Internet-Draft | Roughtime | August 2024 |
Ladd & Dansarie | Expires 3 February 2025 | [Page] |
This document specifies Roughtime - a protocol that aims to achieve rough time synchronization even for clients without any idea of what time it is.¶
This note is to be removed before publishing as an RFC.¶
Status information for this document may be found at https://datatracker.ietf.org/doc/draft-ietf-ntp-roughtime/.¶
Source for this draft and an issue tracker can be found at https://github.com/wbl/roughtime-draft.¶
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 https://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 3 February 2025.¶
Copyright (c) 2024 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 Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
Time synchronization is essential to Internet security as many security protocols and other applications require synchronization [RFC738]. Unfortunately, widely deployed protocols such as the Network Time Protocol (NTP) [RFC5905] lack essential security features, and even newer protocols like Network Time Security (NTS) [RFC8915] lack mechanisms to ensure that the servers behave correctly. Furthermore, clients may lack even a basic idea of the time, creating bootstrapping problems. Roughtime is intended to permit devices to obtain a rough idea of the current time from fairly static configuration consisting of a key and a server.¶
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.¶
Roughtime is a protocol for rough time synchronization that enables clients to provide cryptographic proof of server malfeasance. It does so by having responses from servers include a signature over a value derived from a nonce in the client request. This provides cryptographic proof that the timestamp was issued after the server received the client's request. The derived value included in the server's response is the root of a Merkle tree which includes the hash of the client's nonce as the value of one of its leaf nodes. This enables the server to amortize the relatively costly signing operation over a number of client requests.¶
Single server mode: At its most basic level, Roughtime is a one round protocol in which a completely fresh client requests the current time and the server sends a signed response. The response includes a timestamp and a radius used to indicate the server's certainty about the reported time.¶
The server proves freshness of its response as follows. The client's request contains a nonce which the server incorporates into its signed response. The client can verify the server's signatures and - provided that the nonce has sufficient entropy - this proves that the signed response could only have been generated after the nonce.¶
A Roughtime server guarantees that a response to a query sent at t1, received at t2, and with timestamp t3 has been created between t1 and t2. If t3 is not within that interval, a server inconsistency may be detected and used to impeach the server. The propagation of such a guarantee and its use for time synchronization is discussed in Section 7. No delay attacker may affect this: they may only expand the interval between t1 and t2, or of course stop the measurement in the first place.¶
Roughtime messages are maps consisting of one or more (tag, value) pairs. They start with a header, which contains the number of pairs, the tags, and value offsets. The header is followed by a message values section which contains the values associated with the tags in the header. Messages MUST be formatted according to Figure 1 as described in the following sections.¶
Messages MAY be recursive, i.e. the value of a tag can itself be a Roughtime message.¶
0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Number of pairs (uint32) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | . . . N-1 offsets (uint32) . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | . . . N tags (uint32) . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | . . . Values . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
An int32 is a 32 bit signed integer. It is serialized least significant byte first in sign-magnitude representation with the sign bit in the most significant bit. The negative zero value (0x80000000) MUST NOT be used and any message with it is syntactically invalid and MUST be ignored.¶
A uint32 is a 32 bit unsigned integer. It is serialized with the least significant byte first.¶
A uint64 is a 64 bit unsigned integer. It is serialized with the least significant byte first.¶
Tags are used to identify values in Roughtime messages. A tag is a uint32 but can also be represented as a sequence of up to four ASCII characters [RFC20]. ASCII strings shorter than four characters can be unambiguously converted to tags by padding them with zero bytes. Tags MUST NOT contain any other bytes than capital letters (A-Z) or padding zero bytes. For example, the ASCII string "NONC" would correspond to the tag 0x434e4f4e and "ZZZZ" would correspond to 0x5a5a5a5a. Note that when encoded into a message the ASCII values will be in the natural bytewise order.¶
A timestamp is a representation of UTC time as a uint64 count of seconds since 00:00:00 on 1 January 1970 (the Unix epoch), assuming every day has 86400 seconds. This is a constant offset from the NTP timestamp in seconds. Leap seconds do not have an unambiguous representation in a timestamp, and this has implications for the attainable accuracy and setting of the RADI tag.¶
All Roughtime messages start with a header. The first four bytes of the header is the uint32 number of tags N, and hence of (tag, value) pairs.¶
The following 4*(N-1) bytes are offsets, each a uint32. The last 4*N bytes in the header are tags. Offsets refer to the positions of the values in the message values section. All offsets MUST be multiples of four and placed in increasing order. The first post-header byte is at offset 0. The offset array is considered to have a not explicitly encoded value of 0 as its zeroth entry.¶
The value associated with the ith tag begins at offset[i] and ends at offset[i+1]-1, with the exception of the last value which ends at the end of the message. Values may have zero length. All lengths and offsets are in bytes.¶
Tags MUST be listed in the same order as the offsets of their values and MUST also be sorted in ascending order by numeric value. A tag MUST NOT appear more than once in a header.¶
As described in Section 3, clients initiate time synchronization by sending requests containing a nonce to servers who send signed time responses in return. Roughtime packets can be sent between clients and servers either as UDP datagrams or via TCP streams. Servers SHOULD support the UDP transport mode and TCP mode.¶
A Roughtime packet MUST be formatted according to Figure 2 and as described here. The first field is a uint64 with the value 0x4d49544847554f52 ("ROUGHTIM" in ASCII). The second field is a uint32 and contains the length of the third field. The third and last field contains a Roughtime message as specified in Section 5.¶
0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 0x4d49544847554f52 (uint64) | | ("ROUGHTIM") | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Message length (uint32) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | . . . Roughtime message . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Roughtime request and response packets MUST be transmitted in a single datagram when the UDP transport mode is used. Setting the packet's don't fragment bit [RFC791] is OPTIONAL in IPv4 networks.¶
Multiple requests and responses can be exchanged over an established TCP connection. Clients MAY send multiple requests at once and servers MAY send responses out of order. The connection SHOULD be closed by the client when it has no more requests to send and has received all expected responses. Either side SHOULD close the connection in response to synchronization, format, implementation-defined timeouts, or other errors.¶
All requests and responses MUST contain the VER tag. It contains a list of one or more uint32 version numbers. The version of Roughtime specified by this memo has version number 1.¶
NOTE TO RFC EDITOR: remove this paragraph before publication. For testing drafts of this memo, a version number of 0x80000000 plus the draft number is used.¶
A request MUST contain the tags VER and NONC. It SHOULD include the tag SRV. Other tags SHOULD be ignored by the server. A future version of this protocol may mandate additional tags in the message and assign them semantic meaning.¶
The size of the request message SHOULD be at least 1024 bytes when the UDP transport mode is used. To attain this size the ZZZZ tag SHOULD be added to the message. Responding to requests shorter than 1024 bytes is OPTIONAL and servers MUST NOT send responses larger than the requests they are replying to.¶
In a request, the VER tag contains a list of versions. The VER tag MUST include at least one Roughtime version supported by the client. The client MUST ensure that the version numbers and tags included in the request are not incompatible with each other or the packet contents.¶
The version numbers MUST NOT repeat.¶
The value of the NONC tag is a 32 byte nonce. It SHOULD be generated in a manner indistinguishable from random. BCP 106 contains specific guidelines regarding this [RFC4086].¶
The server begins the request handling process with a set of long-term keys. It resolves which long-term key to use with the following procedure:¶
If the request contains a SRV tag, then the server looks up the long-term key indicated by the SRV value. If no such key exists, then the server MUST ignore the request.¶
If the request contains no SRV tag, but the server has just one long-term key, it SHOULD select that key. Otherwise, if the server has multiple long-term keys, then it MUST ignore the request.¶
A response MUST contain the tags SIG, VER, NONC, PATH, SREP, CERT, and INDX.¶
In general, a SIG tag value is a 64 byte Ed25519 signature [RFC8032] over a concatenation of a signature context ASCII string and the entire value of a tag. All context strings MUST include a terminating zero byte.¶
The SIG tag in the root of a response MUST be a signature over the SREP value using the public key contained in CERT. The context string MUST be "RoughTime v1 response signature".¶
In a response, the VER tag MUST contain a single version number. It SHOULD be one of the version numbers supplied by the client in its request. The server MUST ensure that the version number corresponds with the rest of the packet contents.¶
The PATH tag value MUST be a multiple of 32 bytes long and represent a path of 32 byte hash values in the Merkle tree used to generate the ROOT value as described in a Section 6.3. In the case where a response is prepared for a single request and the Merkle tree contains only the root node, the size of PATH MUST be zero.¶
The SREP tag contains a time response. Its value MUST be a Roughtime message with the tags ROOT, MIDP, and RADI.¶
The ROOT tag MUST contain a 32 byte value of a Merkle tree root as described in Section 6.3.¶
The MIDP tag value MUST be the timestamp of the moment of processing.¶
The RADI tag value MUST be a uint32 representing the server's estimate of the accuracy of MIDP in seconds. Servers MUST ensure that the true time is within (MIDP-RADI, MIDP+RADI) at the time they transmit the response message.¶
The value of the RADI tag MUST be at least 3 seconds. Otherwise leap seconds will impact the observed correctness of Roughtime servers.¶
The CERT tag contains a public-key certificate signed with the server's long-term key. Its value is a Roughtime message with the tags DELE and SIG, where SIG is a signature over the DELE value. The context string used to generate SIG MUST be "RoughTime v1 delegation signature--".¶
The DELE tag contains a delegated public-key certificate used by the server to sign the SREP tag. Its value is a Roughtime message with the tags MINT, MAXT, and PUBK. The purpose of the DELE tag is to enable separation of a long-term public key from keys on devices exposed to the public Internet.¶
The MINT tag is the minimum timestamp for which the key in PUBK is trusted to sign responses. MIDP MUST be more than or equal to MINT for a response to be considered valid.¶
The MAXT tag is the maximum timestamp for which the key in PUBK is trusted to sign responses. MIDP MUST be less than or equal to MAXT for a response to be considered valid.¶
The PUBK tag contains a temporary 32 byte Ed25519 public key which is used to sign the SREP tag.¶
The INDX tag value is a uint32 determining the position of NONC in the Merkle tree used to generate the ROOT value as described in Section 6.3.¶
A Merkle tree is a binary tree where the value of each non-leaf node is a hash value derived from its two children. The root of the tree is thus dependent on all leaf nodes.¶
In Roughtime, each leaf node in the Merkle tree represents the nonce in one request. Leaf nodes are indexed left to right, beginning with zero.¶
The values of all nodes are calculated from the leaf nodes and up towards the root node using the first 32 bytes of the output of the SHA-512 hash algorithm [RFC6234]. For leaf nodes, the byte 0x00 is prepended to the nonce before applying the hash function. For all other nodes, the byte 0x01 is concatenated with first the left and then the right child node value before applying the hash function.¶
The value of the Merkle tree's root node is included in the ROOT tag of the response.¶
The index of a request's nonce node is included in the INDX tag of the response.¶
The values of all sibling nodes in the path between a request's nonce node and the root node are stored in the PATH tag so that the client can reconstruct and validate the value in the ROOT tag using its nonce. These values are each 32 bytes and are stored one after the other with no additional padding or structure. The order in which they are stored is described in the next section.¶
This section describes how to compute the root hash of the Merkle tree
from the values in the tags PATH, INDX, and NONC. The bits of INDX are
ordered from least to most significant. H(x)
denotes the first 32
bytes of the SHA-512 hash digest of x and ||
denotes concatenation.¶
The algorithm maintains a current hash value. At initialization, hash
is set to H(0x00 || nonce)
. When no more entries remain in PATH, the
current hash is the hash of the Merkle tree. All remaining bits of
INDX MUST be zero at that time. Otherwise, let node be the next 32
bytes in PATH. If the current bit in INDX is 0 then hash = H(0x01 ||
node || hash)
, else hash = H(0x01 || hash || node)
.¶
PATH is thus the siblings from the leaf to the root.¶
A client MUST check the following properties when it receives a response. We assume the long-term server public key is known to the client through other means.¶
The signature in CERT was made with the long-term key of the server.¶
The DELE timestamps and the MIDP value are consistent.¶
The INDX and PATH values prove NONC was included in the Merkle tree with value ROOT using the algorithm in Section 6.3.1.¶
The signature of SREP in SIG validates with the public key in DELE.¶
A response that passes these checks is said to be valid. Validity of a response does not prove the time is correct, but merely that the server signed it, and thus promises that it began to compute the signature at a time in the interval (MIDP-RADI, MIDP+RADI).¶
We assume that there is a bound PHI on the frequency error in the clock on the machine. Given a measurement taken at a local time t, we know the true time is in (t-delta-sigma, t-delta+sigma). After d seconds have elapsed we know the true time is within (t-delta-sigma-dPHI, t-delta+sigma+dPHI). A simple and effective way to mix with NTP or PTP discipline of the clock is to trim the observed intervals in NTP to fit entirely within this window or reject measurements that fall to far outside. This assumes time has not been stepped. If the NTP process decides to step the time, it MUST use Roughtime to ensure the new truetime estimate that will be stepped to is consistent with the true time. Should this window become too large, another Roughtime measurement is called for. The definition of "too large" is implementation defined. Implementations MAY use other, more sophisticated means of adjusting the clock respecting Roughtime information. Other applications such as X.509 verification may wish to apply different rules.¶
Servers SHOULD send back a fraction of responses that are syntactically invalid or contain invalid signatures as well as incorrect times. Clients MUST properly reject such responses. Servers MUST NOT send back responses with incorrect times and valid signatures. Either signature MAY be invalid for this application.¶
To carry out a Roughtime measurement, a client must be equipped with a list of servers, a minimum of three of which are operational, not run by the same parties. It must also have a means of reporting to the provider of such a list, such as an OS vendor or software vendor, a failure report as described below.¶
The client randomly permutes three servers from the list, and
sequentially queries them. The first probe uses a NONC that is
randomly generated. The second query uses H(resp || rand)
where rand
is a random 32 byte value and resp is the entire response to the first
probe. The third query uses H(resp || rand)
for a different 32 byte
value. If the times reported are consistent with the causal ordering,
and the delay is within a system provided parameter, the measurement
succeeds. If they are not consistent, there has been malfeasance and
the client SHOULD store a report for evaluation, alert the operator,
and make another measurement.¶
A malfeasance report is a JSON [RFC8259] object with keys "nonces", containing an array of the rand values as base64-encoded [RFC4648] strings, and "responses", containing an array of the responses as base64-encoded strings.¶
Malfeasence reports MAY be transported by any means to the relevant vendor or server operator for discussion. A malfeasance report is cryptographic proof that the responses arrived in that order, and can be used to demonstrate that at least one server sent the wrong time. The venues for sharing such reports and what to do about them are outside the scope of this document.¶
Since the only supported signature scheme, Ed25519, is not quantum resistant, the Roughtime version described in this memo will not survive the advent of quantum computers.¶
Maintaining a list of trusted servers and adjudicating violations of the rules by servers is not discussed in this document and is essential for security. Roughtime clients MUST regularly update their view of which servers are trustworthy in order to benefit from the detection of misbehavior.¶
Validating timestamps made on different dates requires knowledge of leap seconds in order to calculate time intervals correctly.¶
Servers carry out a significant amount of computation in response to clients, and thus may experience vulnerability to denial of service attacks.¶
This protocol does not provide any confidentiality. Given the nature of timestamps, such impact is minor.¶
The compromise of a PUBK's private key, even past MAXT, is a problem as the private key can be used to sign invalid times that are in the range MINT to MAXT, and thus violate the good behavior guarantee of the server.¶
Servers MUST NOT send response packets larger than the request packets sent by clients, in order to prevent amplification attacks.¶
This protocol is designed to obscure all client identifiers. Servers necessarily have persistent long-term identities essential to enforcing correct behavior. Generating nonces in a nonrandom manner can cause leaks of private data or enable tracking of clients as they move between networks.¶
It is expected that clients identify a server by its long-term public key. In multi-tenancy environments, where multiple servers may be listening on the same IP or port space, the protocol is designed so that the client indicates which server it expects to respond. This is done with the SRV tag.¶
IANA is requested to allocate the following entry in the Service Name and Transport Protocol Port Number Registry:¶
Service Name: Roughtime Transport Protocol: tcp,udp Assignee: IESG <iesg@ietf.org> Contact: IETF Chair <chair@ietf.org> Description: Roughtime time synchronization Reference: [[this memo]] Port Number: [[TBD1]], selected by IANA from the User Port range¶
IANA is requested to create a new registry entitled "Roughtime Version Registry". Entries shall have the following fields:¶
Version ID (REQUIRED): a 32-bit unsigned integer¶
Version name (REQUIRED): A short text string naming the version being identified.¶
Reference (REQUIRED): A reference to a relevant specification document.¶
The policy for allocation of new entries SHOULD be: IETF Review.¶
The initial contents of this registry shall be as follows:¶
Version ID | Version name | Reference |
---|---|---|
0x0 | Reserved | [[this memo]] |
0x1 | Roughtime version 1 | [[this memo]] |
0x2-0x7fffffff | Unassigned | |
0x80000000-0xffffffff | Reserved for Private | [[this memo]] |
or Experimental use |
IANA is requested to create a new registry entitled "Roughtime Tag Registry". Entries SHALL have the following fields:¶
Tag (REQUIRED): A 32-bit unsigned integer in hexadecimal format.¶
ASCII Representation (REQUIRED): The ASCII representation of the tag in accordance with Section 5.1.4 of this memo, if applicable.¶
Reference (REQUIRED): A reference to a relevant specification document.¶
The policy for allocation of new entries in this registry SHOULD be: Specification Required.¶
The initial contents of this registry SHALL be as follows:¶
Tag | ASCII Representation | Reference |
---|---|---|
0x00474953 | SIG | [[this memo]] |
0x00565253 | SRV | [[this memo]] |
0x00524556 | VER | [[this memo]] |
0x434e4f4e | NONC | [[this memo]] |
0x454c4544 | DELE | [[this memo]] |
0x48544150 | PATH | [[this memo]] |
0x49444152 | RADI | [[this memo]] |
0x4b425550 | PUBK | [[this memo]] |
0x5044494d | MIDP | [[this memo]] |
0x50455253 | SREP | [[this memo]] |
0x544e494d | MINT | [[this memo]] |
0x544f4f52 | ROOT | [[this memo]] |
0x54524543 | CERT | [[this memo]] |
0x5458414d | MAXT | [[this memo]] |
0x58444e49 | INDX | [[this memo]] |
0x5a5a5a5a | ZZZZ | [[this memo]] |
Aanchal Malhotra and Adam Langley authored early drafts of this memo. Thomas Peterson corrected multiple nits. Peter Löthberg, Tal Mizrahi, Ragnar Sundblad, Kristof Teichel, and the other members of the NTP working group contributed comments and suggestions.¶