Persistent Symmetric Keys in OpenPGP

Status of This Memo

This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶

This Internet-Draft will expire on 8 January 2024.¶

Copyright Notice

Copyright (c) 2023 IETF Trust and the persons identified as the document authors. All rights reserved.¶

This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶

Table of Contents

1. Introduction

The OpenPGP standard [RFC4880] has supported symmetric encryption for data packets using session keys since its inception, as well as symmetric encryption using password-derived keys. This document extends the use of symmetric cryptography by adding support for persistent symmetric keys which can be stored in a transferable private key, and used to symmetrically encrypt session keys, for long-term storage and archival of messages. This document uses authenticated encryption with associated data (AEAD) as proposed by the OpenPGP crypto refresh [crypto-refresh].¶

The OpenPGP standard also supports the use of digital signatures for authentication and integrity but no similar symmetric mechanism exists in the standard. This document introduces hash-based message authentication codes (HMAC) as a symmetric counterpart to digital signatures, for long-term storage and archival of attestations of authenticity and certification.¶

Rather than introducing new packets for storing persistent symmetric keys, the existing Secret-Key packets are reused for this purpose. To indicate the type of keys, two "Public-Key Algorithms" (AEAD and HMAC) are registered, with IDs 64 and 65.¶

Similarly, we reuse the Signature packet for "symmetric signatures". For session keys encrypted with persistent symmetric keys, while a "Symmetric-Key Encrypted Session Key packet" exists, its semantics don't match our goals, as it's intended to encrypt the session key with a user-provided password, and doesn't offer a way to store a reference to a persistent key. Therefore, we reuse the "Public-Key Encrypted Session Key packet" instead, which does offer the desired semantics. Nevertheless, given this usage, the naming of these packets may be confusing, so we propose to rename them to "Password Encrypted Session Key packet" and "Key Encrypted Session Key packet", instead.¶

2. Conventions Used in This Document

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119]. Any implementation that adheres to the format and methods specified in this document is called a compliant application. Compliant applications are a subset of the broader set of OpenPGP applications described in [RFC4880] and the OpenPGP crypto refresh [crypto-refresh]. Any [RFC2119] keyword within this document applies to compliant applications only.¶

3. Motivation

When compared to asymmetric cryptography, symmetric cryptography can provide improved performance and equivalent security with smaller keys. In contexts that do not require asymmetric cryptography, such as secure data storage where the same user encrypts and decrypts data, symmetric cryptography can be used to take advantage of these benefits.¶

Additionally, asymmetric algorithms included in OpenPGP are vulnerable to attacks that might become possible on quantum computers [Shor]. Symmetric cryptography is also affected by quantum computing but to a lesser extent, which can be countered by using larger keys [Grover]. While the standardization of quantum-secure asymmetric cryptography in OpenPGP is ongoing [PQCinOpenPGP], and will be required to secure communications, there is a large body of existing messages encrypted with classical algorithms. Once persistent symmetric keys are available, these messages can be protected against future compromises efficiently by symmetrically re-encrypting the session key, and storing the message symmetrically encrypted for long-term storage and archival.¶

4. Persistent Symmetric Key Algorithms

This document defines two new "public-key algorithms" for use with OpenPGP, extending the table in section 9.1 of [RFC4880].¶

These algorithm IDs can be used in Secret-Key Packets and Secret-Subkey Packets to hold symmetric key material, in Public-Key Encrypted Session Key Packets (for AEAD) to hold session keys encrypted using AEAD, and in Signature Packets (for HMAC) to store HMAC-based signatures.¶

As the secret key material is required for all cryptographic operations with symmetric keys, implementations SHOULD NOT use these algorithm IDs in Public-Key Packets or Public-Subkey Packets, and SHOULD NOT export Public-Key Packets from Secret-Key Packets holding symmetric key material.¶

5. Other Changes

To reflect the usage of symmetric algorithms, we propose to rename Public-Key Encrypted Session Key Packet (Tag 1) to Key Encrypted Session Key Packet (shorthand "PESK"), and rename Symmetric-Key Encrypted Session Key Packet (Tag 3) to Password Encrypted Session Key Packet (shorthand "SESK"). These names reflect the semantics and intended use of the packets, as opposed to the cryptographic algorithms used.¶

6. Security Considerations

Security considerations are discussed throughout the document where appropriate.¶

7. IANA Considerations

8. Acknowledgements

An initial version of this draft was written by Dan Ristea (Proton AG), with guidance from Dr Philipp Jovanovic (University College London).¶

Author's Address