifier:
B9:A5:F6:CB:11:E1:07:A4:49:2C:A7:08:C6:7C:10:BC:
87:B3:74:26
X509v3 Authority Key Identifier:
keyid:B9:A5:F6:CB:11:E1:07:A4:49:2C:A7:08:C6:7C:
10:BC:87:B3:74:26
Signature Algorithm: ecdsa-with-SHA256
30:64:02:30:20:83:06:ce:8d:98:a4:54:7a:66:4c:4a:3a:70:
c2:52:36:5a:52:8d:59:7d:20:9b:2a:69:14:58:87:38:d8:55:
79:dd:fd:29:38:95:1e:91:93:76:b4:f5:66:29:44:b4:02:30:
6f:38:f9:af:12:ed:30:d5:85:29:7c:b1:16:58:bd:67:91:43:
c4:0d:30:f9:d8:1c:ac:2f:06:dd:bc:d5:06:42:2c:84:a2:04:
ea:02:a4:5f:17:51:26:fb:d9:2f:d2:5c
-----BEGIN CERTIFICATE-----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=
-----END CERTIFICATE-----C.1.4. Registrar Key Pair
okay so the registrar is basically the domain owner's rep 💼✨ this key signs the registrar voucher-requests and it's also the thing that terminates the TLS connection from the pledge 🔐
<CODE BEGINS> file "jrc_prime256v1.key"
-----BEGIN EC PRIVATE KEY-----
MHcCAQEEIFZodk+PC5Mu24+ra0sbOjKzan+dW5rvDAR7YuJUOC1YoAoGCCqGSM49
AwEHoUQDQgAElmVQcjS6n+Xd5l/28IFv6UiegQwSBztGj5dkK2MAjQIPV8l8lH+E
jLIOYdbJiI0VtEIf1/Jqt+TOBfinTNOLOg==
-----END EC PRIVATE KEY-----
<CODE ENDS>the public key gets shown in a pledge voucher-request so everyone can tell they're close by 🦋 proximity check, bestie 🫶
<CODE BEGINS> file "jrc_prime256v1.cert"
Certificate:
Data:
Version: 3 (0x2)
Serial Number: 1066965842 (0x3f989b52)
Signature Algorithm: ecdsa-with-SHA256
Issuer: DC = ca, DC = sandelman,
CN = fountain-test.example.com Unstrung Fountain Root CA
Validity
Not Before: Feb 25 21:31:54 2020 GMT
Not After : Feb 24 21:31:54 2022 GMT
Subject: DC = ca, DC = sandelman,
CN = fountain-test.example.com
Subject Public Key Info:
Public Key Algorithm: id-ecPublicKey
Public-Key: (256 bit)
pub:
04:96:65:50:72:34:ba:9f:e5:dd:e6:5f:f6:f0:81:
6f:e9:48:9e:81:0c:12:07:3b:46:8f:97:64:2b:63:
00:8d:02:0f:57:c9:7c:94:7f:84:8c:b2:0e:61:d6:
c9:88:8d:15:b4:42:1f:d7:f2:6a:b7:e4:ce:05:f8:
a7:4c:d3:8b:3a
ASN1 OID: prime256v1
NIST CURVE: P-256
X509v3 extensions:
X509v3 Extended Key Usage: critical
CMC Registration Authority
X509v3 Key Usage: critical
Digital Signature
Signature Algorithm: ecdsa-with-SHA256
30:65:02:30:66:4f:60:4c:55:48:1e:96:07:f8:dd:1f:b9:c8:
12:8d:45:36:87:9b:23:c0:bc:bb:f1:cb:3d:26:15:56:6f:5f:
1f:bf:d5:1c:0e:6a:09:af:1b:76:97:99:19:23:fd:7e:02:31:
00:bc:ac:c3:41:b0:ba:0d:af:52:f9:9c:6e:7a:7f:00:1d:23:
c8:62:01:61:bc:4b:c5:c0:47:99:35:0a:0c:77:61:44:01:4a:
07:52:70:57:00:75:ff:be:07:0e:98:cb:e5
-----BEGIN CERTIFICATE-----
MIIB/DCCAYKgAwIBAgIEP5ibUjAKBggqhkjOPQQDAjBtMRIwEAYKCZImiZPyLGQB
GRYCY2ExGTAXBgoJkiaJk/IsZAEZFglzYW5kZWxtYW4xPDA6BgNVBAMMM2ZvdW50
YWluLXRlc3QuZXhhbXBsZS5jb20gVW5zdHJ1bmcgRm91bnRhaW4gUm9vdCBDQTAe
Fw0yMDAyMjUyMTMxNTRaFw0yMjAyMjQyMTMxNTRaMFMxEjAQBgoJkiaJk/IsZAEZ
FgJjYTEZMBcGCgmSJomT8ixkARkWCXNhbmRlbG1hbjEiMCAGA1UEAwwZZm91bnRh
aW4tdGVzdC5leGFtcGxlLmNvbTBZMBMGByqGSM49AgEGCCqGSM49AwEHA0IABJZl
UHI0up/l3eZf9vCBb+lInoEMEgc7Ro+XZCtjAI0CD1fJfJR/hIyyDmHWyYiNFbRC
H9fyarfkzgX4p0zTizqjKjAoMBYGA1UdJQEB/wQMMAoGCCsGAQUFBwMcMA4GA1Ud
DwEB/wQEAwIHgDAKBggqhkjOPQQDAgNoADBlAjBmT2BMVUgelgf43R+5yBKNRTaH
myPAvLvxyz0mFVZvXx+/1RwOagmvG3aXmRkj/X4CMQC8rMNBsLoNr1L5nG56fwAd
I8hiAWG8S8XAR5k1Cgx3YUQBSgdScFcAdf++Bw6Yy+U=
-----END CERTIFICATE-----
<CODE ENDS>C.1.5. Pledge Key Pair
the pledge comes with its IDevID key pair built in at manufacturing time 🏭💖
<CODE BEGINS> file "idevid_00-D0-E5-F2-00-02.key"
-----BEGIN EC PRIVATE KEY-----
MHcCAQEEIBHNh6r8QRevRuo+tEmBJeFjQKf6bpFA/9NGoltv+9sNoAoGCCqGSM49
AwEHoUQDQgAEA6N1Q4ezfMAKmoecrfb0OBMc1AyEH+BATkF58FsTSyBxs0SbSWLx
FjDOuwB9gLGn2TsTUJumJ6VPw5Z/TP4hJw==
-----END EC PRIVATE KEY-----
<CODE ENDS>this certificate is what the registrar uses to find the MASA 🔍🌸
<CODE BEGINS> file "idevid_00-D0-E5-F2-00-02.cert"
Certificate:
Data:
Version: 3 (0x2)
Serial Number: 521731815 (0x1f18fee7)
Signature Algorithm: ecdsa-with-SHA256
Issuer: CN = highway-test.example.com CA
Validity
Not Before: Apr 27 18:29:30 2021 GMT
Not After : Dec 31 00:00:00 2999 GMT
Subject: serialNumber = 00-D0-E5-F2-00-02
Subject Public Key Info:
Public Key Algorithm: id-ecPublicKey
Public-Key: (256 bit)
pub:
04:03:a3:75:43:87:b3:7c:c0:0a:9a:87:9c:ad:f6:
f4:38:13:1c:d4:0c:84:1f:e0:40:4e:41:79:f0:5b:
13:4b:20:71:b3:44:9b:49:62:f1:16:30:ce:bb:00:
7d:80:b1:a7:d9:3b:13:50:9b:a6:27:a5:4f:c3:96:
7f:4c:fe:21:27
ASN1 OID: prime256v1
NIST CURVE: P-256
X509v3 extensions:
X509v3 Subject Key Identifier:
45:88:CC:96:96:00:64:37:B0:BA:23:65:64:64:54:08:
06:6C:56:AD
X509v3 Basic Constraints:
CA:FALSE
1.3.6.1.5.5.7.1.32:
..highway-test.example.com:9443
Signature Algorithm: ecdsa-with-SHA256
30:65:02:30:62:2a:db:be:34:f7:1b:cb:85:de:26:8e:43:00:
f9:0d:88:c8:77:a8:dd:3c:08:40:54:bc:ec:3d:b6:dc:70:2b:
c3:7f:ca:19:21:9a:a0:ab:c5:51:8e:aa:df:36:de:8b:02:31:
00:b2:5d:59:f8:47:c7:ed:03:97:a8:c0:c7:a8:81:fa:a8:86:
ed:67:64:37:51:7a:6e:9c:a3:82:4d:6d:ad:bc:f3:35:9e:9d:
6a:a2:6d:7f:7f:25:1c:03:ef:f0:ba:9b:71
-----BEGIN CERTIFICATE-----
MIIBrzCCATWgAwIBAgIEHxj+5zAKBggqhkjOPQQDAjAmMSQwIgYDVQQDDBtoaWdo
d2F5LXRlc3QuZXhhbXBsZS5jb20gQ0EwIBcNMjEwNDI3MTgyOTMwWhgPMjk5OTEy
MzEwMDAwMDBaMBwxGjAYBgNVBAUTETAwLUQwLUU1LUYyLTAwLTAyMFkwEwYHKoZI
zj0CAQYIKoZIzj0DAQcDQgAEA6N1Q4ezfMAKmoecrfb0OBMc1AyEH+BATkF58FsT
SyBxs0SbSWLxFjDOuwB9gLGn2TsTUJumJ6VPw5Z/TP4hJ6NZMFcwHQYDVR0OBBYE
FEWIzJaWAGQ3sLojZWRkVAgGbFatMAkGA1UdEwQCMAAwKwYIKwYBBQUHASAEHxYd
aGlnaHdheS10ZXN0LmV4YW1wbGUuY29tOjk0NDMwCgYIKoZIzj0EAwIDaAAwZQIw
YirbvjT3G8uF3iaOQwD5DYjId6jdPAhAVLzsPbbccCvDf8oZIZqgq8VRjqrfNt6L
AjEAsl1Z+EfH7QOXqMDHqIH6qIbtZ2Q3UXpunKOCTW2tvPM1np1qom1/fyUcA+/w
uptx
-----END CERTIFICATE-----
<CODE ENDS>C.2. Example Process
ok the JSON examples below are wrapped at 60 columns 📏 which means there are newlines sprinkled in the strings, and that makes them invalid JSON as-is 🙈 the strings would be super long otherwise, so you have to unwrap them before you process anything 🧚
also heads up: the output of asn1parse has been truncated at 68 columns instead of wrapped, just so it's easier to read 👀💅
C.2.1. Pledge to Registrar ✨
okay so, as we covered in Section 5.2, the pledge signs a pledge voucher-request that carries the registrar's public key in the proximity-registrar-cert field 🔐. The base64 is wrapped at 60 characters so it's easier to read on the page 💅
<CODE BEGINS> file "vr_00-D0-E5-F2-00-02.b64"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<CODE ENDS>here's the ASN.1 breakdown of that artifact 🧐
file: examples/vr_00-D0-E5-F2-00-02.b64
0:d=0 hl=4 l=1648 cons: SEQUENCE
4:d=1 hl=2 l= 9 prim: OBJECT :pkcs7-signedData
15:d=1 hl=4 l=1633 cons: cont [ 0 ]
19:d=2 hl=4 l=1629 cons: SEQUENCE
23:d=3 hl=2 l= 1 prim: INTEGER :01
26:d=3 hl=2 l= 13 cons: SET
28:d=4 hl=2 l= 11 cons: SEQUENCE
30:d=5 hl=2 l= 9 prim: OBJECT :sha256
41:d=3 hl=4 l= 905 cons: SEQUENCE
45:d=4 hl=2 l= 9 prim: OBJECT :pkcs7-data
56:d=4 hl=4 l= 890 cons: cont [ 0 ]
60:d=5 hl=4 l= 886 prim: OCTET STRING :{"ietf-voucher-request:v
950:d=3 hl=4 l= 434 cons: cont [ 0 ]
954:d=4 hl=4 l= 430 cons: SEQUENCE
958:d=5 hl=4 l= 309 cons: SEQUENCE
962:d=6 hl=2 l= 3 cons: cont [ 0 ]
964:d=7 hl=2 l= 1 prim: INTEGER :02
967:d=6 hl=2 l= 4 prim: INTEGER :0D83AFD9
973:d=6 hl=2 l= 10 cons: SEQUENCE
975:d=7 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
985:d=6 hl=2 l= 38 cons: SEQUENCE
987:d=7 hl=2 l= 36 cons: SET
989:d=8 hl=2 l= 34 cons: SEQUENCE
991:d=9 hl=2 l= 3 prim: OBJECT :commonName
996:d=9 hl=2 l= 27 prim: UTF8STRING :highway-test.example.com
1025:d=6 hl=2 l= 32 cons: SEQUENCE
1027:d=7 hl=2 l= 13 prim: UTCTIME :210413203739Z
1042:d=7 hl=2 l= 15 prim: GENERALIZEDTIME :29991231000000Z
1059:d=6 hl=2 l= 28 cons: SEQUENCE
1061:d=7 hl=2 l= 26 cons: SET
1063:d=8 hl=2 l= 24 cons: SEQUENCE
1065:d=9 hl=2 l= 3 prim: OBJECT :serialNumber
1070:d=9 hl=2 l= 17 prim: UTF8STRING :00-D0-E5-F2-00-02
1089:d=6 hl=2 l= 89 cons: SEQUENCE
1091:d=7 hl=2 l= 19 cons: SEQUENCE
1093:d=8 hl=2 l= 7 prim: OBJECT :id-ecPublicKey
1102:d=8 hl=2 l= 8 prim: OBJECT :prime256v1
1112:d=7 hl=2 l= 66 prim: BIT STRING
1180:d=6 hl=2 l= 89 cons: cont [ 3 ]
1182:d=7 hl=2 l= 87 cons: SEQUENCE
1184:d=8 hl=2 l= 29 cons: SEQUENCE
1186:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Subject Key Ident
1191:d=9 hl=2 l= 22 prim: OCTET STRING [HEX DUMP]:04144588CC9696
1215:d=8 hl=2 l= 9 cons: SEQUENCE
1217:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Basic Constraints
1222:d=9 hl=2 l= 2 prim: OCTET STRING [HEX DUMP]:3000
1226:d=8 hl=2 l= 43 cons: SEQUENCE
1228:d=9 hl=2 l= 8 prim: OBJECT :1.3.6.1.5.5.7.1.32
1238:d=9 hl=2 l= 31 prim: OCTET STRING [HEX DUMP]:161D6869676877
1271:d=5 hl=2 l= 10 cons: SEQUENCE
1273:d=6 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
1283:d=5 hl=2 l= 103 prim: BIT STRING
1388:d=3 hl=4 l= 260 cons: SET
1392:d=4 hl=4 l= 256 cons: SEQUENCE
1396:d=5 hl=2 l= 1 prim: INTEGER :01
1399:d=5 hl=2 l= 46 cons: SEQUENCE
1401:d=6 hl=2 l= 38 cons: SEQUENCE
1403:d=7 hl=2 l= 36 cons: SET
1405:d=8 hl=2 l= 34 cons: SEQUENCE
1407:d=9 hl=2 l= 3 prim: OBJECT :commonName
1412:d=9 hl=2 l= 27 prim: UTF8STRING :highway-test.example.com
1441:d=6 hl=2 l= 4 prim: INTEGER :0D83AFD9
1447:d=5 hl=2 l= 11 cons: SEQUENCE
1449:d=6 hl=2 l= 9 prim: OBJECT :sha256
1460:d=5 hl=2 l= 105 cons: cont [ 0 ]
1462:d=6 hl=2 l= 24 cons: SEQUENCE
1464:d=7 hl=2 l= 9 prim: OBJECT :contentType
1475:d=7 hl=2 l= 11 cons: SET
1477:d=8 hl=2 l= 9 prim: OBJECT :pkcs7-data
1488:d=6 hl=2 l= 28 cons: SEQUENCE
1490:d=7 hl=2 l= 9 prim: OBJECT :signingTime
1501:d=7 hl=2 l= 15 cons: SET
1503:d=8 hl=2 l= 13 prim: UTCTIME :210413214323Z
1518:d=6 hl=2 l= 47 cons: SEQUENCE
1520:d=7 hl=2 l= 9 prim: OBJECT :messageDigest
1531:d=7 hl=2 l= 34 cons: SET
1533:d=8 hl=2 l= 32 prim: OCTET STRING [HEX DUMP]:49C21C9889B223
1567:d=5 hl=2 l= 10 cons: SEQUENCE
1569:d=6 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
1579:d=5 hl=2 l= 71 prim: OCTET STRING [HEX DUMP]:3045022100A662and the JSON inside the voucher-request, all wrapped up nice and neat 🎀
{"ietf-voucher-request:voucher":{"assertion":"proximity","cr
eated-on":"2021-04-13T17:43:23.747-04:00","serial-number":"0
0-D0-E5-F2-00-02","nonce":"-_XE9zK9q8Ll1qylMtLKeg","proximit
y-registrar-cert":"MIIB/DCCAYKgAwIBAgIEP5ibUjAKBggqhkjOPQQDA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"}}C.2.2. Registrar to MASA 🚀
okay bestie, as we saw in Section 5.5, the registrar signs its own registrar voucher-request 🫶 and it tucks the pledge's voucher-request inside the prior-signed-voucher-request so the MASA can see the whole chain 🔗✨
<CODE BEGINS> file "parboiled_vr_00-D0-E5-F2-00-02.b64"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<CODE ENDS>Okay so here's the ASN.1 decoding of that artifact 🎀✨ it's the part where the voucher-request gets unpacked into something you can actually read, bestie 🧚
The file is file: examples/parboiled_vr_00_D0-E5-02-00-2D.b64 💖
Okay bestie, here's the signed voucher-request in all its ASN.1 glory 🎀✨
0:d=0 hl=4 l=3939 cons: SEQUENCE
4:d=1 hl=2 l= 9 prim: OBJECT :pkcs7-signedData
15:d=1 hl=4 l=3924 cons: cont [ 0 ]
19:d=2 hl=4 l=3920 cons: SEQUENCE
23:d=3 hl=2 l= 1 prim: INTEGER :01
26:d=3 hl=2 l= 13 cons: SET
28:d=4 hl=2 l= 11 cons: SEQUENCE
30:d=5 hl=2 l= 9 prim: OBJECT :sha256
41:d=3 hl=4 l=2424 cons: SEQUENCE
45:d=4 hl=2 l= 9 prim: OBJECT :pkcs7-data
56:d=4 hl=4 l=2409 cons: cont [ 0 ]
60:d=5 hl=4 l=2405 prim: OCTET STRING :{"ietf-voucher-request:v
2469:d=3 hl=4 l=1135 cons: cont [ 0 ]
2473:d=4 hl=4 l= 508 cons: SEQUENCE
2477:d=5 hl=4 l= 386 cons: SEQUENCE
2481:d=6 hl=2 l= 3 cons: cont [ 0 ]
2483:d=7 hl=2 l= 1 prim: INTEGER :02
2486:d=6 hl=2 l= 4 prim: INTEGER :3F989B52
2492:d=6 hl=2 l= 10 cons: SEQUENCE
2494:d=7 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
2504:d=6 hl=2 l= 109 cons: SEQUENCE
2506:d=7 hl=2 l= 18 cons: SET
2508:d=8 hl=2 l= 16 cons: SEQUENCE
2510:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
2522:d=9 hl=2 l= 2 prim: IA5STRING :ca
2526:d=7 hl=2 l= 25 cons: SET
2528:d=8 hl=2 l= 23 cons: SEQUENCE
2530:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
2542:d=9 hl=2 l= 9 prim: IA5STRING :sandelman
2553:d=7 hl=2 l= 60 cons: SET
2555:d=8 hl=2 l= 58 cons: SEQUENCE
2557:d=9 hl=2 l= 3 prim: OBJECT :commonName
2562:d=9 hl=2 l= 51 prim: UTF8STRING :fountain-test.example.co
2615:d=6 hl=2 l= 30 cons: SEQUENCE
2617:d=7 hl=2 l= 13 prim: UTCTIME :200225213154Z
2632:d=7 hl=2 l= 13 prim: UTCTIME :220224213154Z
2647:d=6 hl=2 l= 83 cons: SEQUENCE
2649:d=7 hl=2 l= 18 cons: SET
2651:d=8 hl=2 l= 16 cons: SEQUENCE
2653:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
2665:d=9 hl=2 l= 2 prim: IA5STRING :ca
2669:d=7 hl=2 l= 25 cons: SET
2671:d=8 hl=2 l= 23 cons: SEQUENCE
2673:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
2685:d=9 hl=2 l= 9 prim: IA5STRING :sandelman
2696:d=7 hl=2 l= 34 cons: SET
2698:d=8 hl=2 l= 32 cons: SEQUENCE
2700:d=9 hl=2 l= 3 prim: OBJECT :commonName
2705:d=9 hl=2 l= 25 prim: UTF8STRING :fountain-test.example.co
2732:d=6 hl=2 l= 89 cons: SEQUENCE
2734:d=7 hl=2 l= 19 cons: SEQUENCE
2736:d=8 hl=2 l= 7 prim: OBJECT :id-ecPublicKey
2745:d=8 hl=2 l= 8 prim: OBJECT :prime256v1
2755:d=7 hl=2 l= 66 prim: BIT STRING
2823:d=6 hl=2 l= 42 cons: cont [ 3 ]
2825:d=7 hl=2 l= 40 cons: SEQUENCE
2827:d=8 hl=2 l= 22 cons: SEQUENCE
2829:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Extended Key Usag
2834:d=9 hl=2 l= 1 prim: BOOLEAN :255
2837:d=9 hl=2 l= 12 prim: OCTET STRING [HEX DUMP]:300A06082B0601
2851:d=8 hl=2 l= 14 cons: SEQUENCE
2853:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Key Usage
2858:d=9 hl=2 l= 1 prim: BOOLEAN :255
2861:d=9 hl=2 l= 4 prim: OCTET STRING [HEX DUMP]:03020780
2867:d=5 hl=2 l= 10 cons: SEQUENCE
2869:d=6 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
2879:d=5 hl=2 l= 104 prim: BIT STRING
2985:d=4 hl=4 l= 619 cons: SEQUENCE
2989:d=5 hl=4 l= 498 cons: SEQUENCE
2993:d=6 hl=2 l= 3 cons: cont [ 0 ]
2995:d=7 hl=2 l= 1 prim: INTEGER :02
2998:d=6 hl=2 l= 4 prim: INTEGER :296B0659
3004:d=6 hl=2 l= 10 cons: SEQUENCE
3006:d=7 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
3016:d=6 hl=2 l= 109 cons: SEQUENCE
3018:d=7 hl=2 l= 18 cons: SET
3020:d=8 hl=2 l= 16 cons: SEQUENCE
3022:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
3034:d=9 hl=2 l= 2 prim: IA5STRING :ca
3038:d=7 hl=2 l= 25 cons: SET
3040:d=8 hl=2 l= 23 cons: SEQUENCE
3042:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
3054:d=9 hl=2 l= 9 prim: IA5STRING :sandelman
3065:d=7 hl=2 l= 60 cons: SET
3067:d=8 hl=2 l= 58 cons: SEQUENCE
3069:d=9 hl=2 l= 3 prim: OBJECT :commonName
3074:d=9 hl=2 l= 51 prim: UTF8STRING :fountain-test.example.co
3127:d=6 hl=2 l= 30 cons: SEQUENCE
3129:d=7 hl=2 l= 13 prim: UTCTIME :200225213145Z
3144:d=7 hl=2 l= 13 prim: UTCTIME :220224213145Z
3159:d=6 hl=2 l= 109 cons: SEQUENCE
3161:d=7 hl=2 l= 18 cons: SET
3163:d=8 hl=2 l= 16 cons: SEQUENCE
3165:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
3177:d=9 hl=2 l= 2 prim: IA5STRING :ca
3181:d=7 hl=2 l= 25 cons: SET
3183:d=8 hl=2 l= 23 cons: SEQUENCE
3185:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
3197:d=9 hl=2 l= 9 prim: IA5STRING :sandelman
3208:d=7 hl=2 l= 60 cons: SET
3210:d=8 hl=2 l= 58 cons: SEQUENCE
3212:d=9 hl=2 l= 3 prim: OBJECT :commonName
3217:d=9 hl=2 l= 51 prim: UTF8STRING :fountain-test.example.co
3270:d=6 hl=2 l= 118 cons: SEQUENCE
3272:d=7 hl=2 l= 16 cons: SEQUENCE
3274:d=8 hl=2 l= 7 prim: OBJECT :id-ecPublicKey
3283:d=8 hl=2 l= 5 prim: OBJECT :secp384r1
3290:d=7 hl=2 l= 98 prim: BIT STRING
3390:d=6 hl=2 l= 99 cons: cont [ 3 ]
3392:d=7 hl=2 l= 97 cons: SEQUENCE
3394:d=8 hl=2 l= 15 cons: SEQUENCE
3396:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Basic Constraints
3401:d=9 hl=2 l= 1 prim: BOOLEAN :255
3404:d=9 hl=2 l= 5 prim: OCTET STRING [HEX DUMP]:30030101FF
3411:d=8 hl=2 l= 14 cons: SEQUENCE
3413:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Key Usage
3418:d=9 hl=2 l= 1 prim: BOOLEAN :255
3421:d=9 hl=2 l= 4 prim: OCTET STRING [HEX DUMP]:03020106
3427:d=8 hl=2 l= 29 cons: SEQUENCE
3429:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Subject Key Ident
3434:d=9 hl=2 l= 22 prim: OCTET STRING [HEX DUMP]:0414B9A5F6CB11
3458:d=8 hl=2 l= 31 cons: SEQUENCE
3460:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Authority Key Ide
3465:d=9 hl=2 l= 24 prim: OCTET STRING [HEX DUMP]:30168014B9A5F6
3491:d=5 hl=2 l= 10 cons: SEQUENCE
3493:d=6 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
3503:d=5 hl=2 l= 103 prim: BIT STRING
3608:d=3 hl=4 l= 331 cons: SET
3612:d=4 hl=4 l= 327 cons: SEQUENCE
3616:d=5 hl=2 l= 1 prim: INTEGER :01
3619:d=5 hl=2 l= 117 cons: SEQUENCE
3621:d=6 hl=2 l= 109 cons: SEQUENCE
3623:d=7 hl=2 l= 18 cons: SET
3625:d=8 hl=2 l= 16 cons: SEQUENCE
3627:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
3639:d=9 hl=2 l= 2 prim: IA5STRING :ca
3643:d=7 hl=2 l= 25 cons: SET
3645:d=8 hl=2 l= 23 cons: SEQUENCE
3647:d=9 hl=2 l= 10 prim: OBJECT :domainComponent
3659:d=9 hl=2 l= 9 prim: IA5STRING :sandelman
3670:d=7 hl=2 l= 60 cons: SET
3672:d=8 hl=2 l= 58 cons: SEQUENCE
3674:d=9 hl=2 l= 3 prim: OBJECT :commonName
3679:d=9 hl=2 l= 51 prim: UTF8STRING :fountain-test.example.co
3732:d=6 hl=2 l= 4 prim: INTEGER :3F989B52
3738:d=5 hl=2 l= 11 cons: SEQUENCE
3740:d=6 hl=2 l= 9 prim: OBJECT :sha256
3751:d=5 hl=2 l= 105 cons: cont [ 0 ]
3753:d=6 hl=2 l= 24 cons: SEQUENCE
3755:d=7 hl=2 l= 9 prim: OBJECT :contentType
3766:d=7 hl=2 l= 11 cons: SET
3768:d=8 hl=2 l= 9 prim: OBJECT :pkcs7-data
3779:d=6 hl=2 l= 28 cons: SEQUENCE
3781:d=7 hl=2 l= 9 prim: OBJECT :signingTime
3792:d=7 hl=2 l= 15 cons: SET
3794:d=8 hl=2 l= 13 prim: UTCTIME :210413214323Z
3809:d=6 hl=2 l= 47 cons: SEQUENCE
3811:d=7 hl=2 l= 9 prim: OBJECT :messageDigest
3822:d=7 hl=2 l= 34 cons: SET
3824:d=8 hl=2 l= 32 prim: OCTET STRING [HEX DUMP]:49CEADD5A3946E
3858:d=5 hl=2 l= 10 cons: SEQUENCE
3860:d=6 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
3870:d=5 hl=2 l= 71 prim: OCTET STRING [HEX DUMP]:3045022100C84ENow the JSON tucked inside the voucher-request, the main event 💜
{"ietf-voucher-request:voucher":{"assertion":"proximity","cr
eated-on":"2021-04-13T21:43:23.787Z","serial-number":"00-D0-
E5-F2-00-02","nonce":"-_XE9zK9q8Ll1qylMtLKeg","prior-signed-
voucher-request":"MIIGcAYJKoZIhvcNAQcCoIIGYTCCBl0CAQExDTALBg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"}}Also, quick heads up 🙈: the previous voucher-request is chilling in the prior-signed-voucher-request attribute 🫶
C.2.3. MASA to Registrar 🎀
okay bestie, this is the moment the MASA hands the voucher back to the registrar 💌 and the registrar is supposed to pass it along to the pledge 🦋 so the whole thing can keep moving ✨
this is the actual voucher artifact, base64 and all 💜
<CODE BEGINS> file "voucher_00-D0-E5-F2-00-02.b64"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<CODE ENDS>Next up is the ASN.1 breakdown of the artifact, which is basically the voucher's internal skeleton laid out line by line 🦴 it's giving very "read the receipts" energy 🧾
file: examples/voucher_00-D0-E5-F2-00-02.b64
0:d=0 hl=4 l=1570 cons: SEQUENCE
4:d=1 hl=2 l= 9 prim: OBJECT :pkcs7-signedData
15:d=1 hl=4 l=1555 cons: cont [ 0 ]
19:d=2 hl=4 l=1551 cons: SEQUENCE
23:d=3 hl=2 l= 1 prim: INTEGER :01
26:d=3 hl=2 l= 13 cons: SET
28:d=4 hl=2 l= 11 cons: SEQUENCE
30:d=5 hl=2 l= 9 prim: OBJECT :sha256
41:d=3 hl=4 l= 888 cons: SEQUENCE
45:d=4 hl=2 l= 9 prim: OBJECT :pkcs7-data
56:d=4 hl=4 l= 873 cons: cont [ 0 ]
60:d=5 hl=4 l= 869 prim: OCTET STRING :{"ietf-voucher:voucher":
933:d=3 hl=4 l= 372 cons: cont [ 0 ]
937:d=4 hl=4 l= 368 cons: SEQUENCE
941:d=5 hl=3 l= 246 cons: SEQUENCE
944:d=6 hl=2 l= 3 cons: cont [ 0 ]
946:d=7 hl=2 l= 1 prim: INTEGER :02
949:d=6 hl=2 l= 4 prim: INTEGER :0B870A31
955:d=6 hl=2 l= 10 cons: SEQUENCE
957:d=7 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
967:d=6 hl=2 l= 38 cons: SEQUENCE
969:d=7 hl=2 l= 36 cons: SET
971:d=8 hl=2 l= 34 cons: SEQUENCE
973:d=9 hl=2 l= 3 prim: OBJECT :commonName
978:d=9 hl=2 l= 27 prim: UTF8STRING :highway-test.example.com
1007:d=6 hl=2 l= 30 cons: SEQUENCE
1009:d=7 hl=2 l= 13 prim: UTCTIME :210413214016Z
1024:d=7 hl=2 l= 13 prim: UTCTIME :230413214016Z
1039:d=6 hl=2 l= 40 cons: SEQUENCE
1041:d=7 hl=2 l= 38 cons: SET
1043:d=8 hl=2 l= 36 cons: SEQUENCE
1045:d=9 hl=2 l= 3 prim: OBJECT :commonName
1050:d=9 hl=2 l= 29 prim: UTF8STRING :highway-test.example.com
1081:d=6 hl=2 l= 89 cons: SEQUENCE
1083:d=7 hl=2 l= 19 cons: SEQUENCE
1085:d=8 hl=2 l= 7 prim: OBJECT :id-ecPublicKey
1094:d=8 hl=2 l= 8 prim: OBJECT :prime256v1
1104:d=7 hl=2 l= 66 prim: BIT STRING
1172:d=6 hl=2 l= 16 cons: cont [ 3 ]
1174:d=7 hl=2 l= 14 cons: SEQUENCE
1176:d=8 hl=2 l= 12 cons: SEQUENCE
1178:d=9 hl=2 l= 3 prim: OBJECT :X509v3 Basic Constraints
1183:d=9 hl=2 l= 1 prim: BOOLEAN :255
1186:d=9 hl=2 l= 2 prim: OCTET STRING [HEX DUMP]:3000
1190:d=5 hl=2 l= 10 cons: SEQUENCE
1192:d=6 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
1202:d=5 hl=2 l= 105 prim: BIT STRING
1309:d=3 hl=4 l= 261 cons: SET
1313:d=4 hl=4 l= 257 cons: SEQUENCE
1317:d=5 hl=2 l= 1 prim: INTEGER :01
1320:d=5 hl=2 l= 46 cons: SEQUENCE
1322:d=6 hl=2 l= 38 cons: SEQUENCE
1324:d=7 hl=2 l= 36 cons: SET
1326:d=8 hl=2 l= 34 cons: SEQUENCE
1328:d=9 hl=2 l= 3 prim: OBJECT :commonName
1333:d=9 hl=2 l= 27 prim: UTF8STRING :highway-test.example.com
1362:d=6 hl=2 l= 4 prim: INTEGER :0B870A31
1368:d=5 hl=2 l= 11 cons: SEQUENCE
1370:d=6 hl=2 l= 9 prim: OBJECT :sha256
1381:d=5 hl=2 l= 105 cons: cont [ 0 ]
1383:d=6 hl=2 l= 24 cons: SEQUENCE
1385:d=7 hl=2 l= 9 prim: OBJECT :contentType
1396:d=7 hl=2 l= 11 cons: SET
1398:d=8 hl=2 l= 9 prim: OBJECT :pkcs7-data
1409:d=6 hl=2 l= 28 cons: SEQUENCE
1411:d=7 hl=2 l= 9 prim: OBJECT :signingTime
1422:d=7 hl=2 l= 15 cons: SET
1424:d=8 hl=2 l= 13 prim: UTCTIME :210413214324Z
1439:d=6 hl=2 l= 47 cons: SEQUENCE
1441:d=7 hl=2 l= 9 prim: OBJECT :messageDigest
1452:d=7 hl=2 l= 34 cons: SET
1454:d=8 hl=2 l= 32 prim: OCTET STRING [HEX DUMP]:55148E0E166153
1488:d=5 hl=2 l= 10 cons: SEQUENCE
1490:d=6 hl=2 l= 8 prim: OBJECT :ecdsa-with-SHA256
1500:d=5 hl=2 l= 72 prim: OCTET STRING [HEX DUMP]:3046022100E854Acknowledgements
We would like to thank the various reviewers for their input, in particular William Atwood, Brian Carpenter, Fuyu Eleven, Eliot Lear, Sergey Kasatkin, Anoop Kumar, Tom Petch, Markus Stenberg, Peter van der Stok, and Thomas Werner. 🫶
Significant reviews were done by Jari Arkko, Christian Huitema, and Russ Housley. 👑
Henk Birkholz contributed the CDDL for the audit-log response. 💅
This document started its life as a two-page idea from Steinthor Bjarnason. 🍓 every big thing starts as a tiny idea, iykyk ✨
In addition, significant review comments were provided by many IESG members, including Adam Roach, Alexey Melnikov, Alissa Cooper, Benjamin Kaduk, Éric Vyncke, Roman Danyliw, and Magnus Westerlund. 🌈
Authors' Addresses
Max Pritikin Cisco
Email: pritikin@cisco.com
Michael C. Richardson Sandelman Software Works
Email: mcr+ietf@sandelman.ca URI: http://www.sandelman.ca/
Toerless Eckert Futurewei Technologies Inc. USA 2330 Central Expy Santa Clara, CA 95050 United States of America
Email: tte+ietf@cs.fau.de
Michael H. Behringer
Email: Michael.H.Behringer@gmail.com
Kent Watsen Watsen Networks
Email: kent+ietf@watsen.net
and that's the whole document, bestie 🦋💖 the end is a full-on glow-up ✨
📦 where this page comes from
- ietf: rfc8995 2021-05
- 💖 girlified by claude-haiku-5-5 on : a language model's rewrite of the original, under its terms