| Network Working Group | P. Wouters |
| Internet-Draft | Red Hat |
| Intended status: Experimental | May 2, 2016 |
| Expires: November 3, 2016 |
Using DANE to Associate OpenPGP public keys with email addresses
draft-ietf-dane-openpgpkey-12
OpenPGP is a message format for email (and file) encryption that lacks a standardized lookup mechanism to securely obtain OpenPGP public keys. DNS-Based Authentication of Named Entities ("DANE") is a method for publishing public keys in DNS. This document specifies a DANE method for publishing and locating OpenPGP public keys in DNS for a specific email address using a new OPENPGPKEY DNS Resource Record. Security is provided via Secure DNS, however the OPENPGPKEY record is not a replacement for verification of authenticity via the "Web Of Trust" or manual verification. The OPENPGPKEY record can be used to encrypt an email that would otherwise have to be sent unencrypted.
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."
This Internet-Draft will expire on November 3, 2016.
Copyright (c) 2016 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 (http://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.
OpenPGP [RFC4880] public keys are used to encrypt or sign email messages and files. To encrypt an email message, or verify a sender's OpenPGP signature, the email client (MUA) or the email server (MTA) needs to locate the recipient's OpenPGP public key.
OpenPGP clients have relied on centralized "well-known" key servers that are accessed using the HTTP Keyserver Protocol [HKP]. Alternatively, users need to manually browse a variety of different front-end websites. These key servers do not require a confirmation of the email address used in the User ID of the uploaded OpenPGP public key. Attackers can - and have - uploaded rogue public keys with other people's email addresses to these key servers.
Once uploaded, public keys cannot be deleted. People who did not pre-sign a key revocation can never remove their OpenPGP public key from these key servers once they have lost access to their private key. This results in receiving encrypted email that cannot be decrypted.
Therefore, these keyservers are not well suited to support MUAs and MTAs to automatically encrypt email - especially in the absence of an interactive user.
This document describes a mechanism to associate a user's OpenPGP public key with their email address, using the OPENPGPKEY DNS RRtype. These records are published in the DNS zone of the user's email address. If the user loses their private key, the OPENPGPKEY DNS record can simply be updated or removed from the zone.
The OPENPGPKEY data is secured using Secure DNS [RFC4035].
The main goal of the OPENPGPKEY resource record is to stop passive attacks against plaintext emails. While it can also thwart some active attacks (such as people uploading rogue keys to keyservers in the hopes that others will encrypt to these rogue keys), this resource record is not a replacement for verifying OpenPGP public keys via the web of trust signatures, or manually via a fingerprint verification.
This specification is one experiment in improving access to public keys for end-to-end email security. There are a range of ways in which this can reasonably be done, for OpenPGP or S/MIME, for example using the DNS, or SMTP, or HTTP. Proposals for each of these have been made with various levels of support in terms of implementation and deployment. For each such experiment, specifications such as this will enable experiments to be carried out that may succeed or that may uncover technical or other impediments to large- or small-scale deployments. The IETF encourages those implementing and deploying such experiments to publicly document their experiences so that future specifications in this space can benefit.
This document defines an RRtype whose use is Experimental. The goal of the experiment is to see whether encrypted email usage will increase if an automated discovery method is available to MTAs and MUAs to help the enduser with email encryption key management.
It is unclear if this RRtype will scale to some of the larger email service deployments. Concerns have been raised about the size of the OPENPGPKEY record and the size of the resulting DNS zone files. This experiment hopefully will give the working group some insight into whether this is a problem or not.
If the experiment is successful, it is expected that the findings of the experiment will result in an updated document for standards track approval.
The OPENPGPKEY RRtype somewhat resembles the generic CERT record defined in [RFC4398]. However, the CERT record uses sub-typing with many different types of keys and certificates. It is suspected that its general application of very different protocols (PKIX versus OpenPGP) has been the cause for lack of implementation and deployment. Furthermore, the CERT record uses sub-typing, which is now considered to be a bad idea for DNS.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [RFC2119].
This document also makes use of standard DNSSEC and DANE terminology. See DNSSEC [RFC4033], [RFC4034], [RFC4035], and DANE [RFC6698] for these terms.
The OPENPGPKEY DNS resource record (RR) is used to associate an end entity OpenPGP Transferable Public Key (see Section 11.1 of [RFC4880] with an email address, thus forming a "OpenPGP public key association". A user that wishes to specify more than one OpenPGP key, for example because they are transitioning to a newer stronger key, can do so by adding multiple OPENPGPKEY records. A single OPENPGPKEY DNS record MUST only contain one OpenPGP key.
The type value allocated for the OPENPGPKEY RR type is 61. The OPENPGPKEY RR is class independent.
The RDATA portion of an OPENPGPKEY Resource Record contains a single value consisting of a [RFC4880] formatted Transferable Public Key.
o The primary key X
o One User ID Y, which SHOULD match 'hugh@example.com'
o self-signature from X, binding X to Y
o The primary key X
o One User ID Y, which SHOULD match 'hugh@example.com'
o self-signature from X, binding X to Y
o encryption-capable subkey Z
o self-signature from X, binding Z to X
o [ other subkeys if relevant ... ]
o The primary key X
o One User ID Y, which SHOULD match 'hugh@example.com'
o self-signature from X, binding X to Y
o third-party certification from V, binding Y to X
o [ other third-party certifications if relevant ... ]
o encryption-capable subkey Z
o self-signature from X, binding Z to X
o [ other subkeys if relevant ... ]
An OpenPGP Transferable Public Key can be arbitrarily large. DNS records are limited in size. When creating OPENPGPKEY DNS records, the OpenPGP Transferable Public Key should be filtered to only contain appropriate and useful data. At a minimum, an OPENPGPKEY Transferable Public Key for the user hugh@example.com should contain:
When preparing a Transferable Public Key for a specific OPENPGPKEY RDATA format with the goal of minimizing certificate size, a user would typically want to:
The RDATA Wire Format consists of a single OpenPGP Transferable Public Key as defined in Section 11.1 of [RFC4880]. Note that this format is without ASCII armor or base64 encoding.
The RDATA Presentation Format, as visible in master files [RFC1035], consists of a single OpenPGP Transferable Public Key as defined in Section 11.1 of [RFC4880] encoded in base64 as defined in Section 4 of [RFC4648].
The DNS does not allow the use of all characters that are supported in the "local-part" of email addresses as defined in [RFC5322] and [RFC6530]. Therefore, email addresses are mapped into DNS using the following method:
c9[..]d6._openpgpkey.example.com. IN OPENPGPKEY <base64 public key>
For example, to request an OPENPGPKEY resource record for a user whose email address is "hugh@example.com", an OPENPGPKEY query would be placed for the following QNAME: "c93f1e400f26708f98cb19d936620da35eec8f72e57f9eec01c1afd6._openpgpkey.example.com". The corresponding RR in the example.com zone might look like (key shortened for formatting):
Mail systems usually handle variant forms of local-parts. The most common variants are upper and lower case, often automatically corrected when a name is recognized as such. Other variants include systems that ignore "noise" characters such as dots, so that local parts johnsmith and John.Smith would be equivalent. Many systems allow "extensions" such as john-ext or mary+ext where john or mary is treated as the effective local-part, and the ext is passed to the recipient for further handling. This can complicate finding the OPENPGPKEY record associated with the dynamically created email address.
[RFC5321] and its predecessors have always made it clear that only the recipient MTA is allowed to interpret the local-part of an address. Therefor, sending MUAs and MTAs supporting OPENPGPKEY MUST NOT perform any kind of mapping rules based on the email address. In order to improve chances of finding OPENPGP RRs for a particular local-part, domains that allow variant forms (such as treating local-parts as case-insensitive) might publish OPENPGP RRs for all variants of local-parts, might publish variants on first use (for example a webmail provider that also controls DNS for a domain can publish variants as used by owner of a particular local-part) or just publish OPENPGP RRs for the most common variants.
Section 3 above defines how the local-part is used to determine the location in which one looks for an OPENPGPKEY record. Given the variety of local-parts seen in email, designing a good experiment for this is difficult as: a) some current implementations are known to lowercase at least US-ASCII local-parts, b) we know from (many) other situations that any strategy based on guessing and making multiple DNS queries is not going to achieve consensus for good reasons, and c) the underlying issues are just hard - see Section 10.1 of [RFC6530] for discussion of just some of the issues that would need to be tackled to fully address this problem.
However, while this specification is not the place to try to address these issues with local-parts, doing so is also not required to determine the outcome of this experiment. If this experiment succeeds then further work on email addresses with non-ASCII local-parts will be needed and that would be better based on the findings from this experiment, rather than doing nothing or starting this experiment based on a speculative approach to what is a very complex topic.
The OPENPGPKEY record allows an application or service to obtain an OpenPGP public key and use it for verifying a digital signature or encrypting a message to the public key. The DNS answer MUST pass DNSSEC validation; if DNSSEC validation reaches any state other than "Secure" (as specified in [RFC4035]), the DNSSEC validation MUST be treated as a failure.
If no OpenPGP public keys are known for an email address, an OPENPGPKEY DNS lookup MAY be performed to seek the OpenPGP public key that corresponds to that email address. This public key can then be used to verify a received signed message or can be used to send out an encrypted email message. An application whose attempt fails to retrieve a DNSSEC verified OPENPGPKEY RR from the DNS should remember that failure for some time to avoid sending out a DNS request for each email message the application is sending out; such DNS requests constitute a privacy leak.
Locally stored OpenPGP public keys are not automatically refreshed. If the owner of that key creates a new OpenPGP public key, that owner is unable to securely notify all users and applications that have its old OpenPGP public key. Applications and users can perform an OPENPGPKEY lookup to confirm the locally stored OpenPGP public key is still the correct key to use. If the locally stored OpenPGP public key is different from the DNSSEC validated OpenPGP public key currently published in DNS, the confirmation MUST be treated as a failure unless the locally stored OpenPGP key signed the newly published OpenPGP public key found in DNS. An application that can interact with the user MAY ask the user for guidance, otherwise the application will have to apply local policy. For privacy reasons, an application MUST NOT attempt to lookup an OpenPGP key from DNSSEC at every use of that key.
An OpenPGP public key can be associated with multiple email addresses by specifying multiple key uids. The OpenPGP public key obtained from a OPENPGPKEY RR can be used as long as the query and resulting data form a proper email to uid identity association.
CNAME's (see [RFC2181]) and DNAME's (see [RFC6672]) can be followed to obtain an OPENPGPKEY RR, as long as the original recipient's email address appears as one of the OpenPGP public key uids. For example, if the OPENPGPKEY RR query for hugh@example.com (8d57[...]b7._openpgpkey.example.com) yields a CNAME to 8d57[...]b7._openpgpkey.example.net, and an OPENPGPKEY RR for 8d57[...]b7._openpgpkey.example.net exists, then this OpenPGP public key can be used, provided one of the key uids contains "hugh@example.com". This public key cannot be used if it would only contain the key uid "hugh@example.net".
If one of the OpenPGP key uids contains only a single wildcard as the LHS of the email address, such as "*@example.com", the OpenPGP public key may be used for any email address within that domain. Wildcards at other locations (eg hugh@*.com) or regular expressions in key uids are not allowed, and any OPENPGPKEY RR containing these MUST be ignored.
Due to the expected size of the OPENPGPKEY record, applications SHOULD use TCP - not UDP - to perform queries for the OPENPGPKEY Resource Record.
Although the reliability of the transport of large DNS Resource Records has improved in the last years, it is still recommended to keep the DNS records as small as possible without sacrificing the security properties of the public key. The algorithm type and key size of OpenPGP keys should not be modified to accommodate this section.
OpenPGP supports various attributes that do not contribute to the security of a key, such as an embedded image file. It is recommended that these properties not be exported to OpenPGP public keyrings that are used to create OPENPGPKEY Resource Records. Some OpenPGP software, for example GnuPG, support a "minimal key export" that is well suited to use as OPENPGPKEY RDATA. See Appendix A.
DNSSEC is not an alternative for the "web of trust" or for manual fingerprint verification by users. DANE for OpenPGP as specified in this document is a solution aimed to ease obtaining someone's public key. Without manual verification of the OpenPGP key obtained via DANE, this retrieved key should only be used for encryption if the only other alternative is sending the message in plaintext. While this thwarts all passive attacks that simply capture and log all plaintext email content, it is not a security measure against active attacks. A user who publishes an OPENPGPKEY record in DNS still expects senders to perform their due diligence by additional (non-DNSSEC) verification of their public key via other out-of-band methods before sending any confidential or sensitive information.
In other words, the OPENPGPKEY record MUST NOT be used to send sensitive information without additional verification or confirmation that the OpenPGP key actually belongs to the target recipient.
DNSSEC does not protect the queries from Pervasive Monitoring as defined in [RFC7258]. Since DNS queries are currently mostly unencrypted, a query to lookup a target OPENPGPKEY record could reveal that a user using the (monitored) recursive DNS server is attempting to send encrypted email to a target. This information is normally protected by the MUAs and MTAs by using TLS encryption using STARTTLS. The DNS itself can mitigate some privacy concerns, but the user needs to select a trusted DNS server that supports these privay enhancing feaures. Recursive DNS servers can support DNS Query Name Minimalisation [RFC7816] which limits leaking the QNAME to only the recursive DNS server and the the nameservers of the actual zone being queried for. Recursive DNS servers can also support TLS [DNS-OVER-TLS] to ensure the path between the enduser and the recursive DNS server is encrypted.
Various components could be responsible for encrypting an email message to a target recipient. It could be done by the sender's MUA or a MUA plugin or the sender's MTA. Each of these have their own characteristics. A MUA can ask the user to make a decision before continuing. The MUA can either accept or refuse a message. The MTA must deliver the message as-is, or encrypt the message before delivering. Each of these components should attempt to encrypt an unencrypted outgoing message whenever possible.
In theory, two different local-parts could hash to the same value. This document assumes that such a hash collision has a negliable chance of happening.
Organisations that are required to be able to read everyone's encrypted email should publish the escrow key as the OPENPGPKEY record. Mail servers of such organizations MAY optionally re-encrypt the message to the individual's OpenPGP key.
An MTA could be operating in a stand-alone mode, without access to the sender's OpenPGP public keyring, or in a way where it can access the user's OpenPGP public keyring. Regardless, the MTA MUST NOT modify the user's OpenPGP keyring.
An MTA sending an email MUST NOT add the public key obtained from an OPENPGPKEY resource record to a permanent public keyring for future use beyond the TTL.
If the obtained public key is revoked, the MTA MUST NOT use the key for encryption, even if that would result in sending the message in plaintext.
If a message is already encrypted, the MTA SHOULD NOT re-encrypt the message, even if different encryption schemes or different encryption keys would be used.
If the DNS request for an OPENPGPKEY record returned an Indeterminate or Bogus answer as specified in [RFC4035], the MTA MUST NOT send the message and queue the plaintext message for encrypted delivery at a later time. If the problem persists, the email should be returned via the regular bounce methods.
If multiple non-revoked OPENPGPKEY resource records are found, the MTA SHOULD pick the most secure RR based on its local policy.
If the public key for a recipient obtained from the locally stored sender's public keyring differs from the recipient's OPENPGPKEY RR, the MUA SHOULD halt processing the message and interact with the user to resolve the conflict before continuing to process the message.
If the public key for a recipient obtained from the locally stored sender's public keyring contains contradicting properties for the same key obtained from an OPENPGPKEY RR, the MUA SHOULD NOT accept the message for delivery.
If multiple non-revoked OPENPGPKEY resource records are found, the MUA SHOULD pick the most secure OpenPGP public key based on its local policy.
The MUA MAY interact with the user to resolve any conflicts between locally stored keyrings and OPENPGPKEY RRdata.
A MUA that is encrypting a message SHOULD clearly indicate to the user the difference between encrypting to a locally stored and previously user-verified public key and encrypting to a public key obtained via an OPENPGPKEY resource record that was not manually verified by the user in the past.
To prevent amplification attacks, an Authoritative DNS server MAY wish to prevent returning OPENPGPKEY records over UDP unless the source IP address has been confirmed with [EDNS-COOKIE]. Such servers MUST NOT return REFUSED, but answer the query with an empty Answer Section and the truncation flag set ("TC=1").
The hashing of the local-part in this document is not a security feature. Publishing OPENPGPKEY records will create a list of hashes of valid email addresses, which could simplify obtaining a list of valid email addresses for a particular domain. It is desirable to not ease the harvesting of email addresses where possible.
The domain name part of the email address is not used as part of the hash so that hashes can be used in multiple zones deployed using DNAME [RFC6672]. This does makes it slightly easier and cheaper to brute-force the SHA2-256 hashes into common and short local-parts, as single rainbow tables can be re-used across domains. This can be somewhat countered by using NSEC3.
DNS zones that are signed with DNSSEC using NSEC for denial of existence are susceptible to zone-walking, a mechanism that allows someone to enumerate all the OPENPGPKEY hashes in a zone. This can be used in combination with previously hashed common or short local-parts (in rainbow tables) to deduce valid email addresses. DNSSEC-signed zones using NSEC3 for denial of existence instead of NSEC are significantly harder to brute-force after performing a zone-walk.
Users may have a local key store with OpenPGP public keys. An application supporting the use of OPENPGPKEY DNS records MUST NOT modify the local key store without explicit confirmation of the user, as the application is unaware of the user's personal policy for adding, removing or updating their local key store. An application MAY warn the user if an OPENPGPKEY record does not match the OpenPGP public key in the local key store.
Applications that cannot interact with users, such as daemon processes, SHOULD store OpenPGP public keys obtained via OPENPGPKEY up to their DNS TTL value. This avoids repeated DNS lookups that third parties could monitor to determine when an email is being sent to a particular user.
Anyone who can obtain a DNSSEC private key of a domain name via coercion, theft or brute force calculations, can replace any OPENPGPKEY record in that zone and all of the delegated child zones. Any future messages encrypted with the malicious OpenPGP key could then be read.
Therefore, an OpenPGP key obtained via a DNSSEC validated OPENPGPKEY record can only be trusted as much as the DNS domain can be trusted, and is no substitute for in-person OpenPGP key verification or additional Openpgp verification via "Web Of Trust" signatures present on the OpenPGP in question.
[RFC Editor Note: Please remove this entire seciton prior to publication as an RFC.]
This section records the status of known implementations of the protocol defined by this specification at the time of posting of this Internet-Draft, and is based on a proposal described in [RFC6982]. The description of implementations in this section is intended to assist the IETF in its decision processes in progressing drafts to RFCs. Please note that the listing of any individual implementation here does not imply endorsement by the IETF. Furthermore, no effort has been spent to verify the information presented here that was supplied by IETF contributors. This is not intended as, and must not be construed to be, a catalog of available implementations or their features. Readers are advised to note that other implementations may exist. According to RFC 6982, "this will allow reviewers and working groups to assign due consideration to documents that have the benefit of running code, which may serve as evidence of valuable experimentation and feedback that have made the implemented protocols more mature. It is up to the individual working groups to use this information as they see fit."
This document uses a new DNS RR type, OPENPGPKEY, whose value 61 has been allocated by IANA from the Resource Record (RR) TYPEs subregistry of the Domain Name System (DNS) Parameters registry.
The IANA template for OPENPGPKEY is listed in Appendix B. It was submitted to IANA on July 23, 2014, reference number #773394 and approved on August 12, 2014.
This document is based on [RFC4255] and [draft-ietf-dane-smime] whose authors are Paul Hoffman, Jacob Schlyter and W. Griffin. Olafur Gudmundsson provided feedback and suggested various improvements. Willem Toorop contributed the gpg and hexdump command options. Daniel Kahn Gillmor provided the text describing the OpenPGP packet formats and filtering options. Edwin Taylor contributed language improvements for various iterations of this document. Text regarding email mappings was taken from [draft-levine-dns-mailbox] whose author is John Levine.
gpg --export --export-options export-minimal,no-export-attributes \
hugh@example.com | base64
The commonly available GnuPG software can be used to generate a minimum Transferable Public Key for the RRdata portion of an OPENPGPKEY record:
gpg --export --export-options export-minimal,no-export-attributes \
hugh@example.com | wc -c
gpg --export --export-options export-minimal,no-export-attributes \
hugh@example.com | hexdump -e \
'"\t" /1 "%.2x"' -e '/32 "\n"'
<SHA2-256-trunc(hugh)>._openpgpkey.example.com. IN TYPE61 \# \
<numOctets> <keydata in hex>
When DNS software reading or signing the zone file does not yet support the OPENPGPKEY RRtype, the Generic Record Syntax of [RFC3597] can be used to generate the RDATA. One needs to calculate the number of octets and the actual data in hexadecimal:
~> openpgpkey --output rfc hugh@example.com c9[..]d6._openpgpkey.example.com. IN OPENPGPKEY mQCNAzIG[...] ~> openpgpkey --output generic hugh@example.com c9[..]d6._openpgpkey.example.com. IN TYPE61 \# 2313 99008d03[...]
The openpgpkey command in the hash-slinger software can be used to generate complete OPENPGPKEY records
A. Submission Date: 23-07-2014 B.1 Submission Type: [x] New RRTYPE [ ] Modification to RRTYPE B.2 Kind of RR: [x] Data RR [ ] Meta-RR C. Contact Information for submitter (will be publicly posted): Name: Paul Wouters Email Address: pwouters@redhat.com International telephone number: +1-647-896-3464 Other contact handles: paul@nohats.ca D. Motivation for the new RRTYPE application. Publishing RFC-4880 OpenPGP formatted keys in DNS with DNSSEC protection to faciliate automatic encryption of emails in defense against pervasive monitoring. E. Description of the proposed RR type. http://tools.ietf.org/html/draft-ietf-dane-openpgpkey-00#section-2 F. What existing RRTYPE or RRTYPEs come closest to filling that need and why are they unsatisfactory? The CERT RRtype is the closest match. It unfortunately depends on subtyping, and its use in general is no longer recommended. It also has no human usable presentation format. Some usage types of CERT require external URI's which complicates the security model. This was discussed in the dane working group. G. What mnemonic is requested for the new RRTYPE (optional)? OPENPGPKEY H. Does the requested RRTYPE make use of any existing IANA registry or require the creation of a new IANA subregistry in DNS Parameters? If so, please indicate which registry is to be used or created. If a new subregistry is needed, specify the allocation policy for it and its initial contents. Also include what the modification procedures will be. The RDATA part uses the key format specified in RFC-4880, which itself uses https://www.iana.org/assignments/pgp-parameters/pgp-parameters.xhtm This RRcode just uses the formats specified in those registries for its RRdata part. I. Does the proposal require/expect any changes in DNS servers/resolvers that prevent the new type from being processed as an unknown RRTYPE (see [RFC3597])? No. J. Comments: Currently, three software implementations of draft-ietf-dane-openpgpkey are using a private number.