diff --git a/Makefile b/Makefile index 1fb3eacb..ffc72c46 100644 --- a/Makefile +++ b/Makefile @@ -23,11 +23,14 @@ check: check_tools get_vendor_deps build: CGO_ENABLED=0 go build $(BUILD_FLAGS) -tags $(BUILD_TAGS) -o build/tendermint ./cmd/tendermint/ +build_c: + CGO_ENABLED=1 go build $(BUILD_FLAGS) -tags "$(BUILD_TAGS) gcc" -o build/tendermint ./cmd/tendermint/ + build_race: CGO_ENABLED=0 go build -race $(BUILD_FLAGS) -tags $(BUILD_TAGS) -o build/tendermint ./cmd/tendermint install: - CGO_ENABLED=0 go install $(BUILD_FLAGS) -tags $(BUILD_TAGS) ./cmd/tendermint + CGO_ENABLED=0 go install $(BUILD_FLAGS) -tags $(BUILD_TAGS) ./cmd/tendermint ######################################## ### Protobuf diff --git a/docs/introduction/install.md b/docs/introduction/install.md index c7b83b03..f7d78aba 100644 --- a/docs/introduction/install.md +++ b/docs/introduction/install.md @@ -77,8 +77,13 @@ make install ## Compile with CLevelDB support -Install [LevelDB](https://github.com/google/leveldb) (minimum version is 1.7) -with snappy. Example for Ubuntu: +Install [LevelDB](https://github.com/google/leveldb) (minimum version is 1.7). + +Build Tendermint with C libraries: `make build_c`. + +### Ubuntu + +Install LevelDB with snappy: ``` sudo apt-get update diff --git a/docs/spec/abci/abci.md b/docs/spec/abci/abci.md index 287406b7..a1217098 100644 --- a/docs/spec/abci/abci.md +++ b/docs/spec/abci/abci.md @@ -48,16 +48,50 @@ Keys and values in tags must be UTF-8 encoded strings (e.g. ## Determinism -Some methods (`SetOption, Query, CheckTx, DeliverTx`) return -non-deterministic data in the form of `Info` and `Log`. The `Log` is -intended for the literal output from the application's logger, while the -`Info` is any additional info that should be returned. - -All other fields in the `Response*` of all methods must be strictly deterministic. +ABCI applications must implement deterministic finite-state machines to be +securely replicated by the Tendermint consensus. This means block execution +over the Consensus Connection must be strictly deterministic: given the same +ordered set of requests, all nodes will compute identical responses, for all +BeginBlock, DeliverTx, EndBlock, and Commit. This is critical, because the +responses are included in the header of the next block, either via a Merkle root +or directly, so all nodes must agree on exactly what they are. For this reason, it is recommended that applications not be exposed to any external user or process except via the ABCI connections to a consensus engine -like Tendermint Core. +like Tendermint Core. The application must only change its state based on input +from block execution (BeginBlock, DeliverTx, EndBlock, Commit), and not through +any other kind of request. This is the only way to ensure all nodes see the same +transactions and compute the same results. + +If there is some non-determinism in the state machine, consensus will eventually +fail as nodes disagree over the correct values for the block header. The +non-determinism must be fixed and the nodes restarted. + +Sources of non-determinism in applications may include: + +- Hardware failures + - Cosmic rays, overheating, etc. +- Node-dependent state + - Random numbers + - Time +- Underspecification + - Library version changes + - Race conditions + - Floating point numbers + - JSON serialization + - Iterating through hash-tables/maps/dictionaries +- External Sources + - Filesystem + - Network calls (eg. some external REST API service) + +See [#56](https://github.com/tendermint/abci/issues/56) for original discussion. + +Note that some methods (`SetOption, Query, CheckTx, DeliverTx`) return +explicitly non-deterministic data in the form of `Info` and `Log` fields. The `Log` is +intended for the literal output from the application's logger, while the +`Info` is any additional info that should be returned. These are the only fields +that are not included in block header computations, so we don't need agreement +on them. All other fields in the `Response*` must be strictly deterministic. ## Block Execution @@ -217,7 +251,7 @@ Commit are included in the header of the next block. be non-deterministic. - `Info (string)`: Additional information. May be non-deterministic. - - `GasWanted (int64)`: Amount of gas request for transaction. + - `GasWanted (int64)`: Amount of gas requested for transaction. - `GasUsed (int64)`: Amount of gas consumed by transaction. - `Tags ([]cmn.KVPair)`: Key-Value tags for filtering and indexing transactions (eg. by account). diff --git a/docs/spec/abci/apps.md b/docs/spec/abci/apps.md index ac073616..cd88c685 100644 --- a/docs/spec/abci/apps.md +++ b/docs/spec/abci/apps.md @@ -86,18 +86,50 @@ Otherwise it should never be modified. ## Transaction Results -`ResponseCheckTx` and `ResponseDeliverTx` contain the same fields, though they -have slightly different effects. +`ResponseCheckTx` and `ResponseDeliverTx` contain the same fields. -In both cases, `Info` and `Log` are non-deterministic values for debugging/convenience purposes +The `Info` and `Log` fields are non-deterministic values for debugging/convenience purposes that are otherwise ignored. -In both cases, `GasWanted` and `GasUsed` parameters are currently ignored, -though see issues -[#1861](https://github.com/tendermint/tendermint/issues/1861), -[#2299](https://github.com/tendermint/tendermint/issues/2299) and -[#2310](https://github.com/tendermint/tendermint/issues/2310) for how this may -soon change. +The `Data` field must be strictly deterministic, but can be arbitrary data. + +### Gas + +Ethereum introduced the notion of `gas` as an absract representation of the +cost of resources used by nodes when processing transactions. Every operation in the +Ethereum Virtual Machine uses some amount of gas, and gas can be accepted at a market-variable price. +Users propose a maximum amount of gas for their transaction; if the tx uses less, they get +the difference credited back. Tendermint adopts a similar abstraction, +though uses it only optionally and weakly, allowing applications to define +their own sense of the cost of execution. + +In Tendermint, the `ConsensusParams.BlockSize.MaxGas` limits the amount of `gas` that can be used in a block. +The default value is `-1`, meaning no limit, or that the concept of gas is +meaningless. + +Responses contain a `GasWanted` and `GasUsed` field. The former is the maximum +amount of gas the sender of a tx is willing to use, and the later is how much it actually +used. Applications should enforce that `GasUsed <= GasWanted` - ie. tx execution +should halt before it can use more resources than it requested. + +When `MaxGas > -1`, Tendermint enforces the following rules: + + - `GasWanted <= MaxGas` for all txs in the mempool + - `(sum of GasWanted in a block) <= MaxGas` when proposing a block + +If `MaxGas == -1`, no rules about gas are enforced. + +Note that Tendermint does not currently enforce anything about Gas in the consensus, only the mempool. +This means it does not guarantee that committed blocks satisfy these rules! +It is the application's responsibility to return non-zero response codes when gas limits are exceeded. + +The `GasUsed` field is ignored compltely by Tendermint. That said, applications should enforce: + - `GasUsed <= GasWanted` for any given transaction + - `(sum of GasUsed in a block) <= MaxGas` for every block + +In the future, we intend to add a `Priority` field to the responses that can be +used to explicitly prioritize txs in the mempool for inclusion in a block +proposal. See [#1861](https://github.com/tendermint/tendermint/issues/1861). ### CheckTx @@ -142,9 +174,6 @@ If the list is not empty, Tendermint will use it for the validator set. This way the application can determine the initial validator set for the blockchain. -ResponseInitChain also includes ConsensusParams, but these are presently -ignored. - ### EndBlock Updates to the Tendermint validator set can be made by returning @@ -179,14 +208,74 @@ following rules: Note the updates returned in block `H` will only take effect at block `H+2`. +## Consensus Parameters + +ConsensusParams enforce certain limits in the blockchain, like the maximum size +of blocks, amount of gas used in a block, and the maximum acceptable age of +evidence. They can be set in InitChain and updated in EndBlock. + +### BlockSize.MaxBytes + +The maximum size of a complete Amino encoded block. +This is enforced by Tendermint consensus. + +This implies a maximum tx size that is this MaxBytes, less the expected size of +the header, the validator set, and any included evidence in the block. + +Must have `0 < MaxBytes < 100 MB`. + +### BlockSize.MaxGas + +The maximum of the sum of `GasWanted` in a proposed block. +This is *not* enforced by Tendermint consensus. +It is left to the app to enforce (ie. if txs are included past the +limit, they should return non-zero codes). It is used by Tendermint to limit the +txs included in a proposed block. + +Must have `MaxGas >= -1`. +If `MaxGas == -1`, no limit is enforced. + +### EvidenceParams.MaxAge + +This is the maximum age of evidence. +This is enforced by Tendermint consensus. +If a block includes evidence older than this, the block will be rejected +(validators won't vote for it). + +Must have `0 < MaxAge`. + +### Updates + +The application may set the consensus params during InitChain, and update them during +EndBlock. + +#### InitChain + +ResponseInitChain includes a ConsensusParams. +If its nil, Tendermint will use the params loaded in the genesis +file. If it's not nil, Tendermint will use it. +This way the application can determine the initial consensus params for the +blockchain. + +#### EndBlock + +ResponseEndBlock includes a ConsensusParams. +If its nil, Tendermint will do nothing. +If it's not nil, Tendermint will use it. +This way the application can update the consensus params over time. + +Note the updates returned in block `H` will take effect right away for block +`H+1`. + ## Query Query is a generic method with lots of flexibility to enable diverse sets of queries on application state. Tendermint makes use of Query to filter new peers based on ID and IP, and exposes Query to the user over RPC. + Note that calls to Query are not replicated across nodes, but rather query the -local node's state - hence they may provide stale reads. For reads that require -consensus, a transaction is required. +local node's state - hence they may return stale reads. For reads that require +consensus, use a transaction. The most important use of Query is to return Merkle proofs of the application state at some height that can be used for efficient application-specific lite-clients. @@ -235,6 +324,15 @@ using the following paths, with no additional data: If either of these queries return a non-zero ABCI code, Tendermint will refuse to connect to the peer. +### Paths + +Queries are directed at paths, and may optionally include additional data. + +The expectation is for there to be some number of high level paths +differentiating concerns, like `/p2p`, `/store`, and `/app`. Currently, +Tendermint only uses `/p2p`, for filtering peers. For more advanced use, see the +implementation of +[Query in the Cosmos-SDK](https://github.com/cosmos/cosmos-sdk/blob/v0.23.1/baseapp/baseapp.go#L333). ## Crash Recovery diff --git a/docs/tendermint-core/running-in-production.md b/docs/tendermint-core/running-in-production.md index f647bd9b..cd55be53 100644 --- a/docs/tendermint-core/running-in-production.md +++ b/docs/tendermint-core/running-in-production.md @@ -1,5 +1,41 @@ # Running in production +## Database + +By default, Tendermint uses the `syndtr/goleveldb` package for it's in-process +key-value database. Unfortunately, this implementation of LevelDB seems to suffer under heavy load (see +[#226](https://github.com/syndtr/goleveldb/issues/226)). It may be best to +install the real C-implementaiton of LevelDB and compile Tendermint to use +that using `make build_c`. See the [install instructions](../introduction/install) for details. + +Tendermint keeps multiple distinct LevelDB databases in the `$TMROOT/data`: + +- `blockstore.db`: Keeps the entire blockchain - stores blocks, + block commits, and block meta data, each indexed by height. Used to sync new + peers. +- `evidence.db`: Stores all verified evidence of misbehaviour. +- `state.db`: Stores the current blockchain state (ie. height, validators, + consensus params). Only grows if consensus params or validators change. Also + used to temporarily store intermediate results during block processing. +- `tx_index.db`: Indexes txs (and their results) by tx hash and by DeliverTx result tags. + +By default, Tendermint will only index txs by their hash, not by their DeliverTx +result tags. See [indexing transactions](../app-dev/indexing-transactions) for +details. + +There is no current strategy for pruning the databases. Consider reducing +block production by [controlling empty blocks](../tendermint-core/using-tendermint#No-Empty-Blocks) +or by increasing the `consensus.timeout_commit` param. Note both of these are +local settings and not enforced by the consensus. + +We're working on [state +syncing](https://github.com/tendermint/tendermint/issues/828), +which will enable history to be thrown away +and recent application state to be directly synced. We'll need to develop solutions +for archival nodes that allow queries on historical transactions and states. +The Cosmos project has had much success just dumping the latest state of a +blockchain to disk and starting a new chain from that state. + ## Logging Default logging level (`main:info,state:info,*:`) should suffice for @@ -11,6 +47,29 @@ you're trying to debug Tendermint or asked to provide logs with debug logging level, you can do so by running tendermint with `--log_level="*:debug"`. +## Write Ahead Logs (WAL) + +Tendermint uses write ahead logs for the consensus (`cs.wal`) and the mempool +(`mempool.wal`). Both WALs have a max size of 1GB and are automatically rotated.. + +The `consensus.wal` is used to ensure we can recover from a crash at any point +in the consensus state machine. +It writes all consensus messages (timeouts, proposals, block part, or vote) +to a single file, flushing to disk before processing messages from its own +validator. Since Tendermint validators are expected to never sign a conflicting vote, the +WAL ensures we can always recover deterministically to the latest state of the consensus without +using the network or re-signing any consensus messages. + +If your `consensus.wal` is corrupted, see [below](#WAL-Corruption). + +The `mempool.wal` logs all incoming txs before running CheckTx, but is +otherwise not used in any programmatic way. It's just a kind of manual +safe guard. Note the mempool provides no durability guarantees - a tx sent to one or many nodes +may never make it into the blockchain if those nodes crash before being able to +propose it. Clients must monitor their txs by subscribing over websockets, +polling for them, or using `/broadcast_tx_commit`. In the worst case, txs can be +resent from the mempool WAL manually. + ## DOS Exposure and Mitigation Validators are supposed to setup [Sentry Node diff --git a/mempool/mempool.go b/mempool/mempool.go index 543c274d..2096912f 100644 --- a/mempool/mempool.go +++ b/mempool/mempool.go @@ -22,14 +22,14 @@ import ( "github.com/tendermint/tendermint/types" ) -// PreCheckFunc is an optional filter to determine if a transaction should be -// rejected. Invoked before CheckTx. An example would be to ensure that a -// transaction isn't exceeded the block size. +// PreCheckFunc is an optional filter executed before CheckTx and rejects +// transaction if false is returned. An example would be to ensure that a +// transaction doesn't exceeded the block size. type PreCheckFunc func(types.Tx) bool // PostCheckFunc is an optional filter executed after CheckTx and rejects // transaction if false is returned. An example would be to ensure a -// transaction doesn't require more gas than available. +// transaction doesn't require more gas than available for the block. type PostCheckFunc func(types.Tx, *abci.ResponseCheckTx) bool /*