]>

v. L. Anelli 13 20122 Milano MI IT vesely@tana.it

General IETF DNSWL EMAIL Authentication-Results This document describes an additional Email Authentication Method compliant with RFC 8601. The method consists in looking up the sender's IP address in a DNS whitelist. This document is provided for information in case the method is seen in the field, as well as to suggest a useful practice and register the relevant keywords. This document does not consider black lists.

One of the many checks that mail servers carry out is to query domain name system whitelists (DNSWL). That method is fully discussed in . The DNS lookup is based on the connecting client's IP address, IPv4 or IPv6, and returns zero or more A records. The latter are IPv4 IP addresses in the range 127.0.0.0/8. Depending on the query, TXT records with varying content can also be retrieved. Query examples are given in . Since the IP address is known as soon as the connection is accepted, this check can occur very early in an SMTP transaction. Its result can be used to counterweight policies that typically occur at early stages too, such as the Sender Policy Framework (SPF, the last paragraph of Appendix D.3 of is also illustrated in ). In addition, the result of a DNSWL lookup can also be used at later stages; for example, a delivery agent can use it to learn the trustworthiness of a mail relay in order to estimate the spamminess of an email message. The latter possibility needs a place to collect query results for downstream use, which is precisely what the Authentication-Results header field aims at providing. Results often contain additional data, encoded according to DNSWL-specific criteria. The method described in this document considers only whitelists --one of the major branches described by . There are also black/block lists, DNSBL, and combined lists. Since they all have the same structure, the abbreviation DNSxL is used to mean any. The core procedures of a mail transfer agent (MTA) tend to be quite general, leaving particular cases to be handled by add-on modules. In the case of combined lists, the boundary MTA (see ) which carries out the check and possibly stores the result has to be able to discern at least the color of each entry, as that is required to make accept/reject decisions. This document provides for storing the result when the DNSxL record to be reported is a whitelisting one. Data conveyed in A and TXT records can be stored as method's properties. The meaning of such data varies widely at the mercy of the list operator, hence the queried zone has to be stored as well. Mail site operators who configure their MTAs to query specific DNWSLs marry the policies of those lists, as, in effect, they become tantamount to local policies, albeit outsourced. Downstream agents who know DNSWL-specific encoding and understand the meaning of that datum can use it to make delivery or display decisions. For example, a mail filter which detected heuristic evidence of a scam can counterweight such information with the trustworthiness score encoded in the A response, so as to protect agains false positives. MUAs can display those results, or use them to decide how to report abusive messages, if configured to do so. This document describes a usage of TXT fields consistent with other authentication methods, namely to serve the domain name in the TXT record. That way, a downstream filter could also consider whether the sending agent is aligned with the author domain, with semantics similar to . At the time of this writing, this method is implemented by . An outline of the implementation is given in .

The result of the method states how the query did, up to the interpretation of the returned data. The method has four possible results: The query successfully returned applicable records. This result is usually accompanied by one or both of the policy properties described below. Since the list is configured as a DNSWL, agents unable to interpret list-specifc properties can still derive a positive value from the fact that the sender is whitelisted. The query worked but yielded no A record, or returned NXDOMAIN, so the sender is not whitelisted. The DNS evaluation could not be completed due to some error that is likely transient in nature, such as a temporary DNS error, e.g., a DNS RCODE of 2, commonly known as SERVFAIL, or other error condition. A later attempt may produce a final result. The DNS evaluation cannot work because test entries don't work, that is, DNSWL is broken, or because queries are over quota, e.g., a DNS RCODE of 5, commonly known as REFUSED, or a DNSWL-specific property (policy.ip, defined below) with the same meaning. A later attempt is unlikely to produce a final result. Human intervention is required. Note that there is no "fail" result. The following ptype.property items define how the data provided by the whitelist lookup can be saved. DNSWL query root domain, which defines the meaning of the policy.ip property below. Note that an MTA can use a local mirror with a different name. The name stored here has to be the best available reference for all foreseeable downstream consumers. Setting dns.zone to the global zone makes the result intelligible even if the message is handed outside of the internal network. The bit mask value received in type A response, in dotted quad notation. Multiple entries can be arranged in a quoted, comma separated list (quotes are necessary because commas are not allowed in a token.) The TXT record, if any. Multiple records are concatenated in the usual way (explained, for example, in Section 3.3 of ). See for the resulting content and query options. This is a generic property stating whether the relevant data was validated using DNSSEC (). For the present method, the relevant data consists of the reported policy properties above, or, if the method result is "none", their non-existence. This property has three possible values: DNSSEC validation confirms the integrity of data. considers how that is related to the DNS response. The data are not signed. See . Not applicable. No DNSSEC validation can be performed, possibly because the lookup is run through a different means than a security-aware DNS resolver. This does not necessarily imply less security. In particular, "na" is used if the data were downloaded in bulk and then loaded on a local nameserver --which is the case of an MTA querying a local zone different from the reported dns.zone. DNS errors, including validation errors, can also report "na". This is also the value assumed by default.

According to , TXT records describe the reason why IP addresses are listed in a DNSWL. An example of a DNSWL whose TXT records contain the domain name of the organization assignee of the sending IP is given in . The domain name would correspond to the DNS domain name used by or within the administrative domain (ADMD) operating the relevant MTA, sometimes called the "organizational domain". In that case, the authentication provided by this method is equivalent to a DKIM signature () or an SPF check host (), if the DNSWL is trusted. According to a DNSWL's policy, attributing responsibility of an IP address to an organization may require something more than a mere PTR record consistency. If no domain names can be responsibly associated to a given IP address, for example because the IP address was added without direct involvement of the organization concerned, DNSWLs can use a subdomain of .INVALID () where the leftmost label hints at why an address is whitelisted. For example, if the address 192.0.2.38 was added by the list managers solely based on their knowledge, the corresponding TXT record might be AUTOPROMOTED.INVALID, so as to avoid to explicitly identify an entity who didn't opt-in. Following the example of Multicast DNS (see the second paragraph of Section 16 of ) names containing non-ASCII characters can be encoded in UTF-8 using the normalization form canonical composition (NFC) as described in Unicode Format for Network Interchange (). Inclusion of unaltered UTF-8 TXT values in the header entails an environment compatible with EAI . DNS queries with a QTYPE of ANY may lead to inconsistent replies, depending on the cache status. In addition, ANY is not "all", and the provisions for queries that have QTYPE=ANY () don't cover DNSxLs. A mail server can issue two simultaneous queries, A and TXT. Otherwise, a downstream filter can issue a TXT query on its own, if it knows that an A query was successful and that the DNSWL serves useful TXT records. It is unlikely that TXT records exist if a query for QTYPE A brought a result of none.

IANA maintains the "Email Authentication Parameters" registry with several subregistries. IANA is requested to make assignments as set out in the following sections.

IANA is requested to create four new entries in the "Email Authentication Methods" registry as follows.

IANA is requested to create a new entry in the "Email Authentication Property Types" registry as follows. ptype Definition Description dns [This.I-D] The property being reported belongs to the Domain Name System

IANA is requested to create four new entries in the "Email Authentication Result Names" registry as follows. Auth Method Code Specification Status dnswl pass [This.I-D] active dnswl none [This.I-D] active dnswl temperror [This.I-D] active dnswl permerror [This.I-D] active

Some DNSWLs which provide for free access below a given quota are known to return special codes to signal over quota, for example 127.0.0.255. If the MTA cannot interpret that value, that case results in a false positive. It can accept messages that it would otherwise reject. A DNSWL-specific module would realize the fact and call for human intervention. Returning an RCODE 5 (REFUSED) conveys the concept that the query is "unauthorized", and human intervention required.

The dns.sec property is meant to be as secure as DNSSEC results. It makes sense to use it in an environment where the DNSSEC validation can succeed. Section 7 of examines various ways of setting up a stub resolver which either validates DNSSEC locally or trusts the validation provided through a secure channel. For a different class, it is possible to set up a dedicated, caching, dnssec-enabled resolver reachable by the mail server through interprocess communication on 127.0.0.1. In such cases, the property dns.sec=yes corresponds to the Authenticated Data (AD) bit in the DNS response header. When the response contains no DNSSEC data, a security-aware resolver seeks a signed proof of the non-existence of a DS record, at some delegation point. If no error is returned, the zone is unsigned and dns.sec=no can be set. According to the Security Considerations Section of : The absence of DNSSEC data in response to a query with the DO bit set must not be taken to mean no security information is available for that zone as the response may be forged or a non-forged response of an altered (DO bit cleared) query. If the application verifies the DNSSEC signatures on its own, it effectively behaves like a validating stub resolver, and hence can set dns.sec correspondingly. When the data are downloaded in bulk and made available on a trusted channel without using DNSSEC, set dns.sec=na or not at all. DNSWL who publish bulk versions of their data can also sign that data, for example using OpenPGP (). It is the responsibility of system administrators to authenticate the data by downloading and validating the signature. The result of such validation is not reported using dns.sec.

For DNSSEC, the considerations of Section 12 of apply. All of the considerations described in Section 7 of apply. That includes securing against tampering all the channels after the production of this Authentication-Results header field. In addition, the usual caveats apply about importing text from external online sources. Although queried DNSWLs are well known, trusted entities, it is suggested that TXT records be reported only if, upon inspection, their content is deemed actually actionable, and their format compatible with the computing environment.

&RFC2606; &RFC5782; &RFC8601; &RFC1034; &RFC3225; &RFC3629; &RFC4033; &RFC4880; &RFC5198; &RFC5598; &RFC6376; &RFC6530; &RFC6762; &RFC7208; &RFC7489; &RFC8460; &RFC8482;

Authentication-Results: mta.example.org; dkim=pass (whitelisted) header.i=@example.com Authentication-Results: mta.example.org; dnswl=pass dns.zone=list.dnswl.example dns.sec=na policy.ip=127.0.10.1 policy.txt="fwd.example https://dnswl.example/?d=fwd.example" Received-SPF: fail (Address does not pass Sender Policy Framework) client-ip=2001:db8::2:1; envelope-from="sender@example.com"; helo=mail.fwd.example; receiver=mta.example.org; Received: from mail.fwd.example (mail.fwd.example [2001:db8::2:1]) (TLS: TLSv1/SSLv3,128bits,ECDHE-RSA-AES128-GCM-SHA256) by mta.example.org with ESMTPS; Thu, 03 Oct 2019 19:23:11 +0200 id 00000000005DC044.000000005702D87C.000007FC ]]> Trace fields at the top of the header

The message went through a third party, fwd.example, which forwarded it to the final MTA. Such mail path was not arranged beforehand with the involved MTAs, it emerged spontaneously. This message would not have made it to the target without whitelisting, because: the author domain published a strict SPF policy (-all), the forwarder did not alter the bounce address, and the target usually honors reject-on-fail, according to Section 8.4 of . However, the target also implemented the last paragraph of Appendix D.3 of . Its behavior hinges on the following DNS entries:

If mail.fwd.example had connected from address 192.0.2.1, then the query name would have been 1.2.0.192.list.dnswl.example. See full description in At connection time, because the remote IP address is whitelisted, the target MTA did not reject the message before DATA. Instead, it recorded the SPF fail result, and indicated the local policy mechanism which was applied in order to override that result. Subsequent filtering verified DKIM . At later stages, mail filters can reject or quarantine the message based on its content. A deeper knowledge of the policy values obtained from dnswl.example allows to interpret the values of policy.ip and weight them against other factors so as to make better decisions.

Implementation details mentioned in this section have been stable for several years. Yet, this description is necessarily superficial, version dependent, and subject to change. can be configured to lookup DNSBL and DNSWL, with similar command line switches:

zone is the zone to be queried. displayzone is only used for -allow, it is the value to be set in the dns.zone property. var stands for the environment variable whose existence triggers a special action. The default variable names result in a conventional behavior implemented by Courier-MTA. By setting different environment variables, users can customize the behavior. Conventional behavior differs widely between -block and -allow. The former rejects the message, the latter produces Authentication-Results header fields. The n.n.n.n IP address requires a precise A record response. If not given, any response results in setting the corresponding variable. If given, variables are set only if the response matches exactly. Such syntax provides for a very limited interpretation of the information encoded in A records. However, it is considered to be too complicated already. Even specifying a range, an enumeration of values, or a regular expression would require something beyond what a normal user would be willing to manage. Finally, the trailing message, which overrides the 5xx SMTP reply for -block, is not used for -allow, except that its mere presence requires to query TXT records to be registered in policy.txt. SPF is part of Courier-MTA's core. It is configured separately, and provides for an "allowok" keyword to indicate to override rejection in case of SPF failure and -allow whitelisting. A customary whitelist is . It serves A records encoded as follows: 1st octect: 127. 2nd octect: 0. 3rd octect: Category of business, 15 values. 4th octect: Trusworthiness/score, 4 values. They also serve TXT records containing the domain name followed by an URL pointing to further info about the relevant organization, such as what other IP addresses of theirs are being whitelisted. They don't use UTF-8. dnswl.org provides for free registration and free access below 100,000 queries per day. They use the special return code 127.0.0.255 exemplified above to signal over quota. Although Courier-MTA itself does not recognize it, it has a mail filter (zdkimfilter, named after its main usage) where recognition of that code, as well as that of trusworthiness in the 4th octect are hard-coded.

The description of the new ptype proposed in just says "The property being reported belongs to the Domain Name System". That definition can broadly include any tag found in a domain's TXT record. For example, auth methods designers can agree that within a resinfo of a given method, any dns ptype refers to tags in the relevant DNS record, unless otherwise specified. So one could have, say:

While dns.sec is defined above, albeit not for the spf method, the use of tlsrpt in the DKIM record is exemplified in Section 3 of . The tag s= is part of the DKIM TXT record, not to be confused with the selector s=, which is part of a DKIM-Signature. Just like the latter can be reported as header.s because the DKIM header field is in the message header, it may make sense to report the former as dns.s because the DKIM DNS record is in the DNS. NOTE: This is only a hint at what may become a consistent naming convention around the new ptype. In any case, any new property using this ptype requires its own formal definition. This document does NOT define the property dns.s=, let alone the service tlsrpt.