RFC 9325

Recommendations for Secure Use of Transport Layer Security (TLS) and Datagram Transport Layer Security (DTLS)

📅 published November 2022, from ietf

RFC 9325 🎀 Recommendations for Secure Use of Transport Layer Security (TLS) and Datagram Transport Layer Security (DTLS)

okay bestie, let's get into it 💅✨ this one is the latest TLS/DTLS security guidance, and it's so good. it's a Best Current Practice (that's BCP 195 🫶), published in November 2022 by the IETF.

Who wrote it? Y. Sheffer (Intuit), P. Saint-Andre (Independent), and T. Fossati (ARM Limited) 👑 slay, the whole squad.

It obsoletes RFC 7525 🙈 (the older TLS recommendations, which were written when everyone was moving to TLS 1.2), and it updates RFC 5288 and RFC 6066 🔥 because of some newer attacks. The ISSN is 2070-1721 if you're collecting those for your little library 🌸

Here's the vibe check: the abstract says TLS and DTLS protect data for a wide range of application protocols, and attacks on the most-used cipher suites have been happening for years 💀 so this document gives the latest recommendations to keep deployed services safe. These apply to most use cases 🦋

Quick note from the abstract: TLS 1.2 was the big thing when RFC 7525 came out. Years later, the switch to it is mostly done, and TLS 1.3 is widely available now 🌈 So this doc updates the guidance for the new environment.

Status of This Memo 🍓

This memo documents an Internet Best Current Practice. It's a product of the IETF, it represents the community's consensus, got public review, and was approved by the IESG. More info on BCPs is in Section 2 of RFC 7841 💜

You can check the current status, errata, and feedback info at https://www.rfc-editor.org/info/rfc9325 ✨

Copyright (c) 2022 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. Please read them carefully, because they explain your rights and restrictions. Code Components pulled out of this document must include Revised BSD License text, as described in Section 4.e of the Trust Legal Provisions, and come without warranty as described in that license 🫶

Table of Contents 🧚

  1. Introduction
  2. Terminology
  3. General Recommendations
    • 3.1. Protocol Versions
      • 3.1.1. SSL/TLS Protocol Versions
      • 3.1.2. DTLS Protocol Versions
      • 3.1.3. Fallback to Lower Versions
    • 3.2. Strict TLS
    • 3.3. Compression
      • 3.3.1. Certificate Compression
    • 3.4. TLS Session Resumption
    • 3.5. Renegotiation in TLS 1.2
    • 3.6. Post-Handshake Authentication
    • 3.7. Server Name Indication (SNI)
    • 3.8. Application-Layer Protocol Negotiation (ALPN)
    • 3.9. Multi-Server Deployment
    • 3.10. Zero Round-Trip Time (0-RTT) Data in TLS 1.3
  4. Recommendations: Cipher Suites
    • 4.1. General Guidelines
    • 4.2. Cipher Suites for TLS 1.2
      • 4.2.1. Implementation Details
    • 4.3. Cipher Suites for TLS 1.3
    • 4.4. Limits on Key Usage
    • 4.5. Public Key Length
    • 4.6. Truncated HMAC
  5. Applicability Statement
    • 5.1. Security Services
    • 5.2. Opportunistic Security
  6. IANA Considerations
  7. Security Considerations
    • 7.1. Host Name Validation
    • 7.2. AES-GCM
      • 7.2.1. Nonce Reuse in TLS 1.2
    • 7.3. Forward Secrecy
    • 7.4. Diffie-Hellman Exponent Reuse
    • 7.5. Certificate Revocation
  8. References
    • 8.1. Normative References
    • 8.2. Informative References Appendix A. Differences from RFC 7525 Acknowledgments Authors' Addresses

1. Introduction 🦋

okay so TLS and DTLS protect data for a wide variety of application protocols 💖 like HTTP [RFC9112] [RFC9113], IMAP [RFC9051], Post Office Protocol (POP) [STD53], SIP [RFC3261], SMTP [RFC5321], and XMPP [RFC6120] (the Extensible Messaging and Presence Protocol, which is giving chat energy 🫧). These protocols use both the TLS or DTLS handshake protocol and the TLS or DTLS record layer 🔐

the TLS handshake can also be paired with different record layers to make secure transport protocols, and the big example is QUIC [RFC9000] 🚀 but those aren't directly in scope here. still, a lot of these recs might apply if those protocols use the TLS handshake 🌈

Here's the tea 🍵: up until 2015, the industry was getting hit with serious attacks on TLS and DTLS, including attacks on the most common cipher suites and how they operate. AES-CBC [RFC3602] and RC4 [RFC7465] were both once the most widely deployed ciphers, and both got attacked in TLS 💀 The pre-2015 attacks are explained in a companion doc, RFC 7457 [RFC7457], which helps explain why the old recs (RFC 7525 [RFC7525]) say what they say 🧠 That companion hasn't been updated alongside this one. Instead, the newer attacks and their fixes live here 🎀

The TLS community responded to the RFC 7457 attacks in a few ways ✨

  • Detailed guidance came out for TLS 1.2 [RFC5246] and DTLS 1.2 [RFC6347], along with earlier versions. That guidance is in the original RFC 7525 and mostly kept here. It's been mostly adopted by the industry since 2015 🙌
  • Versions of TLS earlier than 1.2 got deprecated [RFC8996] 🪦
  • TLS 1.3 [RFC8446] came out, then DTLS 1.3 [RFC9147] 🎉 These largely fix or mitigate the attacks

Anyone who implements or deploys TLS (or TLS-based protocols) needs to know how to use it securely 🔒 This doc gives guidance for deployed services AND software implementations, assuming the implementer expects their code to run in the environments in Section 5 🦋 The target audience is basically every deployer who wants authentication (one-way or mutual), confidentiality, and data integrity for their communications 💜

The recs take into account the security of different mechanisms, how mature and interoperable they are, and how common they are in implementations at the time of writing. Unless something is explicitly called out as applying to TLS alone or DTLS alone, each recommendation applies to both TLS and DTLS 🫶

This doc tries to keep new guidance for TLS 1.2 implementations to a minimum 🤏 the big goal is to get systems onto TLS 1.3. but that's not always realistic 😭 New attacks and ecosystem changes meant some new requirements for TLS 1.2 environments were needed. Those are summarized in Appendix A 📝

future attacks are definitely coming, and this doc can't cover them, so implementers and deployers should really keep watching for new developments 👀 quantum computers are expected to seriously affect cryptographic primitives, but post-quantum crypto is still a work in progress and it's too early to make recs. once those specs are standardized in the IETF or elsewhere, this doc should be updated to reflect best practices then 🔮

the TLS 1.3 spec fixes lots of the vulns in this doc, so a system deploying TLS 1.3 should have fewer vulns than one on TLS 1.2 or below ✨ That's why this doc replaces RFC 7525, with a clear goal of moving most TLS 1.2 uses over to TLS 1.3 🚀

These are minimum recommendations for TLS in the vast majority of implementation and deployment scenarios, except for unauthenticated TLS (see Section 5) 🙈 Other specs that reference this doc can have stricter requirements based on their own situation (for example, with a specific application protocol), and in that case implementers are advised to follow those stricter rules 👑 This doc is a floor, not a ceiling: where it's feasible, service admins are encouraged to go beyond the minimum support in implementations for the strongest security possible 💪 For example, if a service provider knows the deployed base for an application protocol and has done a cost-benefit analysis of security strength vs. interoperability, they might decide to turn off TLS 1.2 entirely and offer only TLS 1.3 🦋

Algorithm strength and feasible attacks can change really fast, and experience shows a BCP about security is a point-in-time statement ⏰ So readers should look for any errata or updates that apply to this doc 🔍

This doc updates [RFC5288] in view of the [Boeck2016] attack. See Section 7.2.1 for details 🔥

This doc updates [RFC6066] in view of the [ALPACA] attack. See Section 3.7 for details 💅

2. Terminology 🧚

A bunch of security terms in this doc are used in the sense defined in [RFC4949], including "attack", "authentication", "certificate", "cipher", "compromise", "confidentiality", "credential", "data integrity", "encryption", "forward secrecy", "key", "key length", "self-signed certificate", "strength", and "strong" 📚

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this doc are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here 🎀

3. General Recommendations 🌸

This section gives general recs on secure TLS use. Cipher suite recs are in the next section 💖

3.1. Protocol Versions ✨

Part 1 of 9. More coming up, bestie 🫶 stay tuned for Section 3.1 and beyond!

3.1.1. SSL/TLS Protocol Versions

okay so this is a two-part glow-up 💅✨. We have to ditch the old, less secure SSL/TLS versions AND start using the modern, more secure ones. Here's the full tea on which versions to use:

  • Implementations MUST NOT negotiate SSL version 2.

Rationale: Today, SSLv2 is considered insecure [RFC6176]. 💀

  • Implementations MUST NOT negotiate SSL version 3.

Rationale: SSLv3 [RFC6101] was a glow-up from SSLv2 and patched some big security holes, but it couldn't do strong cipher suites 🥲. It also doesn't support TLS extensions, and some of those, like renegotiation_info [RFC5746], are security critical 🚨. On top of that, the Padding Oracle On Downgraded Legacy Encryption (POODLE) attack [POODLE] showed up, and now SSLv3 is widely recognized as fundamentally insecure 🙅‍♀️. See [RFC7568] for further details.

  • Implementations MUST NOT negotiate TLS version 1.0 [RFC2246].

Rationale: TLS 1.0 (published in 1999) doesn't support many modern, strong cipher suites 😩. It also lacks a per-record Initialization Vector (IV) for cipher suites based on cipher block chaining (CBC), and it doesn't warn against common padding errors 🙈. This and other recommendations in this section line up with [RFC8996] 🫶.

  • Implementations MUST NOT negotiate TLS version 1.1 [RFC4346].

Rationale: TLS 1.1 (published in 2006) is a security upgrade over TLS 1.0, but it's still missing certain stronger cipher suites that came in with the standardization of TLS 1.2 in 2008. That includes the cipher suites this document recommends for TLS 1.2 (see Section 4.2 below) ✨.

  • Implementations MUST support TLS 1.2 [RFC5246].

Rationale: TLS 1.2 is implemented and deployed way more widely than TLS 1.3 right now 📊. When you follow this document's recommendations to mitigate known attacks, using TLS 1.2 is just as safe as using TLS 1.3 🔐. Most application protocols that reuse TLS and DTLS don't have an immediate need to migrate solely to TLS 1.3. Lots of application clients depend on TLS libraries or operating systems that don't support TLS 1.3 yet, so proactively deprecating TLS 1.2 would cause significant interoperability issues, and that would harm security more than it helps 😭. Still, it's expected that a future version of this BCP will deprecate TLS 1.2 when that's appropriate 🔮.

  • Implementations SHOULD support TLS 1.3 [RFC8446] and, if implemented, MUST prefer to negotiate TLS 1.3 over earlier versions of TLS.

Rationale: TLS 1.3 is a major overhaul of the protocol and fixes many of the security issues in TLS 1.2 👑. To the extent that an implementation supports TLS 1.2 (even if it defaults to TLS 1.3), it MUST follow the recommendations regarding TLS 1.2 specified in this document.

  • New transport protocols that integrate the TLS/DTLS handshake protocol and/or record layer MUST use only TLS/DTLS 1.3 (for instance, QUIC [RFC9001] took this approach). New application protocols that employ TLS/DTLS for channel or session encryption MUST integrate with both TLS/DTLS versions 1.2 and 1.3; nevertheless, in rare cases where broad interoperability is not a concern, application protocol designers MAY choose to forego TLS 1.2.

Rationale: Secure deployment of TLS 1.3 is significantly easier and less error-prone than secure deployment of TLS 1.2 🧚. When you're designing a new secure transport protocol like QUIC, there's no reason to support TLS 1.2. New application protocols that reuse TLS, though, need to support both TLS 1.3 and TLS 1.2 so they can take advantage of underlying library or operating system support for both versions 🌈.

This BCP applies to TLS 1.3, TLS 1.2, and earlier versions.

It is not safe for readers to assume that the recommendations in this BCP apply to any future version of TLS. 🙅‍♀️

3.1.2. DTLS Protocol Versions

DTLS, an adaptation of TLS for UDP datagrams, was introduced when TLS 1.1 was published. Here are the recommendations for DTLS 🦋:

  • Implementations MUST NOT negotiate DTLS version 1.0 [RFC4347].

Version 1.0 of DTLS correlates to version 1.1 of TLS (see above).

  • Implementations MUST support DTLS 1.2 [RFC6347].

Version 1.2 of DTLS correlates to version 1.2 of TLS (see above). (There is no version 1.1 of DTLS.) 🙃

  • Implementations SHOULD support DTLS 1.3 [RFC9147] and, if implemented, MUST prefer to negotiate DTLS version 1.3 over earlier versions of DTLS.

Version 1.3 of DTLS correlates to version 1.3 of TLS (see above) 💜.

3.1.3. Fallback to Lower Versions

TLS/DTLS 1.2 clients MUST NOT fall back to earlier TLS versions, since those versions have been deprecated [RFC8996]. As a result, the downgrade-protection Signaling Cipher Suite Value (SCSV) mechanism [RFC7507] is no longer needed for clients. In addition, TLS 1.3 implements a new version-negotiation mechanism ✨.

3.2. Strict TLS

The following recommendations are here to help prevent "SSL Stripping" and STARTTLS command injection (attacks that are summarized in [RFC7457]) 🛡️:

  • Many existing application protocols were designed before TLS became common. These protocols usually support TLS in one of two ways: either via a separate port for TLS-only communication (e.g., port 443 for HTTPS) or via a method for dynamically upgrading a channel from unencrypted to TLS protected (e.g., STARTTLS, which is used in protocols such as IMAP and XMPP). Regardless of the mechanism for protecting the communication channel (TLS-only port or dynamic upgrade), what matters is the end state of the channel 🎯. When a protocol defines both a dynamic upgrade method and a separate TLS-only method, then the separate TLS-only method MUST be supported by implementations and MUST be configured by administrators to be used in preference to the dynamic upgrade method. When a protocol supports only a dynamic upgrade method, implementations MUST provide a way for administrators to set a strict local policy that forbids use of plaintext in the absence of a negotiated TLS channel, and administrators MUST use this policy.
  • HTTP client and server implementations intended for use in the World Wide Web (see Section 5) MUST support the HTTP Strict Transport Security (HSTS) header field [RFC6797] so that web servers can advertise that they are willing to accept TLS-only clients. Web servers SHOULD use HSTS to indicate that they are willing to accept TLS-only clients, unless they are deployed in such a way that using HSTS would in fact weaken overall security (e.g., it can be problematic to use HSTS with self-signed certificates, as described in Section 11.3 of [RFC6797]). Similar technologies exist for non-HTTP application protocols, such as Mail Transfer Agent Strict Transport Security (MTA-STS) for mail transfer agents [RFC8461] and methods based on DNS-Based Authentication of Named Entities (DANE) [RFC6698] for SMTP [RFC7672] and XMPP [RFC7712] 🌸.

Rationale: Combining unprotected and TLS-protected communication opens the way to SSL Stripping and similar attacks 😬. An initial part of the communication isn't integrity protected, so an attacker whose goal is to keep the communication in the clear can manipulate it.

3.3. Compression

In order to help prevent compression-related attacks (summarized in Section 2.6 of [RFC7457]) when using TLS 1.2, implementations and deployments SHOULD NOT support TLS-level compression (Section 6.2.2 of [RFC5246]). The only exception is when the application protocol in question has been proven not to be open to such attacks. Even then, extreme caution is warranted because of the potential for future attacks related to TLS compression 🙀. More specifically, the HTTP protocol is known to be vulnerable to compression-related attacks. (This recommendation applies to TLS 1.2 only, because compression has been removed from TLS 1.3.) 🧐

Rationale: TLS compression has been subject to security attacks such as the Compression Ratio Info-leak Made Easy (CRIME) attack 💥.

Implementers should note that compression at higher protocol levels can allow an active attacker to extract cleartext information from the connection. The Browser Reconnaissance and Exfiltration via Adaptive Compression of Hypertext (BREACH) attack is one such case. These issues can only be mitigated outside of TLS, so they're outside the scope of this document. See Section 2.6 of [RFC7457] for further details 🔍.

3.3.1. Certificate Compression

Certificate chains often take up most of the bytes transmitted during the handshake 📦. In order to manage their size, some or all of the following methods can be employed (see also Section 4 of [RFC9191] for further suggestions):

  • Limit the number of names or extensions.
  • Use keys with small public key representations, like the Elliptic Curve Digital Signature Algorithm (ECDSA).
  • Use certificate compression.

To achieve the latter, TLS 1.3 defines the compress_certificate extension in [RFC8879]. See also Section 5 of [RFC8879] for security and privacy considerations associated with its use. Just so we're clear, CRIME-style attacks on TLS compression do not apply to certificate compression 🫶.

Due to the strong likelihood of middlebox interference, compression in the style of [RFC8879] has not been made available in TLS 1.2 🚧. In theory, the cached_info extension defined in [RFC7924] could be used, but it isn't supported widely enough to be considered a practical alternative 🥺.

3.4. TLS Session Resumption 🔁✨

Okay bestie, session resumption is literally why your connections feel so fast 💨 It drastically cuts down the number of full TLS handshakes, so it's an essential performance feature for most deployments 🚀

Stateless session resumption with session tickets is a super popular move 💅 For TLS 1.2, it's specified in [RFC5077]. For TLS 1.3, a more secure mechanism based on a pre-shared key (PSK) is described in Section 4.6.1 of [RFC8446]. If you want the tea on the risks of TLS cryptographic "shortcuts", including session resumption, check out [Springall16] ☕👀

When it's used, the resumption information MUST be authenticated and encrypted so an attacker can't modify it or eavesdrop on it 🔒 There are more recs for session tickets, too:

  • A strong cipher MUST be used when encrypting the ticket (at least as strong as the main TLS cipher suite). 💪
  • Ticket-encryption keys MUST be changed regularly, e.g., once every week, so as not to negate the benefits of forward secrecy (see Section 7.3 for details on forward secrecy). Old ticket-encryption keys MUST be destroyed at the end of the validity period. 🗑️
  • For similar reasons, session ticket validity MUST be limited to a reasonable duration (e.g., half as long as ticket-encryption key validity). ⏳
  • TLS 1.2 does not roll the session key forward within a single session. Thus, to prevent an attack where the server's ticket-encryption key is stolen and used to decrypt the entire content of a session (negating the concept of forward secrecy), a TLS 1.2 server SHOULD NOT resume sessions that are too old, e.g., sessions that have been open longer than two ticket-encryption key rotation periods. 🙅‍♀️

Rationale: Session resumption is another kind of TLS handshake, so it has to be just as secure as the initial handshake 🧐 This document (Section 4) recommends the use of cipher suites that provide forward secrecy, meaning they prevent an attacker who gains momentary access to the TLS endpoint (either client or server) and its secrets from reading either past or future communication. The tickets have to be managed so they don't negate this security property. 🛡️

TLS 1.3 provides the powerful option of forward secrecy even within a long-lived connection that is periodically resumed. Section 2.2 of [RFC8446] recommends that clients SHOULD send a "key_share" when initiating session resumption. To gain forward secrecy, this document recommends that server implementations SHOULD select the "psk_dhe_ke" PSK key exchange mode and respond with a "key_share" to complete an Ephemeral Elliptic Curve Diffie-Hellman (ECDHE) exchange on each session resumption. As a more performant alternative, server implementations MAY refrain from responding with a "key_share" until a certain amount of time (e.g., measured in hours) has passed since the last ECDHE exchange. This implies that the "key_share" operation would not occur for the presumed majority of session resumption requests (which would occur within a few hours), while still ensuring forward secrecy for longer-lived sessions. 🦋

TLS session resumption introduces potential privacy issues, since the server is able to track the client, in some cases indefinitely 👀 See [Sy2018] for more details.

3.5. Renegotiation in TLS 1.2 🔄

The recommendations in this section apply to TLS 1.2 only, because renegotiation has been removed from TLS 1.3 🙅‍♀️

Renegotiation in TLS 1.2 is a handshake that establishes new cryptographic parameters for an existing session. The mechanism existed in TLS 1.2 and in earlier protocol versions, and it was improved after several major attacks, including a plaintext injection attack, CVE-2009-3555 [CVE]. 😬

TLS 1.2 clients and servers MUST implement the renegotiation_info extension, as defined in [RFC5746]. 🎀

TLS 1.2 clients MUST send renegotiation_info in the Client Hello. If the server does not acknowledge the extension, the client MUST generate a fatal handshake_failure alert prior to terminating the connection. 🚨

Rationale: It is not safe for a client to connect to a TLS 1.2 server that does not support renegotiation_info, regardless of whether either endpoint actually implements renegotiation. See also Section 4.1 of [RFC5746]. 🫶

A related attack, resulting from TLS session parameters not being properly authenticated, is a Triple Handshake [Triple-Handshake]. To address this attack, TLS 1.2 implementations MUST support the extended_master_secret extension defined in [RFC7627]. 💜

3.6. Post-Handshake Authentication 🤝

Renegotiation in TLS 1.2 was (partially) replaced in TLS 1.3 by separate post-handshake authentication and key update mechanisms. In the context of protocols that multiplex requests over a single connection (such as HTTP/2 [RFC9113]), post-handshake authentication has the same problems as TLS 1.2 renegotiation. Multiplexed protocols SHOULD follow the advice provided for HTTP/2 in Section 9.2.3 of [RFC9113]. 🧩

3.7. Server Name Indication (SNI) 🏷️

TLS implementations MUST support the Server Name Indication (SNI) extension defined in Section 3 of [RFC6066] for those higher-level protocols that would benefit from it, including HTTPS. However, the actual use of SNI in particular circumstances is a matter of local policy. At the time of writing, a technology for encrypting the SNI (called Encrypted Client Hello) is being worked on in the TLS Working Group [TLS-ECH]. Once that method has been standardized and widely implemented, it will likely be appropriate to recommend its usage in a future version of this BCP. 👀✨

Rationale: SNI supports deployment of multiple TLS-protected virtual servers on a single address, and therefore enables fine-grained security for these virtual servers, by allowing each one to have its own certificate. However, SNI also leaks the target domain for a given connection. This information leak will be closed by use of TLS Encrypted Client Hello once that method has been standardized. 🔐

In order to prevent the attacks described in [ALPACA], a server that does not recognize the presented server name SHOULD NOT continue the handshake and instead SHOULD fail with a fatal-level unrecognized_name(112) alert. Note that this recommendation updates Section 3 of [RFC6066], which stated:

If the server understood the ClientHello extension but does not recognize the server name, the server SHOULD take one of two actions: either abort the handshake by sending a fatal-level unrecognized_name(112) alert or continue the handshake.

Clients SHOULD abort the handshake if the server acknowledges the SNI extension but presents a certificate with a different hostname than the one sent by the client. 🚫

3.8. Application-Layer Protocol Negotiation (ALPN) 🗣️

TLS implementations (both client- and server-side) MUST support the Application-Layer Protocol Negotiation (ALPN) extension [RFC7301]. 🌸

In order to prevent "cross-protocol" attacks, which result from failing to ensure that a message intended for use in one protocol cannot be mistaken for a message for use in another protocol, servers are advised to strictly enforce the behavior prescribed in Section 3.2 of [RFC7301]:

In the event that the server supports no protocols that the client advertises, then the server SHALL respond with a fatal 'no_application_protocol' alert.

Clients SHOULD abort the handshake if the server acknowledges the ALPN extension but does not select a protocol from the client list. Failure to do so can result in attacks such as those described in [ALPACA]. 💥

Protocol developers are strongly encouraged to register an ALPN identifier for their protocols. This applies both to new protocols and to well-established protocols. However, because the latter might have a large deployed base, strict enforcement of ALPN usage may not be feasible when an ALPN identifier is registered for a well-established protocol. 📝

3.9. Multi-Server Deployment 🖥️🖥️

Deployments that involve multiple servers or services can increase the size of the attack surface for TLS. Two scenarios are of interest:

  1. Deployments in which multiple services handle the same domain name via different protocols (e.g., HTTP and IMAP). In this case, an attacker might be able to direct a connecting endpoint to the service offering a different protocol and mount a cross-protocol attack. In a cross-protocol attack, the client and server believe they are using different protocols, which the attacker might exploit if messages sent in one protocol are interpreted as messages in the other protocol with undesirable effects (see [ALPACA] for more detailed information about this class of attacks). To mitigate this threat, service providers SHOULD deploy ALPN (see Section 3.8). In addition, to the extent possible, they SHOULD ensure that multiple services handling the same domain name provide equivalent levels of security that are consistent with the recommendations in this document. Such measures SHOULD include the handling of configurations across multiple TLS servers and protections against compromise of credentials held by those servers. 🕵️‍♀️
  2. Deployments in which multiple servers providing the same service have different TLS configurations. In this case, an attacker might be able to direct a connecting endpoint to a server with a TLS configuration that is more easily exploitable (see [DROWN] for more detailed information about this class of attacks). To mitigate this threat, service providers SHOULD ensure that all servers providing the same service provide equivalent levels of security that are consistent with the recommendations in this document. 🧯

3.10. Zero Round-Trip Time (0-RTT) Data in TLS 1.3 ⚡

The 0-RTT early data feature is new in TLS 1.3. It provides reduced latency when TLS connections are resumed, at the potential cost of certain security properties. As a result, it requires special attention from implementers on both the server and the client side. Typically, this extends to the TLS library as well as protocol layers above it. ⚠️

For HTTP over TLS, refer to [RFC8470] for guidance.

For QUIC on TLS, refer to Section 9.2 of [RFC9001].

For other protocols, generic guidance is given in Section 8 and Appendix E.5 of [RFC8446]. To paraphrase Appendix E.5, applications MUST avoid this feature unless an explicit specification exists for the application protocol in question to clarify when 0-RTT is appropriate and secure. This can take the form of an IETF RFC, a non-IETF standard, or documentation associated with a non-standard protocol. 📜

4. Recommendations: Cipher Suites 🔐🌈

TLS 1.2 provided considerable flexibility in the selection of cipher suites. Unfortunately, the security of some of these cipher suites has degraded over time, to the point where some are known to be insecure (this is one reason why TLS 1.3 restricted such flexibility). Incorrectly configuring a server leads to no or reduced security. This section includes recommendations on the selection and negotiation of cipher suites. 💖

4.1. General Guidelines 💅

Okay so cryptographic algorithms get weaker over time because cryptanalysis keeps getting better 😭. Stuff that was once strong can become weak, so cipher suites using weak algorithms need to be phased out and replaced with more secure ones. That's how the desired security properties keep holding ✨. SSL/TLS has been around for well over 20 years, and a lot of the cipher suites recommended in older versions are now considered weak, or at least not as strong as we'd like 🙈. So this section gives cipher suite selection a glow-up 💖.

  • Implementations MUST NOT negotiate the cipher suites with NULL encryption.

Rationale: The NULL cipher suites don't encrypt traffic, so they provide no confidentiality services. Any entity in the network with access to the connection can view the plaintext of the contents being exchanged by the client and server. Still, this document doesn't discourage software from implementing NULL cipher suites, since they can be useful for testing and debugging 🧚.

  • Implementations MUST NOT negotiate RC4 cipher suites.

Rationale: The RC4 stream cipher has a bunch of cryptographic weaknesses, as documented in [RFC7465]. Note that DTLS already specifically forbids the use of RC4 🙅‍♀️.

  • Implementations MUST NOT negotiate cipher suites offering less than 112 bits of security, including so-called "export-level" encryption (which provides 40 or 56 bits of security).

Rationale: Based on [RFC3766], at least 112 bits of security is needed. 40-bit and 56-bit security (found in so-called "export ciphers") is considered insecure today 💀.

  • Implementations SHOULD NOT negotiate cipher suites that use algorithms offering less than 128 bits of security.

Rationale: Cipher suites that offer 112 or more bits, but less than 128 bits, of security are not considered weak at this time. However, their useful lifespan is expected to be short enough to justify supporting stronger cipher suites right now. 128-bit ciphers are expected to remain secure for at least several years, and 256-bit ciphers until the next fundamental technology breakthrough 🔥. Note that, because of so-called "meet-in-the-middle" attacks [Multiple-Encryption], some legacy cipher suites (e.g., 168-bit Triple DES (3DES)) have an effective key length that is smaller than their nominal key length (112 bits in the case of 3DES). Such cipher suites should be evaluated according to their effective key length.

  • Implementations SHOULD NOT negotiate cipher suites based on RSA key transport, a.k.a. "static RSA".

Rationale: These cipher suites, which have assigned values starting with the string TLS_RSA_WITH_*, have several drawbacks. The biggest one is that they don't support forward secrecy 🙈.

  • Implementations SHOULD NOT negotiate cipher suites based on non-ephemeral (static) finite-field Diffie-Hellman (DH) key agreement. Similarly, implementations SHOULD NOT negotiate non-ephemeral Elliptic Curve DH key agreement.

Rationale: The former cipher suites, which have assigned values prefixed by TLS_DH_*, have several drawbacks, especially the fact that they don't support forward secrecy. The latter (TLS_ECDH_*) also lack forward secrecy and are subject to invalid curve attacks [Jager2015] 😭.

  • Implementations MUST support and prefer to negotiate cipher suites offering forward secrecy. However, TLS 1.2 implementations SHOULD NOT negotiate cipher suites based on ephemeral finite-field Diffie-Hellman key agreement (i.e., TLS_DHE_* suites). This is justified by the known fragility of the construction (see [RACCOON]) and the limitations around negotiation, including using [RFC7919], which has seen very limited uptake.

Rationale: Forward secrecy (sometimes called "perfect forward secrecy") prevents the recovery of information that was encrypted with older session keys. That limits how far back in time data can be decrypted when an attack is successful 🦋. See Sections 7.3 and 7.4 for a detailed discussion.

4.2. Cipher Suites for TLS 1.2 🌸

Given the foregoing considerations, implementation and deployment of the following cipher suites is RECOMMENDED:

  • TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256
  • TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384
  • TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256
  • TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384

As these are Authenticated Encryption with Associated Data (AEAD) algorithms [RFC5116], these cipher suites are supported only in TLS 1.2 and not in earlier protocol versions 👑.

Typically, to prefer these suites, the order of suites needs to be explicitly configured in server software. It would be ideal if server software implementations preferred these suites by default 🫶.

Some devices have hardware support for AES Counter Mode with CBC-MAC (AES-CCM) but not AES Galois/Counter Mode (AES-GCM), so they can't follow the foregoing recommendations about cipher suites. There are even devices that don't support public key cryptography at all, but those are out of scope entirely 🤷‍♀️.

A cipher suite that operates in CBC (cipher block chaining) mode (e.g., TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256) SHOULD NOT be used unless the encrypt_then_mac extension [RFC7366] is also successfully negotiated. This requirement applies to both client and server implementations.

When using ECDSA signatures for authentication of TLS peers, it is RECOMMENDED that implementations use the NIST curve P-256. In addition, to avoid predictable or repeated nonces (which could reveal the long-term signing key), it is RECOMMENDED that implementations implement "deterministic ECDSA" as specified in [RFC6979] and in line with the recommendations in [RFC8446] ✨.

Note that implementations of "deterministic ECDSA" may be vulnerable to certain side-channel and fault injection attacks precisely because of their determinism 😱. While most fault injection attacks described in the literature assume physical access to the device (and are therefore more relevant in Internet of Things (IoT) deployments with poor or non-existent physical security), some can be carried out remotely [Poddebniak2017], e.g., as Rowhammer [Kim2014] variants. In deployments where side-channel attacks and fault injection attacks are a concern, implementation strategies that combine both randomness and determinism (for example, as described in [CFRG-DET-SIGS]) can be used to avoid the risk of successful extraction of the signing key 🔐.

4.2.1. Implementation Details 🎀

Clients SHOULD include TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 as the first proposal to any server. Servers MUST prefer this cipher suite over weaker cipher suites whenever it is proposed, even if it isn't the first proposal. Clients are of course free to offer stronger cipher suites, e.g., using AES-256. When they do, the server SHOULD prefer the stronger cipher suite unless there are compelling reasons (e.g., seriously degraded performance) to choose otherwise 💜.

The previous version of the TLS recommendations [RFC7525] implicitly allowed the old RFC 5246 mandatory-to-implement cipher suite, TLS_RSA_WITH_AES_128_CBC_SHA. At the time of writing, this cipher suite doesn't provide additional interoperability, except with very old clients. As with other cipher suites that don't provide forward secrecy, implementations SHOULD NOT support this cipher suite. Other application protocols specify other cipher suites as mandatory to implement (MTI) 🍓.

[RFC8422] allows clients and servers to negotiate ECDH parameters (curves). Both clients and servers SHOULD include the "Supported Elliptic Curves Extension" [RFC8422]. Clients and servers SHOULD support the NIST P-256 (secp256r1) [RFC8422] and X25519 (x25519) [RFC7748] curves. Note that [RFC8422] deprecates all but the uncompressed point format. Therefore, if the client sends an ec_point_formats extension, the ECPointFormatList MUST contain a single element, uncompressed 🌈.

4.3. Cipher Suites for TLS 1.3 ⭐

This document does not specify any cipher suites for TLS 1.3. Readers are referred to Section 9.1 of [RFC8446] for cipher suite recommendations.

4.4. Limits on Key Usage 🔥

All ciphers have an upper limit on the amount of traffic that can be securely protected with any given key. In the case of AEAD cipher suites, two separate limits are maintained for each key:

  1. Confidentiality limit (CL), i.e., the number of records that can be encrypted.
  2. Integrity limit (IL), i.e., the number of records that are allowed to fail authentication.

The latter applies to DTLS (and also to QUIC) but not to TLS itself, since TLS connections are torn down on the first decryption failure.

When a sender is approaching CL, the implementation SHOULD initiate a new handshake (in TLS 1.3, this can be achieved by sending a KeyUpdate message on the established session) to rotate the session key. When a receiver has reached IL, the implementation SHOULD close the connection. Although these recommendations are a best practice, implementers need to be aware that it isn't always easy to accomplish them in protocols built on top of TLS/DTLS without introducing coordination across layer boundaries. See Section 6 of [RFC9001] for an example of the cooperation that was necessary in QUIC between the crypto and transport layers to support key updates. Note that, in general, application protocols might not be able to emulate that method, given their more constrained interaction with TLS/DTLS. As a result of these complexities, these recommendations are not mandatory 🥺.

For all TLS 1.3 cipher suites, readers are referred to Section 5.5 of [RFC8446] for the values of CL and IL. For all DTLS 1.3 cipher suites, readers are referred to Section 4.5.3 of [RFC9147].

For all AES-GCM cipher suites recommended for TLS 1.2 and DTLS 1.2 in this document, CL can be derived by plugging the corresponding parameters into the inequalities in Section 6.1 of [AEAD-LIMITS] that apply to random, partially implicit nonces, i.e., the nonce construction used in TLS 1.2. Although the obtained figures are slightly higher than those for TLS 1.3, it is RECOMMENDED that the same limit of 2^24.5 records is used for both versions 💅.

For all AES-GCM cipher suites recommended for DTLS 1.2, IL (obtained from the same inequalities referenced above) is 2^28.

4.5. Public Key Length 🔑✨

okay so when you use the cipher suites this doc recommends, the TLS handshake normally has two public keys hanging out: one for the Diffie-Hellman key agreement and one for server authentication. 🫶 If there's a client certificate in the mix, that's a third key joining the party 💅

For the "DHE" cipher suites (the ones built on modular exponential (MODP) Diffie-Hellman groups), DH key lengths of at least 2048 bits are REQUIRED. 🔥

Rationale: ngl this part is lowkey the most important bit 🙈 For various reasons, DH keys in practice usually come in lengths that are powers of two (e.g., 2^10 = 1024 bits, 2^11 = 2048 bits, 2^12 = 4096 bits). A DH key of 1228 bits would be roughly equivalent to only an 80-bit symmetric key [RFC3766], so it's better to go longer than that for the "DHE" family of cipher suites. A DH key of 1926 bits would be roughly equivalent to a 100-bit symmetric key [RFC3766]. A DH key of 2048 bits (equivalent to a 112-bit symmetric key) is the minimum allowed by the latest revision of [NIST.SP.800-56A] as of this writing (see in particular Appendix D of that document).

As noted in [RFC3766], correcting for the emergence of The Weizmann Institute Relation Locator (TWIRL) machine [TWIRL] would imply that 1024-bit DH keys yield about 61 bits of equivalent strength, and that a 2048-bit DH key would yield about 92 bits of equivalent strength. The Logjam attack [Logjam] further demonstrates that 1024-bit Diffie-Hellman parameters should be avoided. 💀

With regard to ECDH keys, implementers are referred to the IANA "TLS Supported Groups" registry (formerly known as the "EC Named Curve Registry") within the "Transport Layer Security (TLS) Parameters" registry [IANA_TLS], and in particular to the "recommended" groups. Curves of less than 224 bits MUST NOT be used. This recommendation is in line with the latest revision of [NIST.SP.800-56A]. 🌸

When using RSA, servers MUST authenticate using certificates with at least a 2048-bit modulus for the public key. In addition, the use of the SHA-256 hash algorithm is RECOMMENDED, and SHA-1 or MD5 MUST NOT be used [RFC9155] (for more details, see also [CAB-Baseline], for which the current version at the time of writing is 1.8.4). Clients MUST indicate to servers that they request SHA-256 by using the "Signature Algorithms" extension defined in TLS 1.2. For TLS 1.3, the same requirement is already specified by [RFC8446]. 🔐

4.6. Truncated HMAC 🎀

Implementations MUST NOT use the Truncated HMAC Extension, defined in Section 7 of [RFC6066]. 🚫

Rationale: the extension does not apply to the AEAD cipher suites recommended above. However, it does apply to most other TLS cipher suites. Its use has been shown to be insecure in [PatersonRS11]. 😭

5. Applicability Statement 🌈

The recommendations of this document primarily apply to the implementation and deployment of application protocols that are most commonly used with TLS and DTLS on the Internet today. Examples include, but are not limited to:

  • Web software and services that wish to protect HTTP traffic with TLS. 🌐
  • Email software and services that wish to protect IMAP, Post Office Protocol version 3 (POP3), or SMTP traffic with TLS. 📧
  • Instant-messaging software and services that wish to protect Extensible Messaging and Presence Protocol (XMPP) or Internet Relay Chat (IRC) traffic with TLS. 💬
  • Realtime media software and services that wish to protect Secure Realtime Transport Protocol (SRTP) traffic with DTLS. 🎥

This document does not modify the implementation and deployment recommendations (e.g., mandatory-to-implement cipher suites) prescribed by existing application protocols that employ TLS or DTLS. If the community that uses such an application protocol wishes to modernize its usage of TLS or DTLS to be consistent with the best practices recommended here, it needs to explicitly update the existing application protocol definition (one example is [RFC7590], which updates [RFC6120]). 🧚

Designers of new application protocols developed through the Internet Standards Process [RFC2026] are expected at minimum to conform to the best practices recommended here, unless they provide documentation of compelling reasons that would prevent such conformance (e.g., widespread deployment on constrained devices that lack support for the necessary algorithms). 👑

Although many of the recommendations provided here might also apply to QUIC insofar that it uses the TLS 1.3 handshake protocol, QUIC and other such secure transport protocols are out of scope of this document. For QUIC specifically, readers are referred to Section 9.2 of [RFC9001]. ✨

This document does not address the use of TLS in constrained-node networks [RFC7228]. For recommendations regarding the profiling of TLS and DTLS for small devices with severe constraints on power, memory, and processing resources, the reader is referred to [RFC7925] and [IOT-PROFILE]. 🔋

5.1. Security Services 🔐

This document provides recommendations for an audience that wishes to secure their communication with TLS to achieve the following:

  • Confidentiality: all application-layer communication is encrypted, with the goal that no party should be able to decrypt it except the intended receiver. 🤫
  • Data integrity: any changes made to the communication in transit are detectable by the receiver. 🛡️
  • Authentication: an endpoint of the TLS communication is authenticated as the intended entity to communicate with. 🪪

With regard to authentication, TLS enables authentication of one or both endpoints in the communication. In the context of opportunistic security [RFC7435], TLS is sometimes used without authentication. As discussed in Section 5.2, considerations for opportunistic security are not in scope for this document.

If deployers deviate from the recommendations given in this document, they need to be aware that they might lose access to one of the foregoing security services. 😬

This document applies only to environments where confidentiality is required. It requires algorithms and configuration options that enforce secrecy of the data in transit.

This document also assumes that data integrity protection is always one of the goals of a deployment. In cases where integrity is not required, it does not make sense to employ TLS in the first place. There are attacks against confidentiality-only protection that utilize the lack of integrity to also break confidentiality (see, for instance, [DegabrieleP07] in the context of IPsec). 🙅‍♀️

This document addresses itself to application protocols that are most commonly used on the Internet with TLS and DTLS. Typically, all communication between TLS clients and TLS servers requires all three of the above security services. This is particularly true where TLS clients are user agents like web browsers or email clients. 💖

This document does not address the rarer deployment scenarios where one of the above three properties is not desired, such as the use case described in Section 5.2. As another scenario where confidentiality is not needed, consider a monitored network where the authorities in charge of the respective traffic domain require full access to unencrypted (plaintext) traffic, and where users collaborate and send their traffic in the clear. 🕵️

5.2. Opportunistic Security 🦋

There are several important scenarios in which the use of TLS is optional, i.e., the client decides dynamically ("opportunistically") whether to use TLS with a particular server or to connect in the clear. This practice, often called "opportunistic security", is described at length in [RFC7435] and is often motivated by a desire for backward compatibility with legacy deployments. 🕰️

In these scenarios, some of the recommendations in this document might be too strict, since adhering to them could cause fallback to cleartext, a worse outcome than using TLS with an outdated protocol version or cipher suite. 😵

6. IANA Considerations 📋

This document has no IANA actions. 🎉

7. Security Considerations 🚨

This entire document discusses the security practices directly affecting applications using the TLS protocol. This section contains broader security considerations related to technologies used in conjunction with or by TLS. The reader is referred to the Security Considerations sections of TLS 1.3 [RFC8446], DTLS 1.3 [RFC9147], TLS 1.2 [RFC5246], and DTLS 1.2 [RFC6347] for further context. 📚

7.1. Host Name Validation 🕵️‍♀️

Application authors, take note: some TLS implementations do not validate host names. If the TLS implementation you're using does not validate host names, you might need to write your own validation code or consider using a different TLS implementation. 🚩

It is noted that the requirements regarding host name validation (and, in general, binding between the TLS layer and the protocol that runs above it) vary between different protocols. For HTTPS, these requirements are defined by Sections 4.3.3, 4.3.4, and 4.3.5 of [RFC9110]. 🌐

Host name validation is security-critical for all common TLS use cases. Without it, TLS ensures that the certificate is valid and guarantees possession of the private key, but it does not ensure that the connection terminates at the desired endpoint. Readers are referred to [RFC6125] for further details regarding generic host name validation in the TLS context. In addition, that RFC contains a long list of application protocols, some of which implement a policy very different from HTTPS. 😱

If the host name is discovered indirectly and insecurely (e.g., by a cleartext DNS query for an SRV or Mail Exchange (MX) record), it SHOULD NOT be used as a reference identifier [RFC6125] even when it matches the presented certificate. This proviso does not apply if the host name is discovered securely (for further discussion, see [RFC7673] and [RFC7672]). 🔒

Host name validation typically applies only to the leaf "end entity" certificate. Naturally, in order to ensure proper authentication in the context of the PKI, application clients need to verify the entire certification path in accordance with [RFC5280]. 🧩

7.2. AES-GCM

Okay bestie, AES-GCM is the authenticated encryption algorithm that [Section 4.2](#) recommends, and if you want the security stuff that's specific to using it with TLS, go check out Section 6 of [RFC5288] 🎀✨

7.2.1. Nonce Reuse in TLS 1.2

So this one is lowkey scary 💀. [Boeck2016] documents TLS stacks out there that accidentally reuse the AES-GCM nonce. That means AES-GCM can get implemented insecurely, and then TLS sessions using an AES-GCM cipher suite become vulnerable to attacks like [Joux2006]. The CVEs are CVE-2016-0270, CVE-2016-10213, CVE-2016-10212, and CVE-2017-5933 🙈

Good news though: TLS 1.3 fixed this 👑 It enforces a deterministic way to generate nonces from the record sequence numbers and shared secrets, for all its AEAD cipher suites (AES-GCM included). TLS 1.2 implementations, though, could still pick their own nonce generation methods, and some of those could be insecure 😭

So here's the move 💅 TLS 1.2 implementations are RECOMMENDED to use the 64-bit sequence number to fill in the nonce_explicit part of the GCM nonce, as described in the first two paragraphs of Section 5.3 of [RFC8446]. This is a stronger recommendation than the one in Section 3 of [RFC5288], which said that using 64-bit sequence numbers for the nonce_explicit field is optional. So yes, this one updates that section 🦋

Also, at the time of writing, there are no cipher suites defined for nonce-reuse-resistant algorithms like AES-GCM-SIV [RFC8452]. Not yet, anyway 🥺

7.3. Forward Secrecy

Okay, forward secrecy (also called "perfect forward secrecy" or "PFS", and defined in [RFC4949]) is basically a defense against an attacker who records your encrypted convos, where the session keys are only encrypted with the communicating parties' long-term keys 🔐

Here's the scary part: if the attacker gets those long-term keys later, they can decrypt the session keys and therefore the entire conversation. Yikes 😱

For TLS and DTLS, long-term key compromise is honestly not that unlikely 🙃 It can happen, for example, because of:

  • A client or server getting attacked some other way, and the private key being pulled out.
  • A long-term key being pulled from a device that was sold or decommissioned without being wiped first.
  • A long-term key used on a device as a default key [Heninger2012].
  • A key generated by a trusted third party like a CA, and later retrieved from it through extortion or compromise [Soghoian2011].
  • A cryptographic breakthrough, or asymmetric keys that are too short [Kleinjung2010].
  • Social engineering attacks against system administrators.
  • Private keys collected from badly protected backups.

So forward secrecy makes sure that, in those cases, it's not feasible for an attacker to work out the session keys, even if they got the long-term keys some time after the conversation happened. It also protects against an attacker who has the long-term keys but just stays passive during the conversation 🧚

Forward secrecy is usually achieved by using the Diffie-Hellman scheme to derive session keys. Both parties keep private secrets and send parameters over the network as modular powers over certain cyclic groups. The properties of the Discrete Logarithm Problem (DLP) let the parties derive the session keys without an eavesdropper being able to do the same 🌈 There's currently no known attack against DLP if sufficiently large parameters are chosen. There's also a variant that uses elliptic curves instead of the originally proposed modular arithmetic. Given the current state of the art, Elliptic Curve Diffie-Hellman seems more efficient, permits shorter key lengths, and allows less freedom for implementation errors than finite-field Diffie-Hellman ✨

Unfortunately, a bunch of TLS/DTLS cipher suites were defined without forward secrecy, like TLS_RSA_WITH_AES_256_CBC_SHA256 😤 So this document advocates strict use of forward-secrecy-only ciphers 💖

7.4. Diffie-Hellman Exponent Reuse

For performance reasons, TLS implementations often reuse Diffie-Hellman and Elliptic Curve Diffie-Hellman exponents across multiple connections. Girl, this can cause major security issues 🚨

  • If exponents get reused for too long (in some cases, even just a few hours), an attacker who gets access to the host can decrypt previous connections. So exponent reuse basically negates forward secrecy 💀
  • TLS implementations that reuse exponents should test the DH public key they receive for group membership, to avoid some known attacks. TLS doesn't standardize these tests at the time of writing, though general guidance is in [NIST.SP.800-56A] and many protocol implementations already have it 🫶
  • Under certain conditions, using static finite-field DH keys, or ephemeral finite-field DH keys reused across multiple connections, can lead to timing attacks (like the ones described in [RACCOON]) on the shared secrets used in Diffie-Hellman key exchange ⏱️
  • An "invalid curve" attack can be mounted against Elliptic Curve DH if the victim doesn't verify that the received point lies on the correct curve. If the victim reuses the DH secrets, the attacker can repeat the probe with varying points and recover the full secret (see [Antipa2003] and [Jager2015]) 🕵️‍♀️

So, to address these concerns:

  • TLS implementations SHOULD NOT use static finite-field DH keys, and SHOULD NOT reuse ephemeral finite-field DH keys across multiple connections 🙅‍♀️
  • Server implementations that want to reuse Elliptic Curve DH keys SHOULD either use a "safe curve" [SAFECURVES] (e.g., X25519) or perform the checks described in [NIST.SP.800-56A] on the received points 🎀

7.5. Certificate Revocation

Okay, so no complete and efficient solution exists yet for checking whether common public key certificates have been revoked [RFC5280]. Still, these are the current state-of-the-art considerations and recommendations 🥹

  • Certificate revocation is a key tool for recovering from attacks on the TLS implementation, and for misissued certificates. TLS implementations MUST implement a strategy to distrust revoked certificates 🚫
  • Certificate Revocation Lists (CRLs) are the most widely supported way to distribute revocation info, but they have known scaling challenges that limit their usefulness, even with workarounds like partitioned CRLs and delta CRLs. The more modern [CRLite] and its follow-on Let's Revoke [LetsRevoke] build on Certificate Transparency [RFC9162] logs and aggressive compression to make practical use of CRL infrastructure possible. But at the time of writing, neither is deployed for client-side revocation processing at scale 🥲
  • Proprietary mechanisms that embed revocation lists in the web browser's configuration database can't scale beyond the few most heavily used web servers 🙃
  • The Online Certification Status Protocol (OCSP) [RFC6960], in its basic form, has both scaling and privacy issues 👀 Also, clients typically "soft-fail", meaning they don't abort the TLS connection if the OCSP server doesn't respond. (However, this might be a workaround to avoid denial-of-service attacks if an OCSP responder is taken offline.) For a recent survey of OCSP deployment in the web PKI, see [Chung18] 📊
  • The TLS Certificate Status Request extension (Section 8 of [RFC6066]), commonly called "OCSP stapling", fixes the operational issues with OCSP. But it's still ineffective against an active on-path attacker, because the attacker can simply ignore the client's request for a stapled OCSP response 😩
  • [RFC7633] defines a certificate extension that says clients MUST expect stapled OCSP responses for the certificate, and MUST abort the handshake ("hard-fail") if such a response isn't available 🛑
  • OCSP stapling as used in TLS 1.2 doesn't extend to intermediate certificates in a certificate chain. The Multiple Certificate Status extension [RFC6961] addresses this, but it's seen little deployment and was deprecated by [RFC8446]. So even though [RFC7525] recommended this extension for TLS 1.2, this document no longer recommends it 🥀
  • TLS 1.3 (Section 4.4.2.1 of [RFC8446]) lets you associate OCSP information with intermediate certificates, using an extension to the CertificateEntry structure. Still, using this is impractical, because many certification authorities (CAs) either don't publish OCSP for CA certificates, or publish OCSP reports with a lifetime that's too long to be useful 😭
  • Both CRLs and OCSP depend on fairly reliable Internet connectivity, which might not be available to certain kinds of nodes. A common example is newly provisioned devices that need a secure connection in order to boot up for the first time 🤖

For the common use cases of public key certificates in TLS, servers SHOULD support the following as a best practice, given the current state of the art and as a foundation for a possible future solution: OCSP [RFC6960] and OCSP stapling using the status_request extension defined in [RFC6066]. Note that the exact mechanism for embedding the status_request extension differs between TLS 1.2 and 1.3 🧩 As a matter of local policy, server operators MAY request that CAs issue must-staple [RFC7633] certificates for the server and/or for client authentication, but we recommend reviewing the operational conditions before deciding on this approach 💜

The considerations in this section do not apply to scenarios where the DNS-Based Authentication of Named Entities (DANE) TLSA resource record [RFC6698] is used to signal to a client which certificate a server considers valid and good to use for TLS connections. 🌸

8. References

8.1. Normative References 💖

okay so these are the receipts bestie, the stuff the spec leans on to work 🫶 these are the normative refs, and you literally can't skip them 🔥

8.2. Informative References 🌈

these ones are the bonus reading, the research and drafts that explain why stuff works the way it does 🧚 lowkey my favourite section ngl 😭

Okay bestie, here's the rest of the informative reading list 📚✨ Every citation is copied exactly as written, because these are the receipts 🫶

   [Logjam]   Adrian, D., Bhargavan, K., Durumeric, Z., Gaudry, P.,
              Green, M., Halderman, J., Heninger, N., Springall, D.,
              Thomé, E., Valenta, L., VanderSloot, B., Wustrow, E.,
              Zanella-Béguelin, S., and P. Zimmermann, "Imperfect
              Forward Secrecy: How Diffie-Hellman Fails in Practice",
              Proceedings of the 22nd ACM SIGSAC Conference on Computer
              and Communications Security, pp. 5-17,
              DOI 10.1145/2810103.2813707, October 2015,
              <https://doi.org/10.1145/2810103.2813707>.

   [Multiple-Encryption]
              Merkle, R. and M. Hellman, "On the security of multiple
              encryption", Communications of the ACM, Vol. 24, Issue 7,
              pp. 465-467, DOI 10.1145/358699.358718, July 1981,
              <https://doi.org/10.1145/358699.358718>.

   [NIST.SP.800-56A]
              National Institute of Standards and Technology,
              "Recommendation for Pair-Wise Key-Establishment Schemes
              Using Discrete Logarithm Cryptography", Revision 3, NIST
              Special Publication 800-56A,
              DOI 10.6028/NIST.SP.800-56Ar3, April 2018,
              <https://doi.org/10.6028/NIST.SP.800-56Ar3>.

   [PatersonRS11]
              Paterson, K., Ristenpart, T., and T. Shrimpton, "Tag Size
              Does Matter: Attacks and Proofs for the TLS Record
              Protocol", Proceedings of the 17th International
              conference on The Theory and Application of Cryptology and
              Information Security, pp. 372-389,
              DOI 10.1007/978-3-642-25385-0_20, December 2011,
              <https://doi.org/10.1007/978-3-642-25385-0_20>.

   [Poddebniak2017]
              Poddebniak, D., Somorovsky, J., Schinzel, S., Lochter, M.,
              and P. Rösler, "Attacking Deterministic Signature Schemes
              using Fault Attacks", Conference: 2018 IEEE European
              Symposium on Security and Privacy,
              DOI 10.1109/EuroSP.2018.00031, April 2018,
              <https://eprint.iacr.org/2017/1014.pdf>.

   [POODLE]   US-CERT, "SSL 3.0 Protocol Vulnerability and POODLE
              Attack", October 2014,
              <https://www.us-cert.gov/ncas/alerts/TA14-290A>.

   [RACCOON]  Merget, R., Brinkmann, M., Aviram, N., Somorovsky, J.,
              Mittmann, J., and J. Schwenk, "Raccoon Attack: Finding and
              Exploiting Most-Significant-Bit-Oracles in TLS-DH(E)",
              30th USENIX Security Symposium (USENIX Security 21), 2021,
              <https://www.usenix.org/conference/usenixsecurity21/
              presentation/merget>.

   [RFC2026]  Bradner, S., "The Internet Standards Process -- Revision
              3", BCP 9, RFC 2026, DOI 10.17487/RFC2026, October 1996,
              <https://www.rfc-editor.org/info/rfc2026>.

   [RFC2246]  Dierks, T. and C. Allen, "The TLS Protocol Version 1.0",
              RFC 2246, DOI 10.17487/RFC2246, January 1999,
              <https://www.rfc-editor.org/info/rfc2246>.

   [RFC3261]  Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston,
              A., Peterson, J., Sparks, R., Handley, M., and E.
              Schooler, "SIP: Session Initiation Protocol", RFC 3261,
              DOI 10.17487/RFC3261, June 2002,
              <https://www.rfc-editor.org/info/rfc3261>.

   [RFC3602]  Frankel, S., Glenn, R., and S. Kelly, "The AES-CBC Cipher
              Algorithm and Its Use with IPsec", RFC 3602,
              DOI 10.17487/RFC3602, September 2003,
              <https://www.rfc-editor.org/info/rfc3602>.

   [RFC4346]  Dierks, T. and E. Rescorla, "The Transport Layer Security
              (TLS) Protocol Version 1.1", RFC 4346,
              DOI 10.17487/RFC4346, April 2006,
              <https://www.rfc-editor.org/info/rfc4346>.

   [RFC4347]  Rescorla, E. and N. Modadugu, "Datagram Transport Layer
              Security", RFC 4347, DOI 10.17487/RFC4347, April 2006,
              <https://www.rfc-editor.org/info/rfc4347>.

   [RFC4949]  Shirey, R., "Internet Security Glossary, Version 2",
              FYI 36, RFC 4949, DOI 10.17487/RFC4949, August 2007,
              <https://www.rfc-editor.org/info/rfc4949>.

   [RFC5077]  Salowey, J., Zhou, H., Eronen, P., and H. Tschofenig,
              "Transport Layer Security (TLS) Session Resumption without
              Server-Side State", RFC 5077, DOI 10.17487/RFC5077,
              January 2008, <https://www.rfc-editor.org/info/rfc5077>.

   [RFC5116]  McGrew, D., "An Interface and Algorithms for Authenticated
              Encryption", RFC 5116, DOI 10.17487/RFC5116, January 2008,
              <https://www.rfc-editor.org/info/rfc5116>.

   [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>.

   [RFC5321]  Klensin, J., "Simple Mail Transfer Protocol", RFC 5321,
              DOI 10.17487/RFC5321, October 2008,
              <https://www.rfc-editor.org/info/rfc5321>.

   [RFC6101]  Freier, A., Karlton, P., and P. Kocher, "The Secure
              Sockets Layer (SSL) Protocol Version 3.0", RFC 6101,
              DOI 10.17487/RFC6101, August 2011,
              <https://www.rfc-editor.org/info/rfc6101>.

   [RFC6120]  Saint-Andre, P., "Extensible Messaging and Presence
              Protocol (XMPP): Core", RFC 6120, DOI 10.17487/RFC6120,
              March 2011, <https://www.rfc-editor.org/info/rfc6120>.

   [RFC6698]  Hoffman, P. and J. Schlyter, "The DNS-Based Authentication
              of Named Entities (DANE) Transport Layer Security (TLS)
              Protocol: TLSA", RFC 6698, DOI 10.17487/RFC6698, August
              2012, <https://www.rfc-editor.org/info/rfc6698>.

   [RFC6797]  Hodges, J., Jackson, C., and A. Barth, "HTTP Strict
              Transport Security (HSTS)", RFC 6797,
              DOI 10.17487/RFC6797, November 2012,
              <https://www.rfc-editor.org/info/rfc6797>.

   [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>.

   [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>.

   [RFC7457]  Sheffer, Y., Holz, R., and P. Saint-Andre, "Summarizing
              Known Attacks on Transport Layer Security (TLS) and
              Datagram TLS (DTLS)", RFC 7457, DOI 10.17487/RFC7457,
              February 2015, <https://www.rfc-editor.org/info/rfc7457>.

   [RFC7507]  Moeller, B. and A. Langley, "TLS Fallback Signaling Cipher
              Suite Value (SCSV) for Preventing Protocol Downgrade
              Attacks", RFC 7507, DOI 10.17487/RFC7507, April 2015,
              <https://www.rfc-editor.org/info/rfc7507>.

   [RFC7525]  Sheffer, Y., Holz, R., and P. Saint-Andre,
              "Recommendations for Secure Use of Transport Layer
              Security (TLS) and Datagram Transport Layer Security
              (DTLS)", BCP 195, RFC 7525, DOI 10.17487/RFC7525, May
              2015, <https://www.rfc-editor.org/info/rfc7525>.

   [RFC7568]  Barnes, R., Thomson, M., Pironti, A., and A. Langley,
              "Deprecating Secure Sockets Layer Version 3.0", RFC 7568,
              DOI 10.17487/RFC7568, June 2015,
              <https://www.rfc-editor.org/info/rfc7568>.

   [RFC7590]  Saint-Andre, P. and T. Alkemade, "Use of Transport Layer
              Security (TLS) in the Extensible Messaging and Presence
              Protocol (XMPP)", RFC 7590, DOI 10.17487/RFC7590, June
              2015, <https://www.rfc-editor.org/info/rfc7590>.

   [RFC7633]  Hallam-Baker, P., "X.509v3 Transport Layer Security (TLS)
              Feature Extension", RFC 7633, DOI 10.17487/RFC7633,
              October 2015, <https://www.rfc-editor.org/info/rfc7633>.

   [RFC7672]  Dukhovni, V. and W. Hardaker, "SMTP Security via
              Opportunistic DNS-Based Authentication of Named Entities
              (DANE) Transport Layer Security (TLS)", RFC 7672,
              DOI 10.17487/RFC7672, October 2015,
              <https://www.rfc-editor.org/info/rfc7672>.

   [RFC7673]  Finch, T., Miller, M., and P. Saint-Andre, "Using DNS-
              Based Authentication of Named Entities (DANE) TLSA Records
              with SRV Records", RFC 7673, DOI 10.17487/RFC7673, October
              2015, <https://www.rfc-editor.org/info/rfc7673>.

   [RFC7712]  Saint-Andre, P., Miller, M., and P. Hancke, "Domain Name
              Associations (DNA) in the Extensible Messaging and
              Presence Protocol (XMPP)", RFC 7712, DOI 10.17487/RFC7712,
              November 2015, <https://www.rfc-editor.org/info/rfc7712>.

   [RFC7919]  Gillmor, D., "Negotiated Finite Field Diffie-Hellman
              Ephemeral Parameters for Transport Layer Security (TLS)",
              RFC 7919, DOI 10.17487/RFC7919, August 2016,
              <https://www.rfc-editor.org/info/rfc7919>.

   [RFC7924]  Santesson, S. and H. Tschofenig, "Transport Layer Security
              (TLS) Cached Information Extension", RFC 7924,
              DOI 10.17487/RFC7924, July 2016,
              <https://www.rfc-editor.org/info/rfc7924>.

   [RFC7925]  Tschofenig, H., Ed. and T. Fossati, "Transport Layer
              Security (TLS) / Datagram Transport Layer Security (DTLS)
              Profiles for the Internet of Things", RFC 7925,
              DOI 10.17487/RFC7925, July 2016,
              <https://www.rfc-editor.org/info/rfc7925>.

   [RFC8452]  Gueron, S., Langley, A., and Y. Lindell, "AES-GCM-SIV:
              Nonce Misuse-Resistant Authenticated Encryption",
              RFC 8452, DOI 10.17487/RFC8452, April 2019,
              <https://www.rfc-editor.org/info/rfc8452>.

   [RFC8461]  Margolis, D., Risher, M., Ramakrishnan, B., Brotman, A.,
              and J. Jones, "SMTP MTA Strict Transport Security (MTA-
              STS)", RFC 8461, DOI 10.17487/RFC8461, September 2018,
              <https://www.rfc-editor.org/info/rfc8461>.

   [RFC8470]  Thomson, M., Nottingham, M., and W. Tarreau, "Using Early
              Data in HTTP", RFC 8470, DOI 10.17487/RFC8470, September
              2018, <https://www.rfc-editor.org/info/rfc8470>.

   [RFC8879]  Ghedini, A. and V. Vasiliev, "TLS Certificate
              Compression", RFC 8879, DOI 10.17487/RFC8879, December
              2020, <https://www.rfc-editor.org/info/rfc8879>.

   [RFC9000]  Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based
              Multiplexed and Secure Transport", RFC 9000,
              DOI 10.17487/RFC9000, May 2021,
              <https://www.rfc-editor.org/info/rfc9000>.

   [RFC9001]  Thomson, M., Ed. and S. Turner, Ed., "Using TLS to Secure
              QUIC", RFC 9001, DOI 10.17487/RFC9001, May 2021,
              <https://www.rfc-editor.org/info/rfc9001>.

Appendix A. Differences from RFC 7525 📜✨

Okay bestie, this revision of the Best Current Practices has SO many changes 💅 This appendix is all about the normative ones, the stuff that actually changes what you're required to do 🫶

  • High-level differences:
    • Described the expectations from new TLS-incorporating transport protocols and from new application protocols layered on TLS.
    • Clarified items (e.g., renegotiation) that only apply to TLS 1.2.
    • Changed the status of TLS 1.0 and 1.1 from "SHOULD NOT" to "MUST NOT".
    • Added TLS 1.3 at a "SHOULD" level.
    • Made similar changes to DTLS.
    • Included specific guidance for multiplexed protocols.
    • MUST-level implementation requirement for ALPN and more specific SHOULD-level guidance for ALPN and SNI.
    • Clarified discussion of strict TLS policies, including MUST-level recommendations.
    • Limits on key usage.
    • New attacks since [RFC7457]: ALPACA, Raccoon, Logjam, and "Nonce-Disrespecting Adversaries".
    • RFC 6961 (OCSP status_request_v2) has been deprecated.
    • MUST-level requirement for server-side RSA certificates to have a 2048-bit modulus at a minimum, replacing a "SHOULD".
  • Differences specific to TLS 1.2:
    • SHOULD-level guidance on AES-GCM nonce generation.
    • SHOULD NOT use (static or ephemeral) finite-field DH key agreement.
    • SHOULD NOT reuse ephemeral finite-field DH keys across multiple connections.
    • SHOULD NOT use static Elliptic Curve DH key exchange.
    • 2048-bit DH is now a "MUST" and ECDH minimal curve size is 224 (vs. 192 previously).
    • Support for extended_master_secret is now a "MUST" (previously it was a soft recommendation, as the RFC had not been published at the time). Also removed other, more complicated, related mitigations.
    • MUST-level restriction on session ticket validity, replacing a "SHOULD".
    • SHOULD-level restriction on the TLS session duration, depending on the rotation period of an [RFC5077] ticket key.
    • Dropped TLS_DHE_RSA_WITH_AES from the recommended ciphers.
    • Added TLS_ECDHE_ECDSA_WITH_AES to the recommended ciphers.
    • SHOULD NOT use the old MTI cipher suite, TLS_RSA_WITH_AES_128_CBC_SHA.
    • Recommended curve X25519 alongside NIST P-256.
  • Differences specific to TLS 1.3:
    • New TLS 1.3 capabilities: 0-RTT.
    • Removed capabilities: renegotiation and compression.
    • Added mention of TLS Encrypted Client Hello, but no recommendation for use until it is finalized.
    • SHOULD-level requirement for forward secrecy in TLS 1.3 session resumption.
    • Generic MUST-level guidance to avoid 0-RTT unless it is documented for the particular protocol.

Acknowledgments 🌸

Huge thanks to Alexey Melnikov, Alvaro Retana, Andrei Popov, Ben Kaduk, Christian Huitema, Corey Bonnell, Cullen Jennings, Daniel Kahn Gillmor, David Benjamin, Eric Rescorla, Éric Vyncke, Francesca Palombini, Hannes Tschofenig, Hubert Kario, Ilari Liusvaara, John Preuß Mattsson, John R. Levine, Julien Élie, Lars Eggert, Leif Johansson, Magnus Westerlund, Martin Duke, Martin Thomson, Mohit Sahni, Nick Sullivan, Nimrod Aviram, Paul Wouters, Peter Gutmann, Rich Salz, Robert Sayre, Robert Wilton, Roman Danyliw, Ryan Sleevi, Sean Turner, Stephen Farrell, Tim Evans, Valery Smyslov, Viktor Dukhovni, and Warren Kumari for the helpful comments and discussions that shaped this document 💖

The authors gratefully acknowledge the contribution of Ralph Holz, who was a coauthor of RFC 7525, the previous version of the TLS recommendations 🫶

See RFC 7525 for additional acknowledgments specific to the previous version of the TLS recommendations 🧚

Authors' Addresses 🦋

Yaron Sheffer
Intuit
Email: yaronf.ietf@gmail.com

Peter Saint-Andre
Independent
Email: stpeter@stpeter.im

Thomas Fossati
ARM Limited
Email: thomas.fossati@arm.com
📦 where this page comes from