From 24cc6b0db8ef9c718925d14b329f21938e5d2b1b Mon Sep 17 00:00:00 2001 From: Rasmus Dahlberg Date: Tue, 20 Apr 2021 12:28:28 +0200 Subject: started on our in-progress (re)design documents --- doc/api.md | 247 ++++++++++++++++++++ doc/design.md | 32 +++ doc/formats.md | 160 ------------- doc/schema/consistency_proof.schema.json | 30 +++ doc/schema/example/consistency_proof.json | 7 + doc/schema/example/inclusion_proof.json | 7 + doc/schema/example/leaves.json | 14 ++ doc/schema/example/sth.json | 11 + doc/schema/inclusion_proof.schema.json | 30 +++ doc/schema/leaves.schema.json | 38 +++ doc/schema/sth.schema.json | 50 ++++ doc/sketch.md | 372 ------------------------------ 12 files changed, 466 insertions(+), 532 deletions(-) create mode 100644 doc/api.md create mode 100644 doc/design.md delete mode 100644 doc/formats.md create mode 100644 doc/schema/consistency_proof.schema.json create mode 100644 doc/schema/example/consistency_proof.json create mode 100644 doc/schema/example/inclusion_proof.json create mode 100644 doc/schema/example/leaves.json create mode 100644 doc/schema/example/sth.json create mode 100644 doc/schema/inclusion_proof.schema.json create mode 100644 doc/schema/leaves.schema.json create mode 100644 doc/schema/sth.schema.json delete mode 100644 doc/sketch.md (limited to 'doc') diff --git a/doc/api.md b/doc/api.md new file mode 100644 index 0000000..760663b --- /dev/null +++ b/doc/api.md @@ -0,0 +1,247 @@ +# System Transparency Logging: API v0 +This document describes details of the System Transparency logging API, +version 0. The broader picture is not explained here. We assume that you have +read the System Transparency design document. It can be found [here](https://github.com/system-transparency/stfe/blob/design/doc/design.md). + +**Warning.** +This is a work-in-progress document that may be moved or modified. + +## Overview +The log implements an HTTP(S) API: +- Requests that add data to the log use the HTTP POST method. The HTTP content +type is `application/x-www-form-urlencoded`. The posted data are key-value +pairs. Binary data must be base64-encoded. +- Requests that retrieve data from the log use the HTTP GET method. The HTTP +content type is `application/x-www-form-urlencoded`. Input parameters are +key-value pairs. +- Responses are JSON objects. The HTTP content type is `application/json`. +- Error messages are human-readable strings. The HTTP content type is +`text/plain`. + +We decided to use these web formats for requests and responses because the log +is running as an HTTP(S) service. In other words, anyone that interacts with +the log is most likely using these formats already. The other benefit is that +all requests and responses are human-readable. This makes it easier to +understand the protocol, troubleshoot issues, and copy-paste. We favored +compatibility and understandability over a wire-efficient format. + +Note that we are not using JSON for signed and/or logged data. In other words, +a submitter that wishes to distribute log responses to their user base in a +different format may do so. The forced (de)serialization parser on _end-users_ +is a small subset of Trunnel. Trunnel is an "idiot-proof" wire-format that the +Tor project uses. + +## Primitives +### Cryptography +The log uses the same Merkle tree hash strategy as [RFC 6962, §2](https://tools.ietf.org/html/rfc6962#section-2). +The hash functions must be [SHA256](https://csrc.nist.gov/csrc/media/publications/fips/180/4/final/documents/fips180-4-draft-aug2014.pdf). +The log must sign tree heads using [Ed25519](https://tools.ietf.org/html/rfc8032). +The log's witnesses must also sign tree heads using Ed25519. + +All other parts that are not Merkle tree related also use SHA256 as the hash +function. Using more than one hash function would increases the overall attack +surface: two hash functions must be collision resistant instead of one. + +We recommend that submitters sign using Ed25519. We also support RSA with +[deterministic](https://tools.ietf.org/html/rfc8017#section-8.2) +or [probabilistic](https://tools.ietf.org/html/rfc8017#section-8.1) +padding. Supporting RSA is suboptimal, but excluding it would make the log +useless for many possible adopters. + +### Serialization +We use the [Trunnel](https://gitweb.torproject.org/trunnel.git) [description language](https://www.seul.org/~nickm/trunnel-manual.html) +to define (de)serialization of data structures that need to be signed or +inserted into the Merkle tree. Trunnel is more expressive than the +[SSH wire format](https://tools.ietf.org/html/rfc4251#section-5). +It is about as expressive as the [TLS presentation language](https://tools.ietf.org/html/rfc8446#section-3). +A notable difference is that Trunnel supports integer constraints. The Trunnel +language is also readable by humans _and_ machines. "Obviously correct code" +can be generated in C and Go. + +A fair summary of our Trunnel usage is as follows. + +All integers are 64-bit, unsigned, and in network byte order. A fixed-size byte +array is put into the serialization buffer in-order, starting from the first +byte. These basic types are concatenated to form a collection. You should not +need a general-purpose Trunnel (de)serialization parser to work with this +format. If you have one, you may use it though. The main point of using +Trunnel is that it makes a simple format explicit and unambiguous. + +TODO: URL-encode _or_ JSON? I think we should only need one. Always doing HTTP +POST would also ensure that input parameters don't show up in web server logs. + +#### Merkle tree head +Tree heads are signed by the log and its witnesses. It contains a timestamp, a +tree size, and a root hash. The timestamp is included so that monitors can +ensure _liveliness_. It is the time since the UNIX epoch (January 1, 1970 +00:00:00 UTC) in milliseconds. The tree size specifies the current number of +leaves. The root hash fixes the structure and content of the Merkle tree. + +``` +struct tree_head { + u64 timestamp; + u64 tree_size; + u8 root_hash[32]; +}; +``` + +The serialized tree head must be signed using Ed25519. A witness must only sign +the log's tree head if it is consistent with prior history and the timestamp is +roughly correct. A timestamp is roughly correct if it is not backdated or +future-dated more than 12 hours. + +#### Merkle tree leaf +The log supports a single leaf type. It contains a checksum, a signature +scheme, a signature that the submitter computed over that checksum, and the hash +of the public verification key that can be used to verify the signature. + +``` +const ALG_ED25519 = 1; // RFC 8032 +const ALG_RSASSA_PKCS1_V1_5 = 2; // RFC 8017 +const ALG_RSASSA_PSS = 3; // RFC 8017 + +struct tree_leaf { + u8 checksum[32]; + u64 signature_scheme IN [ + ALG_ED25519, + ALG_RSASSA_PKCS1_V1_5, + ALG_RSASSA_PSS, + ]; + union signature[signature_scheme] { + ALG_ED25519: u8 ed25519[32]; + default: u8 rsa[512]; + } + u8 key_hash[32]; +} +``` + +A key-hash is included in the leaf so that it can be attributed to the signing +entity. A hash, rather than the full public verification key, is used to force +the verifier to locate the appropriate key and make an explicit trust decision. + +## Public endpoints +Every log has a base URL that identifies it uniquely. The only constraint is +that it must be a valid HTTP(S) URL that can have the `/st/v0/` suffix +appended. For example, a complete endpoint URL could be +`https://log.example.com/2021/st/v0/get-signed-tree-head`. + +### get-signed-tree-head +``` +GET /st/v0/get-signed-tree-head +``` + +Input key-value pairs: +- `type`: either the string "latest", "stable", or "cosigned". + - "latest": ask for the most recent signed tree head. + - "stable": ask for a recent signed tree head that is fixed for some period + of time. + - "cosigned": ask for a recent cosigned tree head. + +Output: +- On success: status 200 OK and a signed tree head. The response body is +defined by the following [schema](https://github.com/system-transparency/stfe/blob/design/doc/schema/sth.schema.json). +- On failure: a different status code and a human-readable error message. + +### get-proof-by-hash +``` +POST /st/v0/get-proof-by-hash +``` + +Input key-value pairs: +- `leaf_hash`: a base64-encoded leaf hash that identifies which `tree_leaf` the +log should prove inclusion for. The leaf hash is computed using the RFC 6962 +hashing strategy. In other words, `H(0x00 | tree_leaf)`. +- `tree_size`: the tree size of a tree head that the proof should be based on. + +Output: +- On success: status 200 OK and an inclusion proof. The response body is +defined by the following [schema](https://github.com/system-transparency/stfe/blob/design/doc/schema/inclusion_proof.schema.json). +- On failure: a different status code and a human-readable error message. + +### get-consistency-proof +``` +POST /st/v0/get-consistency-proof +``` + +Input key-value pairs: +- `new_size`: the tree size of a newer tree head. +- `old_size`: the tree size of an older tree head that the log should prove is +consistent with the newer tree head. + +Output: +- On success: status 200 OK and a consistency proof. The response body is +defined by the following [schema](https://github.com/system-transparency/stfe/blob/design/doc/schema/consistency_proof.schema.json). +- On failure: a different status code and a human-readable error message. + +### get-leaves +``` +POST /st/v0/get-leaves +``` + +Input key-value pairs: +- `start_size`: zero-based index of the first leaf to retrieve. +- `end_size`: index of the last leaf to retrieve. + +Output: +- On success: status 200 OK and a list of leaves. The response body is +defined by the following [schema](https://github.com/system-transparency/stfe/blob/design/doc/schema/leaves.schema.json). +- On failure: a different status code and a human-readable error message. + +The log may truncate the list of returned leaves. However, it must not be an +empty list on success. + +### add-leaf +``` +POST /st/v0/add-leaf +``` + +Input key-value pairs: +- `leaf_checksum`: the checksum that the submitter wants to log in base64. +- `signature_scheme`: the signature scheme that the submitter wants to use. +- `tree_leaf_signature`: the submitter's `tree_leaf` signature in base64. +- `verification_key`: the submitter's public verification key. It is serialized +as described in the corresponding RFC, then base64-encoded. +- `domain_hint`: a domain name that indicates where the public verification-key +hash can be downloaded in base64. Supported methods: DNS and HTTPS +(TODO: docdoc). + +Output: +- On success: HTTP 200. The log will _try_ to incorporate the submitted leaf +into its Merkle tree. +- On failure: a different status code and a human-readable error message. + +The submitted entry will not be accepted if the signature is invalid or if the +downloaded verification-key hash does not match. The submitted entry may also +not be accepted if the second-level domain name exceeded its rate limit. By +coupling every add-leaf request with a second-level domain, it becomes more +difficult to spam the log. You would need an excessive number of domain names. +This becomes costly if free domain names are rejected. + +The log does not publish domain-name to key bindings because key management is +more complex than that. + +Public logging should not be assumed until an inclusion proof is available. An +inclusion proof should not be relied upon unless it leads up to a trustworthy +signed tree head. Witness cosigning can make a tree head trustworthy. + +TODO: the log may allow no `domain_hint`? Especially useful for v0 testing. + +### add-cosignature +``` +POST /st/v0/add-cosignature +``` + +Input key-value pairs: +- `signature`: a base64-encoded signature over a `tree_head` that is fixed for +some period of time. The cosigning witness retrieves the tree head using the +`get-signed-tree-head` endpoint with the "stable" type. +- `key_hash`: a base64-encoded hash of the public verification key that can be +used to verify the signature. + +Output: +- HTTP status 200 OK on success. Otherwise a different status code and a +human-readable error message. + +The key-hash can be used to identify which witness signed the log's tree head. +A key-hash, rather than the full verification key, is used to force the verifier +to locate the appropriate key and make an explicit trust decision. diff --git a/doc/design.md b/doc/design.md new file mode 100644 index 0000000..f966d03 --- /dev/null +++ b/doc/design.md @@ -0,0 +1,32 @@ +# System Transparency Logging: Design v0 +We propose System Transparency logging. It is similar to Certificate +Transparency, expect that cryptographically signed checksums are logged as +opposed to X.509 certificates. Publicly logging signed checksums allow anyone +to discover which keys signed what. As such, malicious and unintended key-usage +can be _discovered_. We present our design and discuss how two possible +use-cases influenced it: binary transparency and reproducible builds. + +**Target audience.** +You are most likely interested in transparency logs or supply-chain security. + +**Preliminaries.** +You have basic understanding of cryptographic primitives like digital +signatures, hash functions, and Merkle trees. You roughly know what problem +Certificate Transparency solves and how. You may never have heard the term +_gossip-audit model_, or know how it is related to trust assumptions and +detectability properties. + +**Warning.** +This is a work-in-progress document that may be moved or modified. + +## Introduction +Transparency logs make it possible to detect unwanted events. For example, + are there any (mis-)issued TLS certificates [\[CT\]](https://tools.ietf.org/html/rfc6962), + did you get a different Go module than everyone else [\[ChecksumDB\]](https://go.googlesource.com/proposal/+/master/design/25530-sumdb.md), + or is someone running unexpected commands on your server [\[AuditLog\]](https://transparency.dev/application/reliably-log-all-actions-performed-on-your-servers/). +System Transparency logging makes signed checksums transparent. The goal is to +_detect_ unwanted key-usage without making assumptions about the signed data. + +## Threat model and (non-)goals + +## Design diff --git a/doc/formats.md b/doc/formats.md deleted file mode 100644 index bffd05f..0000000 --- a/doc/formats.md +++ /dev/null @@ -1,160 +0,0 @@ -# Formats -This document defines data structures and data formats. - -## Overview -Here we give an overview of our presentation language / serialization rules. - -All integers are represented by 64-bit unsigned integers in network byte order. - -Variable length lists have an integer specifying its length. Then each list -item is enumerated. - -TODO: fixme. - -## Items -Every item type start with a versioned format specifier. Protocol version 1 -uses format specifiers in the range 1--X. - -### Request data structures -Log endpoints that take input data use the following request data structures. - -#### `get_entries_v1` -``` -0 Format 8 16 24 -+----------+----------------+----------------+ -| 1 | Start Size | End Size | -+----------+----------------+----------------+ - uint64 uint64 uint64 -``` -- Format is always 1 for items of type `get_entries_v1`. -- Start size specifies the index of the first Merkle tree leaf to retrieve. -- End size specifies the index of the last Merkle tree leaf to retrieve. - -#### `get_proof_by_hash_v1` -``` -0 Format 8 16 48 -+----------+----------------+----------------+ -| 2 | Tree size | Leaf hash | -+----------+----------------+----------------+ - uint64 uint64 fixed byte array -``` -- Format is always 2 for items of type `get_proof_by_hash_v1`. -- Leaf hash is computed as described in [RFC 6962/bis, §2.1.1](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-35#section-2.1.1). -- Tree size specifies which Merkle tree root inclusion should be proven for. - -#### `get_consistency_proof_v1` -``` -0 Format 8 16 24 -+----------+----------------+----------------+ -| 3 | Old size | New size | -+----------+----------------+----------------+ - uint64 uint64 uint64 -``` -- Format is always 3 for items of type `get_consistency_proof_v1`. -- Old size specifies the tree size of an older Merkle tree head. -- New size specifies the tree size of a newer Merkle tree head. - -### Proof and log data structures -#### `inclusion_proof_v1` -``` - --zero or more node hashes--> -0 Format 8 48 56 64 72 72+Length -+----------+----------------+----------------+----------------+----------------+--------//--------+ -| 4 | Identifier | Tree size | Leaf index | Length | Node hashes | -+----------+----------------+----------------+----------------+----------------+--------//--------+ - uint64 ed25519_v1 uint64 uint64 uint64 list body -``` -- Format is always 4 for items of type `inclusion_proof_v1`. -- Identifier identifies the log uniquely as an `ed25519_v1` item. -- Tree size is the size of the Merkle tree that the proof is based on. -- Leaf index is a zero-based index of the log entry that the proof is based on. -- The remaining part is a list of node hashes. - - Length specifies the full byte size of the list. It must be `32 * m`, - where `m >= 0`. This means that an inclusion needs zero or more node - hashes to be well-formed. - - Node hash is a node hash in the Merkle tree that the proof is based on. - -Remark: the list of node hashes is generated and verified as in [RFC 6962/bis, -§2.1.3](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-35#section-2.1.3). - -#### `consistency_proof_v1` -``` - --zero or more node hashes--> -0 Format 8 48 56 64 72 72+Length -+----------+----------------+----------------+----------------+----------------+--------//--------+ -| 5 | Identifier | Old size | New size | Length | Node hashes | -+----------+----------------+----------------+----------------+----------------+--------//--------+ - uint64 ed25519_v1 uint64 uint64 uint64 list body -``` -- Format is always 5 for items of type `consistency_proof_v1`. -- Identifier identifies the log uniquely as an `ed25519_v1` item. -- Old size is the tree size of the older Merkle tree. -- New size is the tree size of the newer Merkle tree. -- The remaining part is a list of node hashes. - - Length specifies the full byte size of the list. It must be `32 * m`, - where `m >= 0`. This means that a consistenty proof needs zero or more node - hashes to be well-formed. - - Node hash is a node hash from the older or the newer Merkle tree. - -Remark: the list of node hashes is generated and verified as in [RFC 6962/bis, -§2.1.4](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-35#section-2.1.4). - -#### `signed_tree_head_v1` -``` - ----one or more signature-identifier pairs-------> -0 Format 8 16 24 56 64 128 168 64+Length -+----------+----------------+----------------+----------------+----------------+----------------+----------------+--//--+ -| 6 | Timestamp | Tree size | Root hash | Length | Signature | Identifier | .... | -+----------+----------------+----------------+----------------+----------------+----------------+----------------+--//--+ - uint64 uint64 uint64 fixed byte array uint64 fixed byte array ed25519_v1 cont. list body -``` -- Format is always 6 for items of type `signed_tree_head_v1`. -- Timestamp is the time since the UNIX epoch (January 1, 1970 00:00:00 UTC) in -milliseconds. -- Tree size is the number of leaves in the current Merkle tree. -- Root hash is the root hash of the current Merkle tree. -- The remaining part is a list of signature-identifier pairs. - - Length specifies the full byte size of the list. It must be `104 * m`, - where `m > 1`. This means that a signed tree head needs at least one - signature-identifier pair to be well-formed. - - Signature is an Ed25519 signature over bytes 0--56. The signature is - encodes as in [RFC 8032, §3.3](https://tools.ietf.org/html/rfc8032#section-3.3). - - Identifier identifies the signer uniquely as an `ed25519_v1` item. - -Remark: there may be multiple signature-identifier pairs if the log is cosigned. - -#### `signed_checksum32_ed25519_v1` -``` -0 Format 8 40 56 56+Length 120+Length 160+Length -+----------+----------------+----------------+-------//---------+----------------+--------//--------+ -| 7 | Checksum | Length | Identifier | Signature | Namespace | -+----------+----------------+----------------+-------//---------+----------------+--------//--------+ - uint64 fixed byte array uint64 byte array fixed byte array ed25519_v1 -``` -- Format is always 7 for items of type `signed_checksum32_ed25519_v1`. -- Checksum is a 32-byte checksum that represents a data item of opaque type. -- Length specified the full byte size of the following identifier. It must be -larger than zero and less than 128. -- Identifier identifies what the checksum represents. The aforementioned length -constraint means that the identifier cannot be omitted or exceed 128 bytes. -- Signature is an Ed25519 signature over bytes 0--56+Length. The signature is -encodes as in [RFC 8032, §3.3](https://tools.ietf.org/html/rfc8032#section-3.3). -- Namespace is an `ed25519_v1` item that identifies the signer uniquely. - -Remark: to keep this checksum entry as simple as possible it does not have a -variable length checksum or any agility with regards to the signing namespace. -This means that we need to have multiple leaf types that follow the pattern -`signed_checksum{32,64}_namespace_v1`. - -### Namespace data structures -#### `ed25519_v1` -``` -0 Format 8 40 -+----------+----------------+ -| 8 | public key | -+----------+----------------+ - uint64 fixed byte array -``` -- The format is always 8 for items of type `ed25519_v1`. -- The public Ed25519 verification key is always 32 bytes. See encoding in [RFC -8032, §3.2](https://tools.ietf.org/html/rfc8032#section-3.2). diff --git a/doc/schema/consistency_proof.schema.json b/doc/schema/consistency_proof.schema.json new file mode 100644 index 0000000..003f3c7 --- /dev/null +++ b/doc/schema/consistency_proof.schema.json @@ -0,0 +1,30 @@ +{ + "$schema": "https://json-schema.org/draft-07/schema#", + "title": "inclusion_proof", + "description": "JSON-formatted inclusion proof, version 0.", + + "type": "object", + "required": [ "new_size", "old_size", "consistency_proof" ], + "properties": { + "new_size": { + "description": "The tree size of the newer Merkle tree head.", + "type": "integer", + "minimum": 0 + }, + "old_size": { + "description": "The tree size of the older Merkle tree head.", + "type": "integer", + "minimum": 0 + }, + "consistency_proof": { + "description": "A list of base64-encoded node hashes that proves consistency", + "type": "array", + "items": { + "description": "A node hash in base64", + "type": "string", + "minLength": 44, + "maxLength": 44 + } + } + } +} diff --git a/doc/schema/example/consistency_proof.json b/doc/schema/example/consistency_proof.json new file mode 100644 index 0000000..0a323b7 --- /dev/null +++ b/doc/schema/example/consistency_proof.json @@ -0,0 +1,7 @@ +{ + "new_size": 2, + "old_size": 1, + "consistency_proof": [ + "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=" + ] +} diff --git a/doc/schema/example/inclusion_proof.json b/doc/schema/example/inclusion_proof.json new file mode 100644 index 0000000..d46d426 --- /dev/null +++ b/doc/schema/example/inclusion_proof.json @@ -0,0 +1,7 @@ +{ + "tree_size": 2, + "leaf_index": 0, + "inclusion_proof": [ + "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=" + ] +} diff --git a/doc/schema/example/leaves.json b/doc/schema/example/leaves.json new file mode 100644 index 0000000..1eed05d --- /dev/null +++ b/doc/schema/example/leaves.json @@ -0,0 +1,14 @@ +[ + { + "checksum": "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=", + "signature_scheme": 1, + "signature": "CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC=", + "key_hash": "DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD=" + }, + { + "checksum": "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=", + "signature_scheme": 2, + "signature": "BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB=", + "key_hash": "CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC=" + } +] diff --git a/doc/schema/example/sth.json b/doc/schema/example/sth.json new file mode 100644 index 0000000..ec3ad11 --- /dev/null +++ b/doc/schema/example/sth.json @@ -0,0 +1,11 @@ +{ + "timestamp": 0, + "tree_size": 0, + "root_hash": "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=", + "signatures": [ + { + "key_hash": "BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB=", + "signature": "CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC=" + } + ] +} diff --git a/doc/schema/inclusion_proof.schema.json b/doc/schema/inclusion_proof.schema.json new file mode 100644 index 0000000..3309d37 --- /dev/null +++ b/doc/schema/inclusion_proof.schema.json @@ -0,0 +1,30 @@ +{ + "$schema": "https://json-schema.org/draft-07/schema#", + "title": "inclusion_proof", + "description": "JSON-formatted inclusion proof, version 0.", + + "type": "object", + "required": [ "tree_size", "leaf_index", "inclusion_proof" ], + "properties": { + "tree_size": { + "description": "The Merkle tree size that the inclusion proof is based on.", + "type": "integer", + "minimum": 0 + }, + "leaf_index": { + "description": "The zero-based index of the leaf that the inclusion proof is for.", + "type": "integer", + "minimum": 0 + }, + "inclusion_proof": { + "description": "A list of base64-encoded node hashes that proves inclusion", + "type": "array", + "items": { + "description": "A node hash in base64", + "type": "string", + "minLength": 44, + "maxLength": 44 + } + } + } +} diff --git a/doc/schema/leaves.schema.json b/doc/schema/leaves.schema.json new file mode 100644 index 0000000..74d7454 --- /dev/null +++ b/doc/schema/leaves.schema.json @@ -0,0 +1,38 @@ +{ + "$schema": "https://json-schema.org/draft-07/schema#", + "title": "list of tree_leaf", + "description": "JSON-formatted tree leaf list, version 0.", + + "type": "array", + "description": "A list Merkle tree leaves", + "items": { + "type": "object", + "required": [ "checksum", "signature_scheme", "signature", "key_hash" ], + "properties": { + "checksum": { + "description": "A cryptographic hash that is computed over some data of opaque type. The result is base64-encoded.", + "type": "string", + "minLength": 44, + "maxLength": 44 + }, + "signature_scheme": { + "description": "An integer that identifies the signature scheme used by the submitter. See API documentation.", + "type": "integer", + "enum": [ 1, 2, 3 ] + }, + "signature": { + "description": "The submitter's signature over the checksum in base64", + "type": "string", + "minLength": 44, + "maxLength": 684 + }, + "key_hash": { + "description": "A public verification-key hash that identifies the signer.", + "type": "string", + "minLength": 44, + "maxLength": 44 + } + } + }, + "minItems": 1 +} diff --git a/doc/schema/sth.schema.json b/doc/schema/sth.schema.json new file mode 100644 index 0000000..86de2d3 --- /dev/null +++ b/doc/schema/sth.schema.json @@ -0,0 +1,50 @@ +{ + "$schema": "https://json-schema.org/draft-07/schema#", + "title": "signed_tree_head_v0", + "description": "JSON-formatted signed tree head, version 0.", + + "type": "object", + "required": [ "timestamp", "tree_size", "root_hash", "signatures" ], + "properties": { + "timestamp": { + "description": "The number of milliseconds since the UNIX epoch (January 1, 1970 00:00:00 UTC).", + "type": "integer", + "minimum": 0 + }, + "tree_size": { + "description": "The number of entries that are stored in the log's Merkle tree.", + "type": "integer", + "minimum": 0 + }, + "root_hash": { + "description": "The log's Merkle tree root hash in base64.", + "type": "string", + "minLength": 44, + "maxLength": 44 + }, + "signatures": { + "description": "A list of signer-signature pairs.", + "type": "array", + "items": { + "description": "A signer-signature pair.", + "type": "object", + "required": [ "key_hash", "signature" ], + "properties": { + "key_hash": { + "description": "A public verification-key hash that identifies the signer in base64.", + "type": "string", + "minLength": 44, + "maxLength": 44 + }, + "signature": { + "description": "The signer's signature over the log's tree_leaf structure in base64.", + "type": "string", + "minLength": 44, + "maxLength": 44 + } + } + }, + "minItems": 1 + } + } +} diff --git a/doc/sketch.md b/doc/sketch.md deleted file mode 100644 index 31964e0..0000000 --- a/doc/sketch.md +++ /dev/null @@ -1,372 +0,0 @@ -# System Transparency Logging -This document provides a sketch of System Transparency (ST) logging. The basic -idea is to insert hashes of system artifacts into a public, append-only, and -tamper-evident transparency log, such that any enforcing client can be sure that -they see the same system artifacts as everyone else. A system artifact could -be a browser update, an operating system image, a Debian package, or more -generally something that is opaque. - -We take inspiration from the Certificate Transparency Front-End -([CTFE](https://github.com/google/certificate-transparency-go/tree/master/trillian/ctfe)) -that implements [RFC 6962](https://tools.ietf.org/html/rfc6962) for -[Trillian](https://transparency.dev). - -## Log parameters -An ST log is defined by the following parameters: -- `log_identifier`: a `Namespace` of type `ed25519_v1` that defines the log's -signing algorithm and public verification key. -- `supported_namespaces`: a list of namespace types that the log supports. -Entities must use a supported namespace type when posting signed data to the -log. -- `base_url`: prefix used by clients that contact the log, e.g., -example.com:1234/log. -- `final_cosigned_tree_head`: an `StItem` of type `cosigned_tree_head_v*`. Not -set until the log is turned into read-only mode in preparation of a shutdown. - -ST logs use the same hash strategy as described in RFC 6962: SHA256 with `0x00` -as leaf node prefix and `0x01` as interior node prefix. - -In contrast to Certificate Transparency (CT) **there is no Maximum Merge Delay -(MMD)**. New entries are merged into the log as soon as possible, and no client -should trust that something is logged until an inclusion proof can be provided -that references a trustworthy STH. Therefore, **there are no "promises" of -public logging** as in CT. - -To produce trustworthy STHs a simple form of [witness -cosigning](https://arxiv.org/pdf/1503.08768.pdf) is built into the log. -Witnesses poll the log for the next stable STH, and verify that it is consistent -before posting a cosignature that can then be served by the log. - -## Acceptance criteria and scope -A log should accept a leaf submission if it is: -- Well-formed, see data structure definitions below. -- Digitally signed by a registered namespace. - -Rate limits may be applied per namespace to combat spam. Namespaces may also be -used by clients to determine which entries belong to who. It is up to the -submitters to communicate trusted namespaces to their own clients. In other -words, there are no mappings from namespaces to identities built into the log. -There is also no revocation of namespaces: **we facilitate _detection_ of -compromised signing keys by making artifact hashes public, which is not to be -confused with _prevention_ or even _recovery_ after detection**. - -## Data structure definitions -Data structures are defined and serialized using the presentation language in -[RFC 5246, §4](https://tools.ietf.org/html/rfc5246). A definition of the log's -Merkle tree can be found in [RFC 6962, -§2](https://tools.ietf.org/html/rfc6962#section-2). - -### Namespace -A _namespace_ is a versioned data structure that contains a public verification -key (or fingerprint), as well as enough information to determine its format, -signing, and verification operations. Namespaces are used as identifiers, both -for the log itself and the parties that submit artifact hashes and cosignatures. - -``` -enum { - reserved(0), - ed25519_v1(1), - (2^16-1) -} NamespaceFormat; - -struct { - NamespaceFormat format; - select (format) { - case ed25519_v1: Ed25519V1; - } message; -} Namespace; -``` - -Our namespace format is inspired by Keybase's -[key-id](https://keybase.io/docs/api/1.0/kid). - -#### Ed25519V1 -At this time the only supported namespace type is based on Ed25519. The -namespace field contains the full verification key. Signing operations and -serialized formats are defined by [RFC -8032](https://tools.ietf.org/html/rfc8032). -``` -struct { - opaque namespace[32]; // public verification key -} Ed25519V1; -``` - -### `StItem` -A general-purpose `TransItem` is defined in [RFC 6962/bis, -§4.5](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-34#section-4.5). -We define our own `TransItem`, but name it `StItem` to emphasize that they are -not the same. - -``` -enum { - reserved(0), - signed_tree_head_v1(1), - cosigned_tree_head_v1(2), - consistency_proof_v1(3), - inclusion_proof_v1(4), - signed_checksum_v1(5), // leaf type - (2^16-1) -} StFormat; - -struct { - StFormat format; - select (format) { - case signed_tree_head_v1: SignedTreeHeadV1; - case cosigned_tree_head_v1: CosignedTreeHeadV1; - case consistency_proof_v1: ConsistencyProofV1; - case inclusion_proof_v1: InclusionProofV1; - case signed_checksum_v1: SignedChecksumV1; - } message; -} StItem; - -struct { - StItem items<0..2^32-1>; -} StItemList; -``` - -#### `signed_tree_head_v1` -We use the same tree head definition as in [RFC 6962/bis, -§4.9](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-34#section-4.9). -The resulting _signed_ tree head is packaged differently: a namespace is used as -log identifier, and it is communicated in a `SignatureV1` structure. -``` -struct { - TreeHeadV1 tree_head; - SignatureV1 signature; -} SignedTreeHeadV1; - -struct { - uint64 timestamp; - uint64 tree_size; - NodeHash root_hash; - Extension extensions<0..2^16-1>; -} TreeHeadV1; -opaque NodeHash<32..2^8-1>; - -struct { - Namespace namespace; - opaque signature<1..2^16-1>; -} SignatureV1; -``` - -#### `cosigned_tree_head_v1` -Transparency logs were designed to be cryptographically verifiable in the -presence of a gossip-audit model that ensures everyone observes _the same -cryptographically verifiable log_. The gossip-audit model is largely undefined -in today's existing transparency logging ecosystems, which means that the logs -must be trusted to play by the rules. We wanted to avoid that outcome in our -ecosystem. Therefore, a gossip-audit model is built into the log. - -The basic idea is that an STH should only be considered valid if it is cosigned -by a number of witnesses that verify the append-only property. Which witnesses -to trust and under what circumstances is defined by a client-side _witness -cosigning policy_. For example, - "require no witness cosigning", - "must have at least `k` signatures from witnesses A...J", and - "must have at least `k` signatures from witnesses A...J where one is from - witness B". - -Witness cosigning policies are beyond the scope of this specification. - -A cosigned STH is composed of an STH and a list of cosignatures. A cosignature -must cover the serialized STH as an `StItem`, and be produced with a witness -namespace of type `ed25519_v1`. - -``` -struct { - SignedTreeHeadV1 signed_tree_head; - SignatureV1 cosignatures<0..2^32-1>; // vector of cosignatures -} CosignedTreeHeadV1; -``` - -#### `consistency_proof_v1` -For the most part we use the same consistency proof definition as in [RFC -6962/bis, -§4.11](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-34#section-4.11). -There are two modifications: our log identifier is a namespace rather than an -[OID](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-34#section-4.4), -and a consistency proof may be empty. - -``` -struct { - Namespace log_id; - uint64 tree_size_1; - uint64 tree_size_2; - NodeHash consistency_path<0..2^16-1>; -} ConsistencyProofV1; -``` - -#### `inclusion_proof_v1` -For the most part we use the same inclusion proof definition as in [RFC -6962/bis, -§4.12](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-34#section-4.12). -There are two modifications: our log identifier is a namespace rather than an -[OID](https://tools.ietf.org/html/draft-ietf-trans-rfc6962-bis-34#section-4.4), -and an inclusion proof may be empty. -``` -struct { - Namespace log_id; - uint64 tree_size; - uint64 leaf_index; - NodeHash inclusion_path<0..2^16-1>; -} InclusionProofV1; -``` - -#### `signed_checksum_v1` -A checksum entry contains a package identifier like `foobar-1.2.3` and an -artifact hash. It is then signed so that clients can distinguish artifact -hashes from two different software publishers A and B. For example, the -`signed_checksum_v1` type can help [enforce public binary logging before -accepting a new software -update](https://wiki.mozilla.org/Security/Binary_Transparency). - -``` -struct { - ChecksumV1 data; - SignatureV1 signature; -} SignedChecksumV1; - -struct { - opaque identifier<1..128>; - opaque checksum<1..64>; -} ChecksumV1; -``` - -It is assumed that clients know how to find the real artifact source (if not -already at hand), such that the logged hash can be recomputed and compared for -equality. The log is not aware of how artifact hashes are computed, which means -that it is up to the submitters to define hash functions, data formats, and -such. - -## Public endpoints -Clients talk to the log using HTTP(S). Successfully processed requests are -responded to with HTTP status code `200 OK`, and any returned data is -serialized. Endpoints without input parameters use HTTP GET requests. -Endpoints that have input parameters HTTP POST a TLS-serialized data structure. -The HTTP content type `application/octet-stream` is used when sending data. - -### add-entry -``` -POST https:///st/v1/add-entry -``` - -Input: -- An `StItem` of type `signed_checksum_v1`. - -No output. - -### add-cosignature -``` -POST https:///st/v1/add-cosignature -``` - -Input: -- An `StItem` of type `cosigned_tree_head_v1`. The list of cosignatures must -be of length one, the witness signature must cover the item's STH, and that STH -must additionally match the log's stable STH that is currently being cosigned. - -No output. - -### get-latest-sth -``` -GET https:///st/v1/get-latest-sth -``` - -No input. - -Output: -- An `StItem` of type `signed_tree_head_v1` that corresponds to the most -recent STH. - -### get-stable-sth -``` -GET https:///st/v1/get-stable-sth -``` - -No input. - -Output: -- An `StItem` of type `signed_tree_head_v1` that corresponds to a stable STH -that witnesses should cosign. The same STH is returned for a period of time. - -### get-cosigned-sth -``` -GET https:///st/v1/get-cosigned-sth -``` - -No input. - -Output: -- An `StItem` of type `cosigned_tree_head_v1` that corresponds to the most -recent cosigned STH. - -### get-proof-by-hash -``` -POST https:///st/v1/get-proof-by-hash -``` - -Input: -``` -struct { - opaque hash[32]; // leaf hash - uint64 tree_size; // tree size that the proof should be based on -} GetProofByHashV1; -``` - -Output: -- An `StItem` of type `inclusion_proof_v1`. - -### get-consistency-proof -``` -POST https:///st/v1/get-consistency-proof -``` - -Input: -``` -struct { - uint64 first; // first tree size that the proof should be based on - uint64 second; // second tree size that the proof should be based on -} GetConsistencyProofV1; -``` - -Output: -- An `StItem` of type `consistency_proof_v1`. - -### get-entries -``` -POST https:///st/v1/get-entries -``` - -Input: -``` -struct { - uint64 start; // 0-based index of first entry to retrieve - uint64 end; // 0-based index of last entry to retrieve in decimal. -} GetEntriesV1; -``` - -Output: -- An `StItem` list where each entry is of type `signed_checksum_v1`. The first -`StItem` corresponds to the start index, the second one to `start+1`, etc. The -log may return fewer entries than requested. - -# Appendix A -In the future other namespace types might be supported. For example, we could -add [RSASSA-PKCS1-v1_5](https://tools.ietf.org/html/rfc3447#section-8.2) as -follows: -1. Add `rsa_v1` format and RSAV1 namespace. This is what we would register on -the server-side such that the server knows the namespace and complete key. -``` -struct { - opaque namespace<32>; // key fingerprint - // + some encoding of public key -} RSAV1; -``` -2. Add `rsassa_pkcs1_5_v1` format and `RSASSAPKCS1_5_v1`. This is what the -submitter would use to communicate namespace and RSA signature mode. -``` -struct { - opaque namespace<32>; // key fingerprint - // + necessary parameters, e.g., SHA256 as hash function -} RSASSAPKCS1_5V1; -``` -- cgit v1.2.3