DNS over DTLS (DNSoD)

The Domain Name System is specified in and . DNS queries and responses are normally exchanged unencrypted and are thus vulnerable to eavesdropping. Such eavesdropping can result in an undesired entity learning domains that a host wishes to access, thus resulting in privacy leakage. DNS privacy problem is further discussed in . Active attackers have long been successful at injecting bogus responses, causing cache poisoning and causing misdirection of the subsequent connection (if attacking A or AAAA records). A popular mitigation against that attack is to use ephemeral and random source ports for DNS queries . This document defines DNS over DTLS (DNSoD, pronounced "dee-enn-sod") which provides confidential DNS communication between stub resolvers and recursive resolvers, stub resolvers and forwarders, forwarders and recursive resolvers. The motivations for proposing DNSoD are that TCP suffers from network head-of-line blocking, where the loss of a packet causes all other TCP segments to not be delivered to the application until the lost packet is re-transmitted. DNSoD, because it uses UDP, does not suffer from network head-of-line blocking. DTLS session resumption consumes 1 round trip whereas TLS session resumption can start only after TCP handshake is complete. Although TCP Fast Open can reduce that handshake, TCP Fast Open is only available on a few OSs, it is not yet ubiquitous.

DNS queries can be sent over UDP or TCP. The scope of this document, however, is only UDP. DNS over TCP could be protected with TLS, as described in . DNS Security Extensions ( DNSSEC ) provides object integrity of DNS resource records, allowing end-users (or their resolver) to verify legitimacy of responses. However, DNSSEC does not protect privacy of DNS requests or responses. DNSoD works in conjunction with DNSSEC, but DNSoD does not replace the need or value of DNSSEC.

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 .

DNSoD MUST run over standard UDP port 853 as defined in . The host should determine if the DNS server supports DNSoD by sending a DTLS ClientHello message. A DNS server that does not support DNSoD will not respond to ClientHello messages sent by the client. If no response is received from that server, and the client has no better round-trip estimate, the client MUST retransmit the DTLS ClientHello according to Section 4.2.4.1 of . After 15 seconds, it MUST cease attempts to re-transmit its ClientHello. The client MAY repeat that procedure in the event the DNS server upgrades to support DNSoD, but such probing SHOULD NOT be done more frequently than every 24 hours and MUST NOT be done more frequently than every 15 minutes. This mechanism requires no additional signaling between the client and server. Behavior for an unsuccessful DTLS connection is discussed in .

Once the DNS client succeeds in receiving HelloVerifyRequest from the server via UDP on the well-known port for DNS over DTLS, it proceeds with DTLS handshake as described in , following the best practices specified in . DNS privacy requires encrypting the query (and response) from passive attacks. Such encryption typically provides integrity protection as a side-effect, which means on-path attackers cannot simply inject bogus DNS responses. However, to provide stronger protection from active attackers pretending to be the server, the server itself needs to be authenticated. To authenticate the server providing DNS privacy, DNS client can use the authenication mechanisms discussed in . This document does not propose new ideas for authentication. After DTLS negotiation completes, the connection will be encrypted and is now protected from eavesdropping.

In DTLS, all data is protected using the same record encoding and mechanisms. When the mechanism described in this document is in effect, DNS messages are encrypted using the standard DTLS record encoding. When a user of DTLS wishes to send an DNS message, it delivers it to the DTLS implementation as an ordinary application data write (e.g., SSL_write()). To reduce client and server workload, clients SHOULD re-use the DTLS session. A single DTLS session can be used to send multiple DNS requests and receive multiple DNS responses. DNSoD client and server can use DTLS heartbeat to verify that the peer still has DTLS state. DTLS session is terminated by the receipt of an authenticated message that closes the connection (e.g., a DTLS fatal alert).

<------- HelloVerifyRequest (contains cookie) ClientHello --------> (contains cookie) (empty SessionTicket extension) ServerHello (empty SessionTicket extension) Certificate* ServerKeyExchange* CertificateRequest* <-------- ServerHelloDone Certificate* ClientKeyExchange CertificateVerify* (ChangeCipherSpec) Finished --------> NewSessionTicket (ChangeCipherSpec) <-------- Finished DNS Request ---------> <--------- DNS Response ]]>

To reduce number of octets of the DTLS handshake, especially the size of the certificate in the ServerHello (which can be several kilobytes), DNS client and server can use raw public keys or Cached Information Extension . Cached Information Extension avoids transmitting the server's certificate and certificate chain if the client has cached that information from a previous TLS handshake. Since pipelined responses can arrive out of order, clients MUST match responses to outstanding queries on the same DTLS connection using the Message ID. If the response contains a question section, the client MUST match the QNAME, QCLASS, and QTYPE fields. Failure by clients to properly match responses to outstanding queries can have serious consequences for interoperability ( , Section 7). It is highly advantageous to avoid server-side DTLS state and reduce the number of new DTLS sessions on the server which can be done with . This also eliminates a round-trip for subsequent DNSoD queries, because with the DTLS session does not need to be re-established. Compared to normal DNS, DTLS adds at least 13 octets of header, plus cipher and authentication overhead to every query and every response. This reduces the size of the DNS payload that can be carried. DNS client and server MUST support the EDNS0 option defined in so that the DNS client can indicate to the DNS server the maximum DNS response size it can reassemble and deliver in the DNS client’s network stack. The client sets its EDNS0 value as if DTLS is not being used. The DNS server must ensure that the DNS response size does not exceed the Path MTU. The DNS server must consider the amount of record expansion expected by the DTLS processing when calculating the size of DNS response that fits within the path MTU. Path MTU MUST be greater than equal to [DNS response size + DTLS overhead of 13 octets + padding size () + authentication overhead of the negotiated DTLS cipher suite + block padding (Section 4.1.1.1 of ]. If the DNS server's response were to exceed that calculated value, the server sends a response that does fit within that value and sets the TC (truncated) bit. The client, upon receiving a response with the TC bit set and wanting to receive the entire response, establishes a DNS-over-TLS connection to the same server, and sends a new DNS request for the same resource record. DNSoD puts an additional computational load on servers. The largest gain for privacy is to protect the communication between the DNS client (the end user's machine) and its caching resolver.

DNS servers are often configured with anycast addresses. While the network is stable, packets transmitted from a particular source to an anycast address will reach the same server that has the cryptographic context from the DNS over DTLS handshake. But when the network configuration changes, a DNS over DTLS packet can be received by a server that does not have the necessary cryptographic context. To encourage the client to initiate a new DTLS handshake, DNS servers SHOULD generate a DTLS Alert message in response to receiving a DTLS packet for which the server does not have any cryptographic context. Upon receipt of an un-authenicated DTLS alert, the DTLS client validates the Alert is within the replay window (Section 4.1.2.6 of ). It is difficult for the DTLS client to validate that the DTLS alert was generated by the DTLS server in response to a request or was generated by an on- or off-path attacker. Thus, upon receipt of an in-window DTLS Alert, the client SHOULD continue re-transmitting the DTLS packet (in the event the Alert was spoofed), and at the same time it SHOULD initiate DTLS session resumption. When the DTLS client receives authenticated DNS response from one of those DTLS sessions, the other DTLS session should be terminated.

Using DNS privacy with an authenticated server is most preferred, DNS privacy with an unauthenticated server is next preferred, and plain DNS is least preferred. This section gives a non-normative discussion on common behaviors and choices. An implementation MAY attempt to obtain DNS privacy by contacting DNS servers on the local network (provided by DHCP) and on the Internet, and make those attempts in parallel to reduce user impact. If DNS privacy cannot be successfully negotiated for whatever reason, the client can do three things, in order from best to worst for privacy: refuse to send DNS queries on this network, which means the client cannot make effective use of this network, as modern networks require DNS; or, use opportunistic security, as described in or, send plain DNS queries on this network, which means no DNS privacy is provided. Heuristics can improve this situation, but only to a degree (e.g., previous success of DNS privacy on this network may be reason to alert the user about failure to establish DNS privacy on this network now). Still, the client (in cooperation with the end user) has to decide to use the network without the protection of DNS privacy.

This specification uses port 853 already allocated in the IANA port number registry as defined in Section 6 of .

The interaction between a DNS client and DNS server requires Datagram Transport Layer Security (DTLS) with a ciphersuite offering confidentiality protection and guidance given in must be followed to avoid attacks on DTLS. DNS clients keeping track of servers known to support DTLS enables clients to detect downgrade attacks. To interfere with DNS over DTLS, an on- or off-path attacker might send an ICMP message towards the DTLS client or DTLS server. As these ICMP messages cannot be authenticated, all ICMP errors should be treated as soft errors . For servers with no connection history and no apparent support for DTLS, depending on their Privacy Profile and privacy requirements, clients may choose to (a) try another server when available, (b) continue without DTLS, or (c) refuse to forward the query. Once a DNSoD client has established a security association with a particular DNS server, and outstanding normal DNS queries with that server (if any) have been received, the DNSoD client MUST ignore any subsequent normal DNS responses from that server, as all subsequent responses should be encrypted. This behavior mitigates all possible attacks described in Measures for Making DNS More Resilient against Forged Answers . A malicious client might attempt to perform a high number of DTLS handshakes with a server. As the clients are not uniquely identified by the protocol and can be obfuscated with IPv4 address sharing and with IPv6 temporary addresses, a server needs to mitigate the impact of such an attack. Such mitigation might involve rate limiting handshakes from a certain subnet or more advanced DoS/DDoS techniques beyond the scope of this paper.

Thanks to Phil Hedrick for his review comments on TCP and to Josh Littlefield for pointing out DNSoD load on busy servers (most notably root servers). The authors would like to thank Simon Josefsson, Daniel Kahn Gillmor, Bob Harold, Ilari Liusvaara, Sara Dickinson, Christian Huitema, Stephane Bortzmeyer and Geoff Huston for discussions and comments on the design of DNSoD.