The GNU Name System

3.1. Record Types

A registry of GNS Record Types is described in Section 10. The registration policy for this registry is "First Come First Served", as described in [RFC8126]. When requesting new entries, careful consideration of the following criteria is strongly advised: FIXME: ausdenken was wir da gerne haetten.¶

3.2. PKEY

In GNS, a delegation of a label to a zone is represented through a PKEY record. A PKEY resource record contains the public key of the zone to delegate to. A PKEY record MUST be the only record under a label. No other records are allowed. A PKEY DATA entry has the following format:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|                   PUBLIC KEY                  |
|                                               |
|                                               |
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+

where:¶

PUBLIC KEY

A 256-bit ECDSA zone key.¶

3.3. GNS2DNS

It is possible to delegate a label back into DNS through a GNS2DNS record. The resource record contains a DNS name for the resolver to continue with in DNS followed by a DNS server. Both names are in the format defined in [RFC1034] for DNS names. A GNS2DNS DATA entry has the following format:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|                    DNS NAME                   |
/                                               /
/                                               /
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+
|                 DNS SERVER NAME               |
/                                               /
/                                               /
|                                               |
+-----------------------------------------------+

where:¶

DNS NAME

The name to continue with in DNS (0-terminated).¶

DNS SERVER NAME

The DNS server to use. May be an IPv4/IPv6 address in dotted decimal form or a DNS name. It may also be a relative GNS name ending with a "+" top-level domain. The value is UTF-8 encoded (also for DNS names) and 0-terminated.¶

3.4. LEHO

Legacy hostname records can be used by applications that are expected to supply a DNS name on the application layer. The most common use case is HTTP virtual hosting, which as-is would not work with GNS names as those may not be globally unique. A LEHO resource record is expected to be found together in a single resource record with an IPv4 or IPv6 address. A LEHO DATA entry has the following format:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|                 LEGACY HOSTNAME               |
/                                               /
/                                               /
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+

where:¶

LEGACY HOSTNAME

A UTF-8 string (which is not 0-terminated) representing the legacy hostname.¶

NOTE: If an application uses a LEHO value in an HTTP request header (e.g. "Host:" header) it must be converted to a punycode representation [RFC5891].¶

3.5. NICK

Nickname records can be used by zone administrators to publish an indication on what label this zone prefers to be referred to. This is a suggestion to other zones what label to use when creating a PKEY (Section 3.2) record containing this zone's public zone key. This record SHOULD only be stored under the empty label "@" but MAY be returned with record sets under any label as a supplemental record. Section 6.2.6 details how a resolver must process supplemental and non-supplemental NICK records. A NICK DATA entry has the following format:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|                  NICKNAME                     |
/                                               /
/                                               /
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+

where:¶

NICKNAME

A UTF-8 string (which is not 0-terminated) representing the preferred label of the zone. This string MUST NOT include a "." character.¶

3.6. BOX

In GNS, every "." in a name delegates to another zone, and GNS lookups are expected to return all of the required useful information in one record set. This is incompatible with the special labels used by DNS for SRV and TLSA records. Thus, GNS defines the BOX record format to box up SRV and TLSA records and include them in the record set of the label they are associated with. For example, a TLSA record for "_https._tcp.foo.gnu" will be stored in the record set of "foo.gnu" as a BOX record with service (SVC) 443 (https) and protocol (PROTO) 6 (tcp) and record TYPE "TLSA". For reference, see also [RFC2782]. A BOX DATA entry has the following format:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|   PROTO   |    SVC    |       TYPE            |
+-----------+-----------------------------------+
|                 RECORD DATA                   |
/                                               /
/                                               /
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+

where:¶

PROTO

the 16-bit protocol number, e.g. 6 for tcp. In network byte order.¶

SVC

the 16-bit service value of the boxed record, i.e. the port number. In network byte order.¶

TYPE

is the 32-bit record type of the boxed record. In network byte order.¶

RECORD DATA

is a variable length field containing the "DATA" format of TYPE as defined for the respective TYPE in DNS.¶

3.7. VPN

A VPN DATA entry has the following format:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|          HOSTING PEER PUBLIC KEY              |
|                (256 bits)                     |
|                                               |
|                                               |
+-----------+-----------------------------------+
|   PROTO   |    SERVICE  NAME                  |
+-----------+                                   +
/                                               /
/                                               /
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+

where:¶

HOSTING PEER PUBLIC KEY

is a 256-bit EdDSA public key identifying the peer hosting the service.¶

PROTO

the 16-bit protocol number, e.g. 6 for TCP. In network byte order.¶

SERVICE NAME

a shared secret used to identify the service at the hosting peer, used to derive the port number requird to connect to the service. The service name MUST be a 0-terminated UTF-8 string.¶

4.1. Key Derivations

Given a label, the DHT key "q" is derived as follows:¶

PRK_h := HKDF-Extract ("key-derivation", zk)
h := HKDF-Expand (PRK_h, label | "gns", 512 / 8)
d_h := h * d mod L
zk_h := h mod L * zk
q := SHA512 (zk_h)

We use a hash-based key derivation function (HKDF) as defined in [RFC5869]. We use HMAC-SHA512 for the extraction phase and HMAC-SHA256 for the expansion phase.¶

PRK_h

is key material retrieved using an HKDF using the string "key-derivation" as salt and the public zone key "zk" as initial keying material.¶

h

is the 512-bit HKDF expansion result. The expansion info input is a concatenation of the label and string "gns".¶

d

is the 256-bit private zone key as defined in Section 2.¶

label

is a UTF-8 string under which the resource records are published.¶

d_h

is a 256-bit private key derived from the "d" using the keying material "h".¶

zk_h

is a 256-bit public key derived from the zone key "zk" using the keying material "h".¶

L

is the prime-order subgroup as defined in Section 2.¶

q

Is the 512-bit DHT key under which the resource records block is published. It is the SHA512 hash over the public key "zk_h" corresponding to the derived private key "d_h".¶

We point out that the multiplication of "zk" with "h" is a point multiplication, while the multiplication of "d" with "h" is a scalar multiplication.¶

4.2. Resource Records Block

GNS records are grouped by their labels and published as a single block in the DHT. The grouped record sets MAY be paired with any number of supplemental records. Supplemental records must have the supplemental flag set (See Section 3). The contained resource records are encrypted using a symmetric encryption scheme. A GNS implementation must publish RRBLOCKs in accordance to the properties and recommendations of the underlying DHT. This may include a periodic refresh publication. A GNS RRBLOCK has the following format:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|                   SIGNATURE                   |
|                                               |
|                                               |
|                                               |
|                                               |
|                                               |
|                                               |
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+
|                  PUBLIC KEY                   |
|                                               |
|                                               |
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+
|         SIZE          |       PURPOSE         |
+-----+-----+-----+-----+-----+-----+-----+-----+
|                   EXPIRATION                  |
+-----+-----+-----+-----+-----+-----+-----+-----+
|                    BDATA                      /
/                                               /
/                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+

where:¶

SIGNATURE

A 512-bit ECDSA deterministic signature compliant with [RFC6979]. The signature is computed over the data following the PUBLIC KEY field. The signature is created using the derived private key "d_h" (see Section 4).¶

PUBLIC KEY

is the 256-bit public key "zk_h" to be used to verify SIGNATURE. The wire format of this value is defined in [RFC8032], Section 5.1.5.¶

SIZE

A 32-bit value containing the length of the signed data following the PUBLIC KEY field in network byte order. This value always includes the length of the fields SIZE (4), PURPOSE (4) and EXPIRATION (8) in addition to the length of the BDATA. While a 32-bit value is used, implementations MAY refuse to publish blocks beyond a certain size significantly below 4 GB. However, a minimum block size of 62 kilobytes MUST be supported.¶

PURPOSE

A 32-bit signature purpose flag. This field MUST be 15 (in network byte order).¶

EXPIRATION

Specifies when the RRBLOCK expires and the encrypted block SHOULD be removed from the DHT and caches as it is likely stale. However, applications MAY continue to use non-expired individual records until they expire. The value MUST be set to the expiration time of the resource record contained within this block with the smallest expiration time. If a records block includes shadow records, then the maximum expiration time of all shadow records with matching type and the expiration times of the non-shadow records is considered. This is a 64-bit absolute date in microseconds since midnight (0 hour), January 1, 1970 in network byte order.¶

BDATA

The encrypted resource records with a total size of SIZE - 16.¶

4.3. Record Data Encryption and Decryption

A symmetric encryption scheme is used to encrypt the resource records set RDATA into the BDATA field of a GNS RRBLOCK. The wire format of the RDATA looks as follows:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|     RR COUNT          |        EXPIRA-        /
+-----+-----+-----+-----+-----+-----+-----+-----+
/         -TION         |       DATA SIZE       |
+-----+-----+-----+-----+-----+-----+-----+-----+
|         TYPE          |          FLAGS        |
+-----+-----+-----+-----+-----+-----+-----+-----+
|                      DATA                     /
/                                               /
/                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+
|                   EXPIRATION                  |
+-----+-----+-----+-----+-----+-----+-----+-----+
|       DATA SIZE       |          TYPE         |
+-----+-----+-----+-----+-----+-----+-----+-----+
|           FLAGS       |        DATA           /
+-----+-----+-----+-----+                       /
/                       +-----------------------/
/                       |                       /
+-----------------------+                       /
/                     PADDING                   /
/                                               /

where:¶

RR COUNT

A 32-bit value containing the number of variable-length resource records which are following after this field in network byte order.¶

EXPIRATION, DATA SIZE, TYPE, FLAGS and DATA

These fields were defined in the resource record format in Section 3. There MUST be a total of RR COUNT of these resource records present.¶

PADDING

The padding MUST contain the value 0 in all octets. The padding MUST ensure that the size of the RDATA WITHOUT the RR COUNT field is a power of two. As a special exception, record sets with (only) a PKEY record type are never padded. Note that a record set with a PKEY record MUST NOT contain other records.¶

The symmetric keys and initialization vectors are derived from the record label and the zone key "zk". For decryption of the resource records block payload, the key material "K" and initialization vector "IV" for the symmetric cipher are derived as follows:¶

PRK_k := HKDF-Extract ("gns-aes-ctx-key", zk)
PRK_iv := HKDF-Extract ("gns-aes-ctx-iv", zk)
K := HKDF-Expand (PRK_k, label, 512 / 8);
IV := HKDF-Expand (PRK_iv, label, 256 / 8)

HKDF is a hash-based key derivation function as defined in [RFC5869]. Specifically, HMAC-SHA512 is used for the extraction phase and HMAC-SHA256 for the expansion phase. The output keying material is 64 octets (512 bit) for the symmetric keys and 32 octets (256 bit) for the initialization vectors. We divide the resulting keying material "K" into a 256-bit AES [RFC3826] key and a 256-bit TWOFISH [TWOFISH] key:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|                    AES KEY                    |
|                                               |
|                                               |
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+
|                  TWOFISH KEY                  |
|                                               |
|                                               |
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+

Similarly, we divide "IV" into a 128-bit initialization vector and a 128-bit initialization vector:¶

0     8     16    24    32    40    48    56
+-----+-----+-----+-----+-----+-----+-----+-----+
|                    AES IV                     |
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+
|                  TWOFISH IV                   |
|                                               |
+-----+-----+-----+-----+-----+-----+-----+-----+

The keys and IVs are used for a CFB128-AES-256 and CFB128-TWOFISH-256 chained symmetric cipher. Both ciphers are used in Cipher FeedBack (CFB) mode [RFC3826].¶

RDATA := AES(K[0:31], IV[0:15],
             TWOFISH(K[32:63], IV[16:31], BDATA))
BDATA := TWOFISH(K[32:63], IV[16:31],
                 AES(K[0:31], IV[0:15], RDATA))

6.1. Recursion

In each step of the recursive name resolution, there is an authoritative zone zk and a name to resolve. The name may be empty. Initially, the authoritative zone is the start zone. If the name is empty, it is interpreted as the apex label "@".¶

From here, the following steps are recursively executed, in order:¶

1. Extract the right-most label from the name to look up.¶
2. Calculate q using the label and zk as defined in Section 4.1.¶
3. Perform a DHT query GET(q) to retrieve the RRBLOCK.¶
4. Verify and process the RRBLOCK and decrypt the BDATA contained in it as defined in Section 4.3.¶

Upon receiving the RRBLOCK from the DHT, apart from verifying the provided signature, the resolver MUST check that the authoritative zone key was used to sign the record: The derived zone key "h*zk" MUST match the public key provided in the RRBLOCK, otherwise the RRBLOCK MUST be ignored and the DHT lookup GET(q) MUST continue.¶

6.2. Record Processing

Record processing occurs at the end of a single recursion. We assume that the RRBLOCK has been cryptographically verified and decrypted. At this point, we must first determine if we have received a valid record set in the context of the name we are trying to resolve:¶

1. Case 1: If the remainder of the name to resolve is empty and the record set does not consist of a PKEY, CNAME or DNS2GNS record, the record set is the result and the recursion is concluded.¶
2. Case 2: If the name to be resolved is of the format "_SERVICE._PROTO" and the record set contains one or more matching BOX records, the records in the BOX records are the result and the recusion is concluded (Section 6.2.4).¶
3. Case 3: If the remainder of the name to resolve is not empty and does not match the "_SERVICE._PROTO" syntax, then the current record set MUST consist of a single PKEY record (Section 6.2.1), a single CNAME record (Section 6.2.3), or one or more GNS2DNS records (Section 6.2.2), which are processed as described in the respective sections below. The record set may include any number of supplemental records. Otherwise, resolution fails and the resolver MUST return an empty record set. Finally, after the recursion terminates, the client preferences for the record type SHOULD be considered. If a VPN record is found and the client requests an A or AAAA record, the VPN record SHOULD be converted (Section 6.2.5) if possible.¶

Query: alice.doe (type=A)
Result:
A: 1.2.3.4
NICK: eve

Query: alice.doe (type=A)
Result:
A: 1.2.3.4
NICK: john (Supplemental)

9.1. Cryptography

The security of cryptographic systems depends on both the strength of the cryptographic algorithms chosen and the strength of the keys used with those algorithms. The security also depends on the engineering of the protocol used by the system to ensure that there are no non-cryptographic ways to bypass the security of the overall system.¶

This document concerns itself with the selection of cryptographic algorithms for use in GNS. The algorithms identified in this document are not known to be broken (in the cryptographic sense) at the current time, and cryptographic research so far leads us to believe that they are likely to remain secure into the foreseeable future. However, this isn't necessarily forever, and it is expected that new revisions of this document will be issued from time to time to reflect the current best practices in this area.¶

GNS uses ECDSA over Curve25519. This is an unconventional choice, as ECDSA is usually used with other curves. However, traditional ECDSA curves are problematic for a range of reasons described in the Curve25519 and EdDSA papers. Using EdDSA directly is also not possible, as a hash function is used on the private key which destroys the linearity that the GNU Name System depends upon. We are not aware of anyone suggesting that using Curve25519 instead of another common curve of similar size would lower the security of ECDSA. GNS uses 256-bit curves because that way the encoded (public) keys fit into a single DNS label, which is good for usability.¶

9.2. Abuse mitigation

GNS names are UTF-8 strings. Consequently, GNS faces similar issues with respect to name spoofing as DNS does for internationalized domain names. In DNS, attackers may register similar sounding or looking names (see above) in order to execute phishing attacks. GNS zone administrators must take into account this attack vector and incorporate rules in order to mitigate it.¶

Further, DNS can be used to combat illegal content on the internet by having the respective domains seized by authorities. However, the same mechanisms can also be abused in order to impose state censorship, which ist one of the motivations behind GNS. Hence, such a seizure is, by design, difficult to impossible in GNS. In particular, GNS does not support WHOIS ([RFC3912]).¶

9.3. Zone management

In GNS, zone administrators need to manage and protect their zone keys. Once a zone key is lost it cannot be recovered. Once it is compromised it cannot be revoked (unless a revocation message was pre-calculated and is still available). Zone administrators, and for GNS this includes end-users, are required to responsibly and dilligently protect their cryptographic keys. Offline signing is in principle possible, but GNS does not support separate zone signing and key-signing keys (as in [RFC6781]) in order to provide usable security.¶

Similarly, users are required to manage their local root zone. In order to ensure integrity and availability or names, users must ensure that their local root zone information is not compromised or outdated. It can be expected that the processing of zone revocations and an initial root zone is provided with a GNS client implementation ("drop shipping"). Extension and customization of the zone is at the full discretion of the user.¶

9.4. Impact of underlying DHT

This document does not specifiy the properties of the underlying distributed hash table (DHT) which is required by any GNS implementation. For implementors, it is important to note that the properties of the DHT are directly inherited by the GNS implementation. This includes both security as well as other non-functional properties such as scalability and performance. Implementors should take great care when selecting or implementing a DHT for use in a GNS implementation. DHTs with string security and performance guarantees exist [R5N]. It should also be taken into consideration that GNS implementations which build upon different DHT overlays are unlikely to be interoperable with each other.¶

9.5. Revocations

Zone administrators are advised to pre-generate zone revocations and securely store the revocation information in case the zone key is lost, compromised or replaced in the furture. Pre-calculated revocations may become invalid due to expirations or protocol changes such as epoch adjustments. Consequently, implementors and users must make precautions in order to manage revocations accordingly.¶

Revocation payloads do NOT include a 'new' key for key replacement. Inclusion of such a key would have two major disadvantages:¶

If revocation is used after a private key was compromised, allowing key replacement would be dangerous: if an adversary took over the private key, the adversary could then broadcast a revocation with a key replacement. For the replacement, the compromised owner would have no chance to issue even a revocation. Thus, allowing a revocation message to replace a private key makes dealing with key compromise situations worse.¶

Sometimes, key revocations are used with the objective of changing cryptosystems. Migration to another cryptosystem by replacing keys via a revocation message would only be secure as long as both cryptosystems are still secure against forgery. Such a planned, non-emergency migration to another cryptosystem should be done by running zones for both ciphersystems in parallel for a while. The migration would conclude by revoking the legacy zone key only once it is deemed no longer secure, and hopefully after most users have migrated to the replacement.¶