en-US/about_MTA-STS.help.txt

TOPIC
   about_MTA-STS


Internet Engineering Task Force (IETF) D. Margolis
Request for Comments: 8461 M. Risher
Category: Standards Track Google, Inc.
ISSN: 2070-1721 B. Ramakrishnan
                                                              Oath, Inc.
                                                              A. Brotman
                                                           Comcast, Inc.
                                                                J. Jones
                                                         Microsoft, Inc.
                                                          September 2018


              SMTP MTA Strict Transport Security (MTA-STS)

Abstract

   SMTP MTA Strict Transport Security (MTA-STS) is a mechanism enabling
   mail service providers (SPs) to declare their ability to receive
   Transport Layer Security (TLS) secure SMTP connections and to specify
   whether sending SMTP servers should refuse to deliver to MX hosts
   that do not offer TLS with a trusted server certificate.

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/rfc8461.
















Margolis, et al. Standards Track [Page 1]
 
RFC 8461 MTA-STS September 2018


Copyright Notice

   Copyright (c) 2018 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.





































Margolis, et al. Standards Track [Page 2]
 
RFC 8461 MTA-STS September 2018


Table of Contents

   1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4
     1.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 4
   2. Related Technologies . . . . . . . . . . . . . . . . . . . . 5
   3. Policy Discovery . . . . . . . . . . . . . . . . . . . . . . 5
     3.1. MTA-STS TXT Records . . . . . . . . . . . . . . . . . . . 6
     3.2. MTA-STS Policies . . . . . . . . . . . . . . . . . . . . 7
     3.3. HTTPS Policy Fetching . . . . . . . . . . . . . . . . . . 10
     3.4. Policy Selection for Smart Hosts and Subdomains . . . . . 11
   4. Policy Validation . . . . . . . . . . . . . . . . . . . . . . 11
     4.1. MX Host Validation . . . . . . . . . . . . . . . . . . . 12
     4.2. Recipient MTA Certificate Validation . . . . . . . . . . 12
   5. Policy Application . . . . . . . . . . . . . . . . . . . . . 12
     5.1. Policy Application Control Flow . . . . . . . . . . . . . 13
   6. Reporting Failures . . . . . . . . . . . . . . . . . . . . . 13
   7. Interoperability Considerations . . . . . . . . . . . . . . . 14
     7.1. SNI Support . . . . . . . . . . . . . . . . . . . . . . . 14
     7.2. Minimum TLS Version Support . . . . . . . . . . . . . . . 14
   8. Operational Considerations . . . . . . . . . . . . . . . . . 15
     8.1. Policy Updates . . . . . . . . . . . . . . . . . . . . . 15
     8.2. Policy Delegation . . . . . . . . . . . . . . . . . . . . 15
     8.3. Removing MTA-STS . . . . . . . . . . . . . . . . . . . . 16
     8.4. Preserving MX Candidate Traversal . . . . . . . . . . . . 17
   9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 17
     9.1. Well-Known URIs Registry . . . . . . . . . . . . . . . . 17
     9.2. MTA-STS TXT Record Fields . . . . . . . . . . . . . . . . 17
     9.3. MTA-STS Policy Fields . . . . . . . . . . . . . . . . . . 18
   10. Security Considerations . . . . . . . . . . . . . . . . . . . 18
     10.1. Obtaining a Signed Certificate . . . . . . . . . . . . . 18
     10.2. Preventing Policy Discovery . . . . . . . . . . . . . . 19
     10.3. Denial of Service . . . . . . . . . . . . . . . . . . . 19
     10.4. Weak Policy Constraints . . . . . . . . . . . . . . . . 20
     10.5. Compromise of the Web PKI System . . . . . . . . . . . . 20
   11. References . . . . . . . . . . . . . . . . . . . . . . . . . 21
     11.1. Normative References . . . . . . . . . . . . . . . . . . 21
     11.2. Informative References . . . . . . . . . . . . . . . . . 23
   Appendix A. MTA-STS Example Record and Policy . . . . . . . . . 25
   Appendix B. Message Delivery Pseudocode . . . . . . . . . . . . 25
   Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . 28
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 29










Margolis, et al. Standards Track [Page 3]
 
RFC 8461 MTA-STS September 2018


1. Introduction

   The STARTTLS extension to SMTP [RFC3207] allows SMTP clients and
   hosts to negotiate the use of a TLS channel for encrypted mail
   transmission.

   While this opportunistic encryption protocol by itself provides a
   high barrier against passive man-in-the-middle traffic interception,
   any attacker who can delete parts of the SMTP session (such as the
   "250 STARTTLS" response) or who can redirect the entire SMTP session
   (perhaps by overwriting the resolved MX record of the delivery
   domain) can perform downgrade or interception attacks.

   This document defines a mechanism for recipient domains to publish
   policies, via a combination of DNS and HTTPS, specifying:

   o whether MTAs sending mail to this domain can expect PKIX-
      authenticated TLS support

   o what a conforming client should do with messages when TLS cannot
      be successfully negotiated

1.1. Terminology

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

   We also define the following terms for further use in this document:

   o MTA-STS Policy: A commitment by the Policy Domain to support TLS
      authenticated with PKIX [RFC5280] for the specified MX hosts.

   o Policy Domain: The domain for which an MTA-STS Policy is defined.
      This is the next-hop domain; when sending mail to
      "alice@example.com", this would ordinarily be "example.com", but
      this may be overridden by explicit routing rules (as described in
      Section 3.4, "Policy Selection for Smart Hosts and Subdomains").

   o Policy Host: The HTTPS host that serves the MTA-STS Policy for a
      Policy Domain. Rules for constructing the hostname are described
      in Section 3.2, "MTA-STS Policies".

   o Sender or Sending MTA: The SMTP MTA sending an email message.





Margolis, et al. Standards Track [Page 4]
 
RFC 8461 MTA-STS September 2018


   o ABNF: Augmented Backus-Naur Form, a syntax for formally specifying
      syntax, defined in [RFC5234] and [RFC7405].

2. Related Technologies

   The DNS-Based Authentication of a Named Entities (DANE) TLSA record
   [RFC7672] is similar, in that DANE is also designed to upgrade
   unauthenticated encryption or plaintext transmission into
   authenticated, downgrade-resistant encrypted transmission. DANE
   requires DNSSEC [RFC4033] for authentication; the mechanism described
   here instead relies on certification authorities (CAs) and does not
   require DNSSEC, at a cost of risking malicious downgrades. For a
   thorough discussion of this trade-off, see Section 10, "Security
   Considerations".

   In addition, MTA-STS provides an optional testing-only mode, enabling
   soft deployments to detect policy failures; partial deployments can
   be achieved in DANE by deploying TLSA records only for some of a
   domain's MXes, but such a mechanism is not possible for the per-
   domain policies used by MTA-STS.

   The primary motivation of MTA-STS is to provide a mechanism for
   domains to ensure transport security even when deploying DNSSEC is
   undesirable or impractical. However, MTA-STS is designed not to
   interfere with DANE deployments when the two overlap; in particular,
   senders who implement MTA-STS validation MUST NOT allow MTA-STS
   Policy validation to override a failing DANE validation.

3. Policy Discovery

   MTA-STS policies are distributed via HTTPS from a "well-known"
   [RFC5785] path served within the Policy Domain, and their presence
   and current version are indicated by a TXT record at the Policy
   Domain. These TXT records additionally contain a policy "id" field,
   allowing Sending MTAs to check that a cached policy is still current
   without performing an HTTPS request.

   To discover if a recipient domain implements MTA-STS, a sender need
   only resolve a single TXT record. To see if an updated policy is
   available for a domain for which the sender has a previously cached
   policy, the sender need only check the TXT record's version "id"
   against the cached value.









Margolis, et al. Standards Track [Page 5]
 
RFC 8461 MTA-STS September 2018


3.1. MTA-STS TXT Records

   The MTA-STS TXT record is a TXT record with the name "_mta-sts" at
   the Policy Domain. For the domain "example.com", this record would
   be "_mta-sts.example.com". MTA-STS TXT records MUST be US-ASCII,
   semicolon-separated key/value pairs containing the following fields:

   o "v" (plaintext, required): Currently, only "STSv1" is supported.

   o "id" (plaintext, required): A short string used to track policy
      updates. This string MUST uniquely identify a given instance of a
      policy, such that senders can determine when the policy has been
      updated by comparing to the "id" of a previously seen policy.
      There is no implied ordering of "id" fields between revisions.

   An example TXT record is as below:

   _mta-sts.example.com. IN TXT "v=STSv1; id=20160831085700Z;"

   The formal definition of the "_mta-sts" TXT record, defined using
   ABNF [RFC7405], is as follows:

   sts-text-record = sts-version 1*(sts-field-delim sts-field)
                     [sts-field-delim]

   sts-field = sts-id / ; Note that sts-id record
                     sts-extension ; is required.

   sts-field-delim = *WSP ";" *WSP

   sts-version = %s"v=STSv1"

   sts-id = %s"id=" 1*32(ALPHA / DIGIT) ; id=...

   sts-extension = sts-ext-name "=" sts-ext-value ; name=value

   sts-ext-name = (ALPHA / DIGIT)
                     *31(ALPHA / DIGIT / "_" / "-" / ".")

   sts-ext-value = 1*(%x21-3A / %x3C / %x3E-7E)
                     ; chars excluding "=", ";", SP, and CTLs

   The TXT record MUST begin with the sts-version field; the order of
   other fields is not significant. If multiple TXT records for
   "_mta-sts" are returned by the resolver, records that do not begin
   with "v=STSv1;" are discarded. If the number of resulting records is
   not one, or if the resulting record is syntactically invalid, senders
   MUST assume the recipient domain does not have an available MTA-STS



Margolis, et al. Standards Track [Page 6]
 
RFC 8461 MTA-STS September 2018


   Policy and skip the remaining steps of policy discovery. (Note that
   the absence of a usable TXT record is not by itself sufficient to
   remove a sender's previously cached policy for the Policy Domain, as
   discussed in Section 5.1, "Policy Application Control Flow".) If the
   resulting TXT record contains multiple strings, then the record MUST
   be treated as if those strings are concatenated without adding
   spaces.

   The "_mta-sts" record MAY return a CNAME that points (directly or via
   other CNAMEs) to a TXT record, in which case senders MUST follow the
   CNAME pointers. This can be used for policy delegation, as described
   in Section 8.2.

3.2. MTA-STS Policies

   The policy itself is a set of key/value pairs (similar to header
   fields in [RFC5322]) served via the HTTPS GET method from the fixed
   "well-known" [RFC5785] path of ".well-known/mta-sts.txt" served by
   the Policy Host. The Policy Host DNS name is constructed by
   prepending "mta-sts" to the Policy Domain.

   Thus, for a Policy Domain of "example.com", the full URL is
   "https://mta-sts.example.com/.well-known/mta-sts.txt".

   When fetching a policy, senders SHOULD validate that the media type
   is "text/plain" to guard against cases where web servers allow
   untrusted users to host non-text content (typically, HTML or images)
   at a user-defined path. All parameters other than charset=utf-8 or
   charset=us-ascii are ignored. Additional "Content-Type" parameters
   are also ignored.

   This resource contains the following CRLF-separated key/value pairs:

   o "version": Currently, only "STSv1" is supported.

   o "mode": One of "enforce", "testing", or "none", indicating the
      expected behavior of a Sending MTA in the case of a policy
      validation failure. See Section 5, "Policy Application", for more
      details about the three modes.

   o "max_age": Max lifetime of the policy (plaintext non-negative
      integer seconds, maximum value of 31557600). Well-behaved clients
      SHOULD cache a policy for up to this value from the last policy
      fetch time. To mitigate the risks of attacks at policy refresh
      time, it is expected that this value typically be in the range of
      weeks or greater.





Margolis, et al. Standards Track [Page 7]
 
RFC 8461 MTA-STS September 2018


   o "mx": Allowed MX patterns. One or more patterns matching allowed
      MX hosts for the Policy Domain. As an example,

                        mx: mail.example.com <CRLF>
                        mx: *.example.net

   indicates that mail for this domain might be handled by MX
   "mail.example.com" or any MX at "example.net". Valid patterns can be
   either fully specified names ("example.com") or suffixes prefixed by
   a wildcard ("*.example.net"). If a policy specifies more than one
   MX, each MX MUST have its own "mx:" key, and each MX key/value pair
   MUST be on its own line in the policy file. In the case of
   Internationalized Domain Names [RFC5891], the "mx" value MUST specify
   the Punycode-encoded A-label [RFC3492] to match against, and not the
   Unicode-encoded U-label. The full semantics of certificate
   validation (including the use of wildcard patterns) are described in
   Section 4.1, "MX Host Validation".

   An example policy is as below:

                         version: STSv1
                         mode: enforce
                         mx: mail.example.com
                         mx: *.example.net
                         mx: backupmx.example.com
                         max_age: 604800

   The formal definition of the policy resource, defined using ABNF
   [RFC7405], is as follows:

sts-policy-record = sts-policy-field *WSP
                           *(sts-policy-term sts-policy-field *WSP)
                           [sts-policy-term]

sts-policy-field = sts-policy-version / ; required once
                           sts-policy-mode / ; required once
                           sts-policy-max-age / ; required once
                           sts-policy-mx /
                           ; required at least once, except when
                           ; mode is "none"
                           sts-policy-extension ; other fields

sts-policy-field-delim = ":" *WSP

sts-policy-version = sts-policy-version-field sts-policy-field-delim
                         sts-policy-version-value

sts-policy-version-field = %s"version"



Margolis, et al. Standards Track [Page 8]
 
RFC 8461 MTA-STS September 2018


sts-policy-version-value = %s"STSv1"

sts-policy-mode = sts-policy-mode-field sts-policy-field-delim
                           sts-policy-mode-value

sts-policy-mode-field = %s"mode"

sts-policy-mode-value = %s"testing" / %s"enforce" / %s"none"

sts-policy-mx = sts-policy-mx-field sts-policy-field-delim
                           sts-policy-mx-value

sts-policy-mx-field = %s"mx"

sts-policy-mx-value = ["*."] Domain

sts-policy-max-age = sts-policy-max-age-field sts-policy-field-delim
                         sts-policy-max-age-value

sts-policy-max-age-field = %s"max_age"

sts-policy-max-age-value = 1*10(DIGIT)

sts-policy-extension = sts-policy-ext-name ; additional
                           sts-policy-field-delim ; extension
                           sts-policy-ext-value ; fields

sts-policy-ext-name = (sts-policy-alphanum)
                           *31(sta-policy-alphanum / "_" / "-" / ".")

sts-policy-term = LF / CRLF

sts-policy-ext-value = sts-policy-vchar
                           [*(%x20 / sts-policy-vchar)
                           sts-policy-vchar]
                           ; chars, including UTF-8 [RFC3629],
                           ; excluding CTLs and no
                           ; leading/trailing spaces

sts-policy-alphanum = ALPHA / DIGIT

sts-policy-vchar = %x21-7E / UTF8-2 / UTF8-3 / UTF8-4

UTF8-2 = <Defined in Section 4 of [RFC3629]>

UTF8-3 = <Defined in Section 4 of [RFC3629]>

UTF8-4 = <Defined in Section 4 of [RFC3629]>



Margolis, et al. Standards Track [Page 9]
 
RFC 8461 MTA-STS September 2018


Domain = <Defined in Section 4.1.2 of [RFC5321]>

   Parsers MUST accept TXT records and policy files that are
   syntactically valid (i.e., valid key/value pairs separated by
   semicolons for TXT records), possibly containing additional key/value
   pairs not specified in this document, in which case unknown fields
   SHALL be ignored. If any non-repeated field -- i.e., all fields
   excepting "mx" -- is duplicated, all entries except for the first
   SHALL be ignored.

3.3. HTTPS Policy Fetching

   Policy bodies are, as described above, retrieved by Sending MTAs via
   HTTPS [RFC2818]. During the TLS handshake initiated to fetch a new
   or updated policy from the Policy Host, the Policy Host HTTPS server
   MUST present an X.509 certificate that is valid for the "mta-sts"
   DNS-ID [RFC6125] (e.g., "mta-sts.example.com") as described below,
   chain to a root CA that is trusted by the Sending MTA, and be non-
   expired. It is expected that Sending MTAs use a set of trusted CAs
   similar to those in widely deployed web browsers and operating
   systems. See [RFC5280] for more details about certificate
   verification.

   The certificate is valid for the Policy Host (i.e., "mta-sts"
   prepended to the Policy Domain) with respect to the rules described
   in [RFC6125], with the following application-specific considerations:

   o Matching is performed only against the DNS-ID identifiers.

   o DNS domain names in server certificates MAY contain the wildcard
      character '*' as the complete left-most label within the
      identifier.

   The certificate MAY be checked for revocation via the Online
   Certificate Status Protocol (OCSP) [RFC6960], certificate revocation
   lists (CRLs), or some other mechanism.

   Policies fetched via HTTPS are only valid if the HTTP response code
   is 200 (OK). HTTP 3xx redirects MUST NOT be followed, and HTTP
   caching (as specified in [RFC7234]) MUST NOT be used.

   Senders may wish to rate-limit the frequency of attempts to fetch the
   HTTPS endpoint even if a valid TXT record for the recipient domain
   exists. In the case where the HTTPS GET fails, implementers SHOULD
   limit further attempts to a period of five minutes or longer per
   version ID, to avoid overwhelming resource-constrained recipients
   with cascading failures.




Margolis, et al. Standards Track [Page 10]
 
RFC 8461 MTA-STS September 2018


   Senders MAY impose a timeout on the HTTPS GET and/or a limit on the
   maximum size of the response body to avoid long delays or resource
   exhaustion during attempted policy updates. A suggested timeout is
   one minute, and a suggested maximum policy size is 64 kilobytes;
   Policy Hosts SHOULD respond to requests with a complete policy body
   within that timeout and size limit.

   If a valid TXT record is found but no policy can be fetched via HTTPS
   (for any reason), and there is no valid (non-expired) previously
   cached policy, senders MUST continue with delivery as though the
   domain has not implemented MTA-STS.

   Conversely, if no "live" policy can be discovered via DNS or fetched
   via HTTPS, but a valid (non-expired) policy exists in the sender's
   cache, the sender MUST apply that cached policy.

   Finally, to mitigate the risk of persistent interference with policy
   refresh, as discussed in-depth in Section 10, MTAs SHOULD proactively
   refresh cached policies before they expire; a suggested refresh
   frequency is once per day. To enable administrators to discover
   problems with policy refresh, MTAs SHOULD alert administrators
   (through the use of logs or similar) when such attempts fail, unless
   the cached policy mode is "none".

3.4. Policy Selection for Smart Hosts and Subdomains

   When sending mail via a "smart host" -- an administratively
   configured intermediate SMTP relay, which is different from the
   message recipient's server as determined from DNS -- compliant
   senders MUST treat the smart host domain as the Policy Domain for the
   purposes of policy discovery and application. This specification
   does not provide a means of associating policies with email addresses
   that employ Address Literals [RFC5321].

   When sending mail to a mailbox at a subdomain, compliant senders MUST
   NOT attempt to fetch a policy from the parent zone. Thus, for mail
   sent to "user@mail.example.com", the policy can be fetched only from
   "mail.example.com", not "example.com".

4. Policy Validation

   When sending to an MX at a domain for which the sender has a valid
   and non-expired MTA-STS Policy, a Sending MTA honoring MTA-STS MUST
   check whether:

   1. At least one of the policy's "mx" patterns matches the selected
       MX host, as described in Section 4.1, "MX Host Validation".




Margolis, et al. Standards Track [Page 11]
 
RFC 8461 MTA-STS September 2018


   2. The recipient mail server supports STARTTLS and offers a PKIX-
       based TLS certificate, during TLS handshake, which is valid for
       that host, as described in Section 4.2, "Recipient MTA
       Certificate Validation".

   When these conditions are not met, a policy is said to fail to
   validate. This section does not dictate the behavior of Sending MTAs
   when the above conditions are not met; see Section 5, "Policy
   Application", for a description of Sending MTA behavior when policy
   validation fails.

4.1. MX Host Validation

   A receiving candidate MX host is valid according to an applied MTA-
   STS Policy if the MX record name matches one or more of the "mx"
   fields in the applied policy. Matching is identical to the rules
   given in [RFC6125], with the restriction that the wildcard character
   '*' may only be used to match the entire left-most label in the
   presented identifier. Thus, the mx pattern "*.example.com" matches
   "mail.example.com" but not "example.com" or "foo.bar.example.com".

4.2. Recipient MTA Certificate Validation

   The certificate presented by the receiving MTA MUST not be expired
   and MUST chain to a root CA that is trusted by the Sending MTA. The
   certificate MUST have a subject alternative name (SAN) [RFC5280] with
   a DNS-ID [RFC6125] matching the hostname, per the rules given in
   [RFC6125]. The MX's certificate MAY also be checked for revocation
   via OCSP [RFC6960], CRLs [RFC6818], or some other mechanism.

5. Policy Application

   When sending to an MX at a domain for which the sender has a valid,
   non-expired MTA-STS Policy, a Sending MTA honoring MTA-STS applies
   the result of a policy validation failure in one of two ways,
   depending on the value of the policy "mode" field:

   1. "enforce": In this mode, Sending MTAs MUST NOT deliver the
       message to hosts that fail MX matching or certificate validation
       or that do not support STARTTLS.

   2. "testing": In this mode, Sending MTAs that also implement the
       TLSRPT (TLS Reporting) specification [RFC8460] send a report
       indicating policy application failures (as long as TLSRPT is also
       implemented by the recipient domain); in any case, messages may
       be delivered as though there were no MTA-STS validation failure.





Margolis, et al. Standards Track [Page 12]
 
RFC 8461 MTA-STS September 2018


   3. "none": In this mode, Sending MTAs should treat the Policy Domain
       as though it does not have any active policy; see Section 8.3,
       "Removing MTA-STS", for use of this mode value.

   When a message fails to deliver due to an "enforce" policy, a
   compliant MTA MUST NOT permanently fail to deliver messages before
   checking, via DNS, for the presence of an updated policy at the
   Policy Domain. (In all cases, MTAs SHOULD treat such failures as
   transient errors and retry delivery later.) This allows implementing
   domains to update long-lived policies on the fly.

5.1. Policy Application Control Flow

   An example control flow for a compliant sender consists of the
   following steps:

   1. Check for a cached policy whose time-since-fetch has not exceeded
       its "max_age". If none exists, attempt to fetch a new policy
       (perhaps asynchronously, so as not to block message delivery).
       Optionally, Sending MTAs may unconditionally check for a new
       policy at this step.

   2. For each candidate MX, in order of MX priority, attempt to
       deliver the message. If a policy is present with an "enforce"
       mode, when attempting to deliver to each candidate MX, ensure
       STARTTLS support and host identity validity as described in
       Section 4, "Policy Validation". If a candidate fails validation,
       continue to the next candidate (if there is one).

   3. A message delivery attempt MUST NOT be permanently failed until
       the sender has first checked for the presence of a new policy (as
       indicated by the "id" field in the "_mta-sts" TXT record). If a
       new policy is not found, existing rules for the case of temporary
       message delivery failures apply (as discussed in [RFC5321],
       Section 4.5.4.1).

6. Reporting Failures

   MTA-STS is intended to be used along with TLSRPT [RFC8460] in order
   to ensure that implementing domains can detect cases of both benign
   and malicious failures and to ensure that failures that indicate an
   active attack are discoverable. As such, senders that also implement
   TLSRPT SHOULD treat the following events as reportable failures:

   o HTTPS policy fetch failures when a valid TXT record is present.

   o Policy fetch failures of any kind when a valid policy exists in
      the policy cache, except if that policy's mode is "none".



Margolis, et al. Standards Track [Page 13]
 
RFC 8461 MTA-STS September 2018


   o Delivery attempts in which a contacted MX does not support
      STARTTLS or does not present a certificate that validates
      according to the applied policy, except if that policy's mode is
      "none".

7. Interoperability Considerations

7.1. SNI Support

   To ensure that the server sends the right certificate chain, the SMTP
   client MUST have support for the TLS Server Name Indication (SNI)
   extension [RFC6066]. When connecting to an HTTP server to retrieve
   the MTA-STS Policy, the SNI extension MUST contain the name of the
   Policy Host (e.g., "mta-sts.example.com"). When connecting to an
   SMTP server, the SNI extension MUST contain the MX hostname.

   HTTP servers used to deliver MTA-STS policies MAY rely on SNI to
   determine which certificate chain to present to the client. HTTP
   servers MUST respond with a certificate chain that matches the policy
   hostname or abort the TLS handshake if unable to do so. Clients that
   do not send SNI information may not see the expected certificate
   chain.

   SMTP servers MAY rely on SNI to determine which certificate chain to
   present to the client. However, servers that have one identity and a
   single matching certificate do not require SNI support. Servers MUST
   NOT enforce the use of SNI by clients, as the client may be using
   unauthenticated opportunistic TLS and may not expect any particular
   certificate from the server. If the client sends no SNI extension or
   sends an SNI extension for an unsupported server name, the server
   MUST simply send a fallback certificate chain of its choice. The
   reason for not enforcing strict matching of the requested SNI
   hostname is that MTA-STS TLS clients may be typically willing to
   accept multiple server names but can only send one name in the SNI
   extension. The server's fallback certificate may match a different
   name that is acceptable to the client, e.g., the original next-hop
   domain.

7.2. Minimum TLS Version Support

   MTAs supporting MTA-STS MUST have support for TLS 1.2 [RFC5246] or
   TLS 1.3 [RFC8446] or higher. The general TLS usage guidance in
   [RFC7525] SHOULD be followed.








Margolis, et al. Standards Track [Page 14]
 
RFC 8461 MTA-STS September 2018


8. Operational Considerations

8.1. Policy Updates

   Updating the policy requires that the owner make changes in two
   places: the "_mta-sts" TXT record in the Policy Domain's DNS zone and
   at the corresponding HTTPS endpoint. As a result, recipients should
   expect that a policy will continue to be used by senders until both
   the HTTPS and TXT endpoints are updated and the TXT record's TTL has
   passed.

   In other words, a sender who is unable to successfully deliver a
   message while applying a cache of the recipient's now-outdated policy
   may be unable to discover that a new policy exists until the DNS TTL
   has passed. Recipients SHOULD therefore ensure that old policies
   continue to work for message delivery during this period of time, or
   risk message delays.

   Recipients SHOULD also update the HTTPS policy body before updating
   the TXT record; this ordering avoids the risk that senders, seeing a
   new TXT record, mistakenly cache the old policy from HTTPS.

8.2. Policy Delegation

   Domain owners commonly delegate SMTP hosting to a different
   organization, such as an ISP or a web host. In such a case, they may
   wish to also delegate the MTA-STS Policy to the same organization,
   which can be accomplished with two changes.

   First, the Policy Domain must point the "_mta-sts" record, via CNAME,
   to the "_mta-sts" record maintained by the provider. This allows the
   provider to control update signaling.

   Second, the Policy Domain must point the "well-known" policy location
   to the provider. This can be done either by setting the "mta-sts"
   record to an IP address or CNAME specified by the provider and by
   giving the provider a TLS certificate that is valid for that host or
   by setting up a "reverse proxy" (also known as a "gateway") server
   for the Policy Domain's Policy Host, configured to serve proxied
   responses from the Policy Host of the provider.

   For example, given a user domain "user.example" hosted by a mail
   provider "provider.example", the following configuration would allow
   policy delegation:

   DNS:

        _mta-sts.user.example. IN CNAME _mta-sts.provider.example.



Margolis, et al. Standards Track [Page 15]
 
RFC 8461 MTA-STS September 2018


   Policy:

         > GET /.well-known/mta-sts.txt Host: mta-sts.user.example
         < HTTP/1.1 200 OK # Response proxies content from
                            # https://mta-sts.provider.example

   Note that in all such cases, the policy endpoint
   ("https://mta-sts.user.example/.well-known/mta-sts.txt" in this
   example) must still present a certificate valid for the Policy Host
   ("mta-sts.user.example"), and not for that host at the provider's
   domain ("mta-sts.provider.example").

   Note that while Sending MTAs MUST NOT use HTTP caching when fetching
   policies via HTTPS, such caching may nonetheless be useful to a
   reverse proxy configured as described in this section. An HTTPS
   policy endpoint expecting to be proxied for multiple hosted domains
   -- as with a large mail hosting provider or similar -- may wish to
   indicate an HTTP Cache-Control "max-age" response directive (as
   specified in [RFC7234]) of 60 seconds as a reasonable value to save
   reverse proxies an unnecessarily high-rate of proxied policy
   fetching.

8.3. Removing MTA-STS

   In order to facilitate clean opt-out of MTA-STS by implementing
   Policy Domains, and to distinguish clearly between failures that
   indicate attacks and those that indicate such opt-outs, MTA-STS
   implements the "none" mode, which allows validated policies to
   indicate authoritatively that the Policy Domain wishes to no longer
   implement MTA-STS and may, in the future, remove the MTA-STS TXT and
   policy endpoints entirely.

   A suggested workflow to implement such an opt out is as follows:

   1. Publish a new policy with "mode" equal to "none" and a small
       "max_age" (e.g., one day).

   2. Publish a new TXT record to trigger fetching of the new policy.

   3. When all previously served policies have expired -- normally this
       is the time the previously published policy was last served plus
       that policy's "max_age", but note that policies older than the
       previously published policy may have been served with a greater
       "max_age" than the previously published policy, allowing
       overlapping policy caches -- safely remove the TXT record and
       HTTPS endpoint.





Margolis, et al. Standards Track [Page 16]
 
RFC 8461 MTA-STS September 2018


8.4. Preserving MX Candidate Traversal

   Implementers of send-time MTA-STS validation in mail transfer agents
   should take note of the risks of modifying the logic of traversing MX
   candidate lists. Because an MTA-STS Policy can be used to prefilter
   invalid MX candidates from the MX candidate list, it is tempting to
   implement a "two-pass" model, where MX candidates are first filtered
   for possible validity according to the MTA-STS Policy, and then the
   remaining candidates are attempted in order as without an MTA-STS
   Policy. This may lead to incorrect implementations, such as message
   loops; instead, it is recommended that implementers traverse the MX
   candidate list as usual, and treat invalid candidates as though they
   were unreachable (i.e., as though there were some transient error
   when trying to deliver to that candidate).

   One consequence of validating MX hosts in order of ordinary candidate
   traversal is that in the event a higher-priority MX is MTA-STS valid
   and a lower-priority MX is not, senders may never encounter the
   lower-priority MX, leading to a risk that policy misconfigurations
   that apply only to "backup" MXes may only be discovered in the case
   of primary MX failure.

9. IANA Considerations

9.1. Well-Known URIs Registry

   A new "well-known" URI as described in Section 3 has been registered
   in the "Well-Known URIs" registry as described below:

   URI Suffix: mta-sts.txt

   Change Controller: IETF

9.2. MTA-STS TXT Record Fields

   IANA has created a new registry titled "MTA-STS TXT Record Fields".
   The initial entries in the registry are:

       +------------+--------------------+-------------------------+
       | Field Name | Description | Reference |
       +------------+--------------------+-------------------------+
       | v | Record version | Section 3.1 of RFC 8461 |
       | id | Policy instance ID | Section 3.1 of RFC 8461 |
       +------------+--------------------+-------------------------+

   New fields are added to this registry using IANA's "Expert Review"
   policy [RFC8126].




Margolis, et al. Standards Track [Page 17]
 
RFC 8461 MTA-STS September 2018


9.3. MTA-STS Policy Fields

   IANA has created a new registry titled "MTA-STS Policy Fields". The
   initial entries in the registry are:

      +------------+----------------------+-------------------------+
      | Field Name | Description | Reference |
      +------------+----------------------+-------------------------+
      | version | Policy version | Section 3.2 of RFC 8461 |
      | mode | Enforcement behavior | Section 3.2 of RFC 8461 |
      | max_age | Policy lifetime | Section 3.2 of RFC 8461 |
      | mx | MX identities | Section 3.2 of RFC 8461 |
      +------------+----------------------+-------------------------+

   New fields are added to this registry using IANA's "Expert Review"
   policy.

10. Security Considerations

   SMTP MTA-STS attempts to protect against an active attacker trying to
   intercept or tamper with mail between hosts that support STARTTLS.
   There are two classes of attacks considered:

   o Foiling TLS negotiation (for example, by deleting the "250
      STARTTLS" response from a server or altering TLS session
      negotiation). This would result in the SMTP session occurring
      over plaintext, despite both parties supporting TLS.

   o Impersonating the destination mail server, whereby the sender
      might deliver the message to an impostor, who could then monitor
      and/or modify messages despite opportunistic TLS. This
      impersonation could be accomplished by spoofing the DNS MX record
      for the recipient domain or by redirecting client connections
      intended for the legitimate recipient server (for example, by
      altering BGP routing tables).

   MTA-STS can thwart such attacks only if the sender is able to
   previously obtain and cache a policy for the recipient domain, and
   only if the attacker is unable to obtain a valid certificate that
   complies with that policy. Below, we consider specific attacks on
   this model.

10.1. Obtaining a Signed Certificate

   SMTP MTA-STS relies on certificate validation via PKIX-based TLS
   identity checking [RFC6125]. Attackers who are able to obtain a
   valid certificate for the targeted recipient mail service (e.g., by
   compromising a CA) are thus able to circumvent STS authentication.



Margolis, et al. Standards Track [Page 18]
 
RFC 8461 MTA-STS September 2018


10.2. Preventing Policy Discovery

   Since MTA-STS uses DNS TXT records for policy discovery, an attacker
   who is able to block DNS responses can suppress the discovery of an
   MTA-STS Policy, making the Policy Domain appear not to have an MTA-
   STS Policy. The sender policy cache is designed to resist this
   attack by decreasing the frequency of policy discovery and thus
   reducing the window of vulnerability; it is nonetheless a risk that
   attackers who can predict or induce policy discovery -- for example,
   by inducing a sending domain to send mail to a never-before-contacted
   recipient while carrying out a man-in-the-middle attack -- may be
   able to foil policy discovery and effectively downgrade the security
   of the message delivery.

   Since this attack depends upon intercepting initial policy discovery,
   implementers SHOULD prefer policy "max_age" values to be as long as
   is practical.

   Because this attack is also possible upon refresh of a cached policy,
   implementers SHOULD NOT wait until a cached policy has expired before
   checking for an update; if senders attempt to refresh the cache
   regularly (for example, by fetching the current live policy in a
   background task that runs daily or weekly, regardless of the state of
   the "_mta-sts" TXT record, and updating their cache's "max age"
   accordingly), an attacker would have to foil policy discovery
   consistently over the lifetime of a cached policy to prevent a
   successful refresh.

   Additionally, MTAs SHOULD alert administrators to repeated policy
   refresh failures long before cached policies expire (through warning
   logs or similar applicable mechanisms), allowing administrators to
   detect such a persistent attack on policy refresh. (However, they
   should not implement such alerts if the cached policy has a "none"
   mode, to allow clean MTA-STS removal, as described in Section 8.3.)

   Resistance to downgrade attacks of this nature -- due to the ability
   to authoritatively determine "lack of a record" even for non-
   participating recipients -- is a feature of DANE, due to its use of
   DNSSEC for policy discovery.

10.3. Denial of Service

   We additionally consider the Denial-of-Service risk posed by an
   attacker who can modify the DNS records for a recipient domain.
   Absent MTA-STS, such an attacker can cause a Sending MTA to cache
   invalid MX records, but only for however long the sending resolver
   caches those records. With MTA-STS, the attacker can additionally
   advertise a new, long "max_age" MTA-STS Policy with "mx" constraints



Margolis, et al. Standards Track [Page 19]
 
RFC 8461 MTA-STS September 2018


   that validate the malicious MX record, causing senders to cache the
   policy and refuse to deliver messages once the victim has resecured
   the MX records.

   This attack is mitigated in part by the ability of a victim domain to
   (at any time) publish a new policy updating the cached, malicious
   policy, though this does require the victim domain to both obtain a
   valid CA-signed certificate and to understand and properly configure
   MTA-STS.

   Similarly, we consider the possibility of domains that deliberately
   allow untrusted users to serve untrusted content on user-specified
   subdomains. In some cases (e.g., the service "tumblr.com"), this
   takes the form of providing HTTPS hosting of user-registered
   subdomains; in other cases (e.g. dynamic DNS providers), this takes
   the form of allowing untrusted users to register custom DNS records
   at the provider's domain.

   In these cases, there is a risk that untrusted users would be able to
   serve custom content at the "mta-sts" host, including serving an
   illegitimate MTA-STS Policy. We believe this attack is rendered more
   difficult by the need for the attacker to also serve the "_mta-sts"
   TXT record on the same domain -- something not, to our knowledge,
   widely provided to untrusted users. This attack is additionally
   mitigated by the aforementioned ability for a victim domain to update
   an invalid policy at any future date.

10.4. Weak Policy Constraints

   Even if an attacker cannot modify a served policy, the potential
   exists for configurations that allow attackers on the same domain to
   receive mail for that domain. For example, an easy configuration
   option when authoring an MTA-STS Policy for "example.com" is to set
   the "mx" equal to "*.example.com"; in this case, recipient domains
   must consider the risk that any user possessing a valid hostname and
   CA-signed certificate (for example, "dhcp-123.example.com") will,
   from the perspective of MTA-STS Policy validation, be a valid MX host
   for that domain.

10.5. Compromise of the Web PKI System

   A number of risks apply to the PKI system that is used for
   certificate authentication, both of the "mta-sts" HTTPS host's
   certificate and the SMTP servers' certificates. These risks are
   broadly applicable within the Web PKI ecosystem and are not specific
   to MTA-STS; nonetheless, they deserve some consideration in this
   context.




Margolis, et al. Standards Track [Page 20]
 
RFC 8461 MTA-STS September 2018


   Broadly speaking, attackers may compromise the system by obtaining
   certificates under fraudulent circumstances (i.e., by impersonating
   the legitimate owner of the victim domain), by compromising a CA or
   Delegate Authority's private keys, by obtaining a legitimate
   certificate issued to the victim domain, and similar.

   One approach commonly employed by web browsers to help mitigate
   against some of these attacks is to allow for revocation of
   compromised or fraudulent certificates via OCSP [RFC6960] or CRLs
   [RFC6818]. Such mechanisms themselves represent trade-offs and are
   not universally implemented; we nonetheless recommend implementers of
   MTA-STS to implement revocation mechanisms that are most applicable
   to their implementations.

11. References

11.1. Normative References

   [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>.

   [RFC2818] Rescorla, E., "HTTP Over TLS", RFC 2818,
              DOI 10.17487/RFC2818, May 2000,
              <https://www.rfc-editor.org/info/rfc2818>.

   [RFC3207] Hoffman, P., "SMTP Service Extension for Secure SMTP over
              Transport Layer Security", RFC 3207, DOI 10.17487/RFC3207,
              February 2002, <https://www.rfc-editor.org/info/rfc3207>.

   [RFC3492] Costello, A., "Punycode: A Bootstring encoding of Unicode
              for Internationalized Domain Names in Applications
              (IDNA)", RFC 3492, DOI 10.17487/RFC3492, March 2003,
              <https://www.rfc-editor.org/info/rfc3492>.

   [RFC3629] Yergeau, F., "UTF-8, a transformation format of ISO
              10646", STD 63, RFC 3629, DOI 10.17487/RFC3629, November
              2003, <https://www.rfc-editor.org/info/rfc3629>.

   [RFC5234] Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax
              Specifications: ABNF", STD 68, RFC 5234,
              DOI 10.17487/RFC5234, January 2008,
              <https://www.rfc-editor.org/info/rfc5234>.







Margolis, et al. Standards Track [Page 21]
 
RFC 8461 MTA-STS September 2018


   [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security
              (TLS) Protocol Version 1.2", RFC 5246,
              DOI 10.17487/RFC5246, August 2008,
              <https://www.rfc-editor.org/info/rfc5246>.

   [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>.

   [RFC5785] Nottingham, M. and E. Hammer-Lahav, "Defining Well-Known
              Uniform Resource Identifiers (URIs)", RFC 5785,
              DOI 10.17487/RFC5785, April 2010,
              <https://www.rfc-editor.org/info/rfc5785>.

   [RFC6066] Eastlake 3rd, D., "Transport Layer Security (TLS)
              Extensions: Extension Definitions", RFC 6066,
              DOI 10.17487/RFC6066, January 2011,
              <https://www.rfc-editor.org/info/rfc6066>.

   [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>.

   [RFC7405] Kyzivat, P., "Case-Sensitive String Support in ABNF",
              RFC 7405, DOI 10.17487/RFC7405, December 2014,
              <https://www.rfc-editor.org/info/rfc7405>.

   [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>.

   [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>.






Margolis, et al. Standards Track [Page 22]
 
RFC 8461 MTA-STS September 2018


   [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>.

   [RFC8460] Margolis, D., Brotman, A., Ramakrishnan, B., Jones, J.,
              and M. Risher, "SMTP TLS Reporting", RFC 8460,
              DOI 10.17487/RFC8460, September 2018,
              <https://www.rfc-editor.org/info/rfc8460>.

11.2. Informative References

   [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S.
              Rose, "DNS Security Introduction and Requirements",
              RFC 4033, DOI 10.17487/RFC4033, March 2005,
              <https://www.rfc-editor.org/info/rfc4033>.

   [RFC5322] Resnick, P., Ed., "Internet Message Format", RFC 5322,
              DOI 10.17487/RFC5322, October 2008,
              <https://www.rfc-editor.org/info/rfc5322>.

   [RFC5891] Klensin, J., "Internationalized Domain Names in
              Applications (IDNA): Protocol", RFC 5891,
              DOI 10.17487/RFC5891, August 2010,
              <https://www.rfc-editor.org/info/rfc5891>.

   [RFC6818] Yee, P., "Updates to the Internet X.509 Public Key
              Infrastructure Certificate and Certificate Revocation List
              (CRL) Profile", RFC 6818, DOI 10.17487/RFC6818, January
              2013, <https://www.rfc-editor.org/info/rfc6818>.

   [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>.

   [RFC7234] Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke,
              Ed., "Hypertext Transfer Protocol (HTTP/1.1): Caching",
              RFC 7234, DOI 10.17487/RFC7234, June 2014,
              <https://www.rfc-editor.org/info/rfc7234>.

   [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>.





Margolis, et al. Standards Track [Page 23]
 
RFC 8461 MTA-STS September 2018


   [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>.















































Margolis, et al. Standards Track [Page 24]
 
RFC 8461 MTA-STS September 2018


Appendix A. MTA-STS Example Record and Policy

   The owner of "example.com" wishes to begin using MTA-STS with a
   policy that will solicit reports from senders without affecting how
   the messages are processed, in order to verify the identity of MXes
   that handle mail for "example.com", confirm that TLS is correctly
   used, and ensure that certificates presented by the recipient MX
   validate.

   MTA-STS Policy indicator TXT RR:

       _mta-sts.example.com. IN TXT "v=STSv1; id=20160831085700Z;"

   MTA-STS Policy file served as the response body at
   "https://mta-sts.example.com/.well-known/mta-sts.txt":

                         version: STSv1
                         mode: testing
                         mx: mx1.example.com
                         mx: mx2.example.com
                         mx: mx.backup-example.com
                         max_age: 1296000

Appendix B. Message Delivery Pseudocode

   Below is pseudocode demonstrating the logic of a compliant Sending
   MTA.

   While this pseudocode implementation suggests synchronous policy
   retrieval in the delivery path, that may be undesirable in a working
   implementation, and we expect some implementers to instead prefer a
   background fetch that does not block delivery when no cached policy
   is present.


   func isEnforce(policy) {
     // Return true if the policy mode is "enforce".
   }

   func isNonExpired(policy) {
     // Return true if the policy is not expired.
   }

   func tryStartTls(connection) {
     // Attempt to open an SMTP STARTTLS connection with the MX.
   }

   func certMatches(connection, host) {



Margolis, et al. Standards Track [Page 25]
 
RFC 8461 MTA-STS September 2018


     // Assume a handy function to return if the server
     // certificate presented in "connection" is valid for "host".
   }

   func policyMatches(candidate, policy) {
     for mx in policy.mx {
       // Literal match.
       if mx == candidate {
         return true
       }
       // Wildcard matches only the leftmost label.
       // Wildcards must always be followed by a '.'.
       if mx[0] == '*' {
         parts = SplitN(candidate, '.', 2) // Split on the first '.'.
         if len(parts) > 1 && parts[1] == mx[2:] {
           return true
         }
       }
     }
     return false
   }

   func tryDeliverMail(connection, message) {
     // Attempt to deliver "message" via "connection".
   }

   func tryGetNewPolicy(domain) {
     // Check for an MTA-STS TXT record for "domain" in DNS, and return
     // the indicated policy.
   }

   func cachePolicy(domain, policy) {
     // Store "policy" as the cached policy for "domain".
   }

   func tryGetCachedPolicy(domain) {
     // Return a cached policy for "domain".
   }

   func reportError(error) {
     // Report an error via TLSRPT.
   }

   func tryMxAccordingTo(message, mx, policy) {
     connection := connect(mx)
     if !connection {
       return false // Can't connect to the MX, so it's not an MTA-STS
                     // error.



Margolis, et al. Standards Track [Page 26]
 
RFC 8461 MTA-STS September 2018


     }
     secure := true
     if !policyMatches(mx, policy) {
       secure = false
       reportError(E_HOST_MISMATCH)
     } else if !tryStartTls(connection) {
       secure = false
       reportError(E_NO_VALID_TLS)
     } else if !certMatches(connection, policy) {
       secure = false
       reportError(E_CERT_MISMATCH)
     }
     if secure || !isEnforce(policy) {
       return tryDeliverMail(connection, message)
     }
     return false
   }

   func tryWithPolicy(message, domain, policy) {
     mxes := getMxForDomain(domain)
     for mx in mxes {
       if tryMxAccordingTo(message, mx, policy) {
         return true
       }
     }
     return false
   }

   func handleMessage(message) {
     domain := ... // domain part after '@' from recipient
     policy := tryGetNewPolicy(domain)
     if policy {
       cachePolicy(domain, policy)
     } else {
       policy = tryGetCachedPolicy(domain)
     }
     if policy {
       return tryWithPolicy(message, domain, policy)
     }
     // Try to deliver the message normally (i.e., without MTA-STS).
   }










Margolis, et al. Standards Track [Page 27]
 
RFC 8461 MTA-STS September 2018


Contributors

   Wei Chuang
   Google, Inc.
   weihaw@google.com

   Viktor Dukhovni
   ietf-dane@dukhovni.de

   Markus Laber
   1&1 Mail & Media Development & Technology GmbH
   markus.laber@1und1.de

   Nicolas Lidzborski
   Google, Inc.
   nlidz@google.com

   Brandon Long
   Google, Inc.
   blong@google.com

   Franck Martin
   LinkedIn, Inc.
   fmartin@linkedin.com

   Klaus Umbach
   1&1 Mail & Media Development & Technology GmbH
   klaus.umbach@1und1.de























Margolis, et al. Standards Track [Page 28]
 
RFC 8461 MTA-STS September 2018


Authors' Addresses

   Daniel Margolis
   Google, Inc.

   Email: dmargolis@google.com


   Mark Risher
   Google, Inc.

   Email: risher@google.com


   Binu Ramakrishnan
   Oath, Inc.

   Email: prbinu@yahoo.com


   Alexander Brotman
   Comcast, Inc.

   Email: alex_brotman@comcast.com


   Janet Jones
   Microsoft, Inc.

   Email: janet.jones@microsoft.com





















Margolis, et al. Standards Track [Page 29]