github.com/aakash4dev/cometbft@v0.38.2/spec/README.md (about) 1 --- 2 order: 1 3 title: Overview 4 parent: 5 title: Spec 6 order: 7 7 --- 8 9 # CometBFT Spec 10 11 This is a markdown specification of CometBFT. 12 It defines the base data structures, how they are validated, 13 and how they are communicated over the network. 14 15 If you find discrepancies between the spec and the code that 16 do not have an associated issue or pull request on github, 17 please submit them to our [bug bounty](https://cometbft.com/security)! 18 19 ## Contents 20 21 - [Overview](#overview) 22 23 ### Data Structures 24 25 - [Encoding and Digests](./core/encoding.md) 26 - [Blockchain](./core/data_structures.md) 27 - [State](./core/state.md) 28 29 ### Consensus Protocol 30 31 - [Consensus Algorithm](./consensus/consensus.md) 32 - [Creating a proposal](./consensus/creating-proposal.md) 33 - [Time](./consensus/bft-time.md) 34 - [Light-Client](./consensus/light-client/README.md) 35 36 ### P2P and Network Protocols 37 38 - [The Base P2P Layer](./p2p/legacy-docs/node.md): multiplex the protocols ("reactors") on authenticated and encrypted TCP connections 39 - [Peer Exchange (PEX)](./p2p/legacy-docs/messages/pex.md): gossip known peer addresses so peers can find each other 40 - [Block Sync](./p2p/legacy-docs/messages/block-sync.md): gossip blocks so peers can catch up quickly 41 - [Consensus](./p2p/legacy-docs/messages/consensus.md): gossip votes and block parts so new blocks can be committed 42 - [Mempool](./p2p/legacy-docs/messages/mempool.md): gossip transactions so they get included in blocks 43 - [Evidence](./p2p/legacy-docs/messages/evidence.md): sending invalid evidence will stop the peer 44 45 ### RPC 46 47 - [RPC SPEC](./rpc/README.md): Specification of the CometBFT remote procedure call interface. 48 49 ### Software 50 51 - [ABCI](./abci/README.md): Details about interactions between the 52 application and consensus engine over ABCI 53 - [Write-Ahead Log](./consensus/wal.md): Details about how the consensus 54 engine preserves data and recovers from crash failures 55 56 ## Overview 57 58 CometBFT provides Byzantine Fault Tolerant State Machine Replication using 59 hash-linked batches of transactions. Such transaction batches are called "blocks". 60 Hence, CometBFT defines a "blockchain". 61 62 Each block in CometBFT has a unique index - its Height. 63 Heights in the blockchain are monotonic. 64 Each block is committed by a known set of weighted Validators. 65 Membership and weighting within this validator set may change over time. 66 CometBFT guarantees the safety and liveness of the blockchain 67 as long as less than 1/3 of the total weight of the Validator set 68 is malicious or faulty. 69 70 A commit in CometBFT is a set of signed messages from more than 2/3 of 71 the total weight of the current Validator set. Validators take turns proposing 72 blocks and voting on them. Once enough votes are received, the block is considered 73 committed. These votes are included in the _next_ block as proof that the previous block 74 was committed - they cannot be included in the current block, as that block has already been 75 created. 76 77 Once a block is committed, it can be executed against an application. 78 The application returns results for each of the transactions in the block. 79 The application can also return changes to be made to the validator set, 80 as well as a cryptographic digest of its latest state. 81 82 CometBFT is designed to enable efficient verification and authentication 83 of the latest state of the blockchain. To achieve this, it embeds 84 cryptographic commitments to certain information in the block "header". 85 This information includes the contents of the block (eg. the transactions), 86 the validator set committing the block, as well as the various results returned by the application. 87 Note, however, that block execution only occurs _after_ a block is committed. 88 Thus, application results can only be included in the _next_ block. 89 90 Also note that information like the transaction results and the validator set are never 91 directly included in the block - only their cryptographic digests (Merkle roots) are. 92 Hence, verification of a block requires a separate data structure to store this information. 93 We call this the `State`. Block verification also requires access to the previous block.