tendermint/node/node.go

1197 lines
36 KiB
Go
Raw Normal View History

package node
import (
"bytes"
"context"
"fmt"
"net"
"net/http"
_ "net/http/pprof"
"os"
"strings"
"time"
"github.com/pkg/errors"
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/client_golang/prometheus/promhttp"
2018-11-14 15:47:41 +03:00
"github.com/rs/cors"
amino "github.com/tendermint/go-amino"
abci "github.com/tendermint/tendermint/abci/types"
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
bcv0 "github.com/tendermint/tendermint/blockchain/v0"
bcv1 "github.com/tendermint/tendermint/blockchain/v1"
cfg "github.com/tendermint/tendermint/config"
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
"github.com/tendermint/tendermint/consensus"
cs "github.com/tendermint/tendermint/consensus"
privval: refactor Remote signers (#3370) This PR is related to #3107 and a continuation of #3351 It is important to emphasise that in the privval original design, client/server and listening/dialing roles are inverted and do not follow a conventional interaction. Given two hosts A and B: Host A is listener/client Host B is dialer/server (contains the secret key) When A requires a signature, it needs to wait for B to dial in before it can issue a request. A only accepts a single connection and any failure leads to dropping the connection and waiting for B to reconnect. The original rationale behind this design was based on security. Host B only allows outbound connections to a list of whitelisted hosts. It is not possible to reach B unless B dials in. There are no listening/open ports in B. This PR results in the following changes: Refactors ping/heartbeat to avoid previously existing race conditions. Separates transport (dialer/listener) from signing (client/server) concerns to simplify workflow. Unifies and abstracts away the differences between unix and tcp sockets. A single signer endpoint implementation unifies connection handling code (read/write/close/connection obj) The signer request handler (server side) is customizable to increase testability. Updates and extends unit tests A high level overview of the classes is as follows: Transport (endpoints): The following classes take care of establishing a connection SignerDialerEndpoint SignerListeningEndpoint SignerEndpoint groups common functionality (read/write/timeouts/etc.) Signing (client/server): The following classes take care of exchanging request/responses SignerClient SignerServer This PR also closes #3601 Commits: * refactoring - work in progress * reworking unit tests * Encapsulating and fixing unit tests * Improve tests * Clean up * Fix/improve unit tests * clean up tests * Improving service endpoint * fixing unit test * fix linter issues * avoid invalid cache values (improve later?) * complete implementation * wip * improved connection loop * Improve reconnections + fixing unit tests * addressing comments * small formatting changes * clean up * Update node/node.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * check during initialization * dropping connecting when writing fails * removing break * use t.log instead * unifying and using cmn.GetFreePort() * review fixes * reordering and unifying drop connection * closing instead of signalling * refactored service loop * removed superfluous brackets * GetPubKey can return errors * Revert "GetPubKey can return errors" This reverts commit 68c06f19b4650389d7e5ab1659b318889028202c. * adding entry to changelog * Update CHANGELOG_PENDING.md Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * updating node.go * review fixes * fixes linter * fixing unit test * small fixes in comments * addressing review comments * addressing review comments 2 * reverting suggestion * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * do not expose brokenSignerDialerEndpoint * clean up logging * unifying methods shorten test time signer also drops * reenabling pings * improving testability + unit test * fixing go fmt + unit test * remove unused code * Addressing review comments * simplifying connection workflow * fix linter/go import issue * using base service quit * updating comment * Simplifying design + adjusting names * fixing linter issues * refactoring test harness + fixes * Addressing review comments * cleaning up * adding additional error check
2019-08-05 17:09:10 +02:00
"github.com/tendermint/tendermint/crypto"
"github.com/tendermint/tendermint/evidence"
cmn "github.com/tendermint/tendermint/libs/common"
"github.com/tendermint/tendermint/libs/log"
limit number of /subscribe clients and queries per client (#3269) * limit number of /subscribe clients and queries per client Add the following config variables (under [rpc] section): * max_subscription_clients * max_subscriptions_per_client * timeout_broadcast_tx_commit Fixes #2826 new HTTPClient interface for subscriptions finalize HTTPClient events interface remove EventSubscriber fix data race ``` WARNING: DATA RACE Read at 0x00c000a36060 by goroutine 129: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe.func1() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:168 +0x1f0 Previous write at 0x00c000a36060 by goroutine 132: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:191 +0x4e0 github.com/tendermint/tendermint/rpc/client.WaitForOneEvent() /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 Goroutine 129 (running) created at: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:164 +0x4b7 github.com/tendermint/tendermint/rpc/client.WaitForOneEvent() /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 Goroutine 132 (running) created at: testing.(*T).Run() /usr/local/go/src/testing/testing.go:878 +0x659 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:119 +0x186 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 ================== ``` lite client works (tested manually) godoc comments httpclient: do not close the out channel use TimeoutBroadcastTxCommit no timeout for unsubscribe but 1s Local (5s HTTP) timeout for resubscribe format code change Subscribe#out cap to 1 and replace config vars with RPCConfig TimeoutBroadcastTxCommit can't be greater than rpcserver.WriteTimeout rpc: Context as first parameter to all functions reformat code fixes after my own review fixes after Ethan's review add test stubs fix config.toml * fixes after manual testing - rpc: do not recommend to use BroadcastTxCommit because it's slow and wastes Tendermint resources (pubsub) - rpc: better error in Subscribe and BroadcastTxCommit - HTTPClient: do not resubscribe if err = ErrAlreadySubscribed * fixes after Ismail's review * Update rpc/grpc/grpc_test.go Co-Authored-By: melekes <anton.kalyaev@gmail.com>
2019-03-11 22:45:58 +04:00
tmpubsub "github.com/tendermint/tendermint/libs/pubsub"
mempl "github.com/tendermint/tendermint/mempool"
"github.com/tendermint/tendermint/p2p"
"github.com/tendermint/tendermint/p2p/pex"
"github.com/tendermint/tendermint/privval"
"github.com/tendermint/tendermint/proxy"
rpccore "github.com/tendermint/tendermint/rpc/core"
ctypes "github.com/tendermint/tendermint/rpc/core/types"
grpccore "github.com/tendermint/tendermint/rpc/grpc"
rpcserver "github.com/tendermint/tendermint/rpc/lib/server"
sm "github.com/tendermint/tendermint/state"
"github.com/tendermint/tendermint/state/txindex"
"github.com/tendermint/tendermint/state/txindex/kv"
"github.com/tendermint/tendermint/state/txindex/null"
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
"github.com/tendermint/tendermint/store"
"github.com/tendermint/tendermint/types"
tmtime "github.com/tendermint/tendermint/types/time"
"github.com/tendermint/tendermint/version"
dbm "github.com/tendermint/tm-db"
)
//------------------------------------------------------------------------------
// DBContext specifies config information for loading a new DB.
type DBContext struct {
ID string
Config *cfg.Config
}
// DBProvider takes a DBContext and returns an instantiated DB.
type DBProvider func(*DBContext) (dbm.DB, error)
// DefaultDBProvider returns a database using the DBBackend and DBDir
// specified in the ctx.Config.
func DefaultDBProvider(ctx *DBContext) (dbm.DB, error) {
dbType := dbm.DBBackendType(ctx.Config.DBBackend)
return dbm.NewDB(ctx.ID, dbType, ctx.Config.DBDir()), nil
}
// GenesisDocProvider returns a GenesisDoc.
// It allows the GenesisDoc to be pulled from sources other than the
// filesystem, for instance from a distributed key-value store cluster.
type GenesisDocProvider func() (*types.GenesisDoc, error)
// DefaultGenesisDocProviderFunc returns a GenesisDocProvider that loads
// the GenesisDoc from the config.GenesisFile() on the filesystem.
func DefaultGenesisDocProviderFunc(config *cfg.Config) GenesisDocProvider {
return func() (*types.GenesisDoc, error) {
return types.GenesisDocFromFile(config.GenesisFile())
}
}
// NodeProvider takes a config and a logger and returns a ready to go Node.
type NodeProvider func(*cfg.Config, log.Logger) (*Node, error)
// DefaultNewNode returns a Tendermint node with default settings for the
// PrivValidator, ClientCreator, GenesisDoc, and DBProvider.
// It implements NodeProvider.
func DefaultNewNode(config *cfg.Config, logger log.Logger) (*Node, error) {
// Generate node PrivKey
nodeKey, err := p2p.LoadOrGenNodeKey(config.NodeKeyFile())
if err != nil {
return nil, err
}
// Convert old PrivValidator if it exists.
oldPrivVal := config.OldPrivValidatorFile()
newPrivValKey := config.PrivValidatorKeyFile()
newPrivValState := config.PrivValidatorStateFile()
if _, err := os.Stat(oldPrivVal); !os.IsNotExist(err) {
oldPV, err := privval.LoadOldFilePV(oldPrivVal)
if err != nil {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return nil, fmt.Errorf("error reading OldPrivValidator from %v: %v\n", oldPrivVal, err)
}
logger.Info("Upgrading PrivValidator file",
"old", oldPrivVal,
"newKey", newPrivValKey,
"newState", newPrivValState,
)
oldPV.Upgrade(newPrivValKey, newPrivValState)
}
return NewNode(config,
privval.LoadOrGenFilePV(newPrivValKey, newPrivValState),
nodeKey,
proxy.DefaultClientCreator(config.ProxyApp, config.ABCI, config.DBDir()),
DefaultGenesisDocProviderFunc(config),
DefaultDBProvider,
DefaultMetricsProvider(config.Instrumentation),
logger,
)
}
// MetricsProvider returns a consensus, p2p and mempool Metrics.
type MetricsProvider func(chainID string) (*cs.Metrics, *p2p.Metrics, *mempl.Metrics, *sm.Metrics)
// DefaultMetricsProvider returns Metrics build using Prometheus client library
// if Prometheus is enabled. Otherwise, it returns no-op Metrics.
func DefaultMetricsProvider(config *cfg.InstrumentationConfig) MetricsProvider {
return func(chainID string) (*cs.Metrics, *p2p.Metrics, *mempl.Metrics, *sm.Metrics) {
if config.Prometheus {
return cs.PrometheusMetrics(config.Namespace, "chain_id", chainID),
p2p.PrometheusMetrics(config.Namespace, "chain_id", chainID),
mempl.PrometheusMetrics(config.Namespace, "chain_id", chainID),
sm.PrometheusMetrics(config.Namespace, "chain_id", chainID)
}
return cs.NopMetrics(), p2p.NopMetrics(), mempl.NopMetrics(), sm.NopMetrics()
}
}
// Option sets a parameter for the node.
type Option func(*Node)
// CustomReactors allows you to add custom reactors (name -> p2p.Reactor) to
// the node's Switch.
//
// WARNING: using any name from the below list of the existing reactors will
// result in replacing it with the custom one.
//
// - MEMPOOL
// - BLOCKCHAIN
// - CONSENSUS
// - EVIDENCE
// - PEX
func CustomReactors(reactors map[string]p2p.Reactor) Option {
return func(n *Node) {
for name, reactor := range reactors {
if existingReactor := n.sw.Reactor(name); existingReactor != nil {
n.sw.Logger.Info("Replacing existing reactor with a custom one",
"name", name, "existing", existingReactor, "custom", reactor)
n.sw.RemoveReactor(name, existingReactor)
}
n.sw.AddReactor(name, reactor)
}
}
}
//------------------------------------------------------------------------------
// Node is the highest level interface to a full Tendermint node.
// It includes all configuration information and running services.
type Node struct {
cmn.BaseService
// config
config *cfg.Config
genesisDoc *types.GenesisDoc // initial validator set
privValidator types.PrivValidator // local node's validator key
// network
transport *p2p.MultiplexTransport
sw *p2p.Switch // p2p connections
addrBook pex.AddrBook // known peers
nodeInfo p2p.NodeInfo
nodeKey *p2p.NodeKey // our node privkey
isListening bool
// services
eventBus *types.EventBus // pub/sub for services
stateDB dbm.DB
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
blockStore *store.BlockStore // store the blockchain to disk
bcReactor p2p.Reactor // for fast-syncing
mempoolReactor *mempl.Reactor // for gossipping transactions
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
mempool mempl.Mempool
consensusState *cs.ConsensusState // latest consensus state
consensusReactor *cs.ConsensusReactor // for participating in the consensus
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
pexReactor *pex.PEXReactor // for exchanging peer addresses
evidencePool *evidence.EvidencePool // tracking evidence
proxyApp proxy.AppConns // connection to the application
rpcListeners []net.Listener // rpc servers
txIndexer txindex.TxIndexer
indexerService *txindex.IndexerService
prometheusSrv *http.Server
}
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
func initDBs(config *cfg.Config, dbProvider DBProvider) (blockStore *store.BlockStore, stateDB dbm.DB, err error) {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
var blockStoreDB dbm.DB
blockStoreDB, err = dbProvider(&DBContext{"blockstore", config})
if err != nil {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return
}
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
blockStore = store.NewBlockStore(blockStoreDB)
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
stateDB, err = dbProvider(&DBContext{"state", config})
if err != nil {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
func createAndStartProxyAppConns(clientCreator proxy.ClientCreator, logger log.Logger) (proxy.AppConns, error) {
proxyApp := proxy.NewAppConns(clientCreator)
proxyApp.SetLogger(logger.With("module", "proxy"))
if err := proxyApp.Start(); err != nil {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return nil, fmt.Errorf("error starting proxy app connections: %v", err)
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return proxyApp, nil
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
func createAndStartEventBus(logger log.Logger) (*types.EventBus, error) {
eventBus := types.NewEventBus()
eventBus.SetLogger(logger.With("module", "events"))
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
if err := eventBus.Start(); err != nil {
return nil, err
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return eventBus, nil
}
func createAndStartIndexerService(config *cfg.Config, dbProvider DBProvider,
eventBus *types.EventBus, logger log.Logger) (*txindex.IndexerService, txindex.TxIndexer, error) {
var txIndexer txindex.TxIndexer
switch config.TxIndex.Indexer {
case "kv":
store, err := dbProvider(&DBContext{"tx_index", config})
if err != nil {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return nil, nil, err
}
2019-08-02 08:53:52 +02:00
switch {
case config.TxIndex.IndexTags != "":
txIndexer = kv.NewTxIndex(store, kv.IndexTags(splitAndTrimEmpty(config.TxIndex.IndexTags, ",", " ")))
2019-08-02 08:53:52 +02:00
case config.TxIndex.IndexAllTags:
txIndexer = kv.NewTxIndex(store, kv.IndexAllTags())
2019-08-02 08:53:52 +02:00
default:
txIndexer = kv.NewTxIndex(store)
}
default:
txIndexer = &null.TxIndex{}
}
indexerService := txindex.NewIndexerService(txIndexer, eventBus)
indexerService.SetLogger(logger.With("module", "txindex"))
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
if err := indexerService.Start(); err != nil {
return nil, nil, err
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return indexerService, txIndexer, nil
}
func doHandshake(stateDB dbm.DB, state sm.State, blockStore sm.BlockStore,
genDoc *types.GenesisDoc, eventBus types.BlockEventPublisher, proxyApp proxy.AppConns, consensusLogger log.Logger) error {
handshaker := cs.NewHandshaker(stateDB, state, blockStore, genDoc)
handshaker.SetLogger(consensusLogger)
handshaker.SetEventBus(eventBus)
if err := handshaker.Handshake(proxyApp); err != nil {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return fmt.Errorf("error during handshake: %v", err)
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return nil
}
privval: refactor Remote signers (#3370) This PR is related to #3107 and a continuation of #3351 It is important to emphasise that in the privval original design, client/server and listening/dialing roles are inverted and do not follow a conventional interaction. Given two hosts A and B: Host A is listener/client Host B is dialer/server (contains the secret key) When A requires a signature, it needs to wait for B to dial in before it can issue a request. A only accepts a single connection and any failure leads to dropping the connection and waiting for B to reconnect. The original rationale behind this design was based on security. Host B only allows outbound connections to a list of whitelisted hosts. It is not possible to reach B unless B dials in. There are no listening/open ports in B. This PR results in the following changes: Refactors ping/heartbeat to avoid previously existing race conditions. Separates transport (dialer/listener) from signing (client/server) concerns to simplify workflow. Unifies and abstracts away the differences between unix and tcp sockets. A single signer endpoint implementation unifies connection handling code (read/write/close/connection obj) The signer request handler (server side) is customizable to increase testability. Updates and extends unit tests A high level overview of the classes is as follows: Transport (endpoints): The following classes take care of establishing a connection SignerDialerEndpoint SignerListeningEndpoint SignerEndpoint groups common functionality (read/write/timeouts/etc.) Signing (client/server): The following classes take care of exchanging request/responses SignerClient SignerServer This PR also closes #3601 Commits: * refactoring - work in progress * reworking unit tests * Encapsulating and fixing unit tests * Improve tests * Clean up * Fix/improve unit tests * clean up tests * Improving service endpoint * fixing unit test * fix linter issues * avoid invalid cache values (improve later?) * complete implementation * wip * improved connection loop * Improve reconnections + fixing unit tests * addressing comments * small formatting changes * clean up * Update node/node.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * check during initialization * dropping connecting when writing fails * removing break * use t.log instead * unifying and using cmn.GetFreePort() * review fixes * reordering and unifying drop connection * closing instead of signalling * refactored service loop * removed superfluous brackets * GetPubKey can return errors * Revert "GetPubKey can return errors" This reverts commit 68c06f19b4650389d7e5ab1659b318889028202c. * adding entry to changelog * Update CHANGELOG_PENDING.md Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * updating node.go * review fixes * fixes linter * fixing unit test * small fixes in comments * addressing review comments * addressing review comments 2 * reverting suggestion * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * do not expose brokenSignerDialerEndpoint * clean up logging * unifying methods shorten test time signer also drops * reenabling pings * improving testability + unit test * fixing go fmt + unit test * remove unused code * Addressing review comments * simplifying connection workflow * fix linter/go import issue * using base service quit * updating comment * Simplifying design + adjusting names * fixing linter issues * refactoring test harness + fixes * Addressing review comments * cleaning up * adding additional error check
2019-08-05 17:09:10 +02:00
func logNodeStartupInfo(state sm.State, pubKey crypto.PubKey, logger, consensusLogger log.Logger) {
// Log the version info.
logger.Info("Version info",
"software", version.TMCoreSemVer,
"block", version.BlockProtocol,
"p2p", version.P2PProtocol,
)
// If the state and software differ in block version, at least log it.
if state.Version.Consensus.Block != version.BlockProtocol {
logger.Info("Software and state have different block protocols",
"software", version.BlockProtocol,
"state", state.Version.Consensus.Block,
)
}
addr := pubKey.Address()
// Log whether this node is a validator or an observer
if state.Validators.HasAddress(addr) {
consensusLogger.Info("This node is a validator", "addr", addr, "pubKey", pubKey)
} else {
consensusLogger.Info("This node is not a validator", "addr", addr, "pubKey", pubKey)
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
func onlyValidatorIsUs(state sm.State, privVal types.PrivValidator) bool {
if state.Validators.Size() > 1 {
return false
}
addr, _ := state.Validators.GetByIndex(0)
return bytes.Equal(privVal.GetPubKey().Address(), addr)
}
func createMempoolAndMempoolReactor(config *cfg.Config, proxyApp proxy.AppConns,
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
state sm.State, memplMetrics *mempl.Metrics, logger log.Logger) (*mempl.Reactor, *mempl.CListMempool) {
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
mempool := mempl.NewCListMempool(
config.Mempool,
proxyApp.Mempool(),
state.LastBlockHeight,
mempl.WithMetrics(memplMetrics),
mempl.WithPreCheck(sm.TxPreCheck(state)),
mempl.WithPostCheck(sm.TxPostCheck(state)),
)
max-bytes PR follow-up (#2318) * ReapMaxTxs: return all txs if max is negative this mirrors ReapMaxBytes behavior See https://github.com/tendermint/tendermint/pull/2184#discussion_r214439950 * increase MaxAminoOverheadForBlock tested with: ``` func TestMaxAminoOverheadForBlock(t *testing.T) { maxChainID := "" for i := 0; i < MaxChainIDLen; i++ { maxChainID += "𠜎" } h := Header{ ChainID: maxChainID, Height: 10, Time: time.Now().UTC(), NumTxs: 100, TotalTxs: 200, LastBlockID: makeBlockID(make([]byte, 20), 300, make([]byte, 20)), LastCommitHash: tmhash.Sum([]byte("last_commit_hash")), DataHash: tmhash.Sum([]byte("data_hash")), ValidatorsHash: tmhash.Sum([]byte("validators_hash")), NextValidatorsHash: tmhash.Sum([]byte("next_validators_hash")), ConsensusHash: tmhash.Sum([]byte("consensus_hash")), AppHash: tmhash.Sum([]byte("app_hash")), LastResultsHash: tmhash.Sum([]byte("last_results_hash")), EvidenceHash: tmhash.Sum([]byte("evidence_hash")), ProposerAddress: tmhash.Sum([]byte("proposer_address")), } b := Block{ Header: h, Data: Data{Txs: makeTxs(10000, 100)}, Evidence: EvidenceData{}, LastCommit: &Commit{}, } bz, err := cdc.MarshalBinary(b) require.NoError(t, err) assert.Equal(t, MaxHeaderBytes+MaxAminoOverheadForBlock-2, len(bz)-1000000-20000-1) } ``` * fix MaxYYY constants calculation by using math.MaxInt64 See https://github.com/tendermint/tendermint/pull/2184#discussion_r214444244 * pass mempool filter as an option See https://github.com/tendermint/tendermint/pull/2184#discussion_r214445869 * fixes after Dev's comments
2018-09-04 11:46:34 +04:00
mempoolLogger := logger.With("module", "mempool")
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
mempoolReactor := mempl.NewReactor(config.Mempool, mempool)
mempoolReactor.SetLogger(mempoolLogger)
if config.Consensus.WaitForTxs() {
mempool.EnableTxsAvailable()
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return mempoolReactor, mempool
}
func createEvidenceReactor(config *cfg.Config, dbProvider DBProvider,
stateDB dbm.DB, logger log.Logger) (*evidence.EvidenceReactor, *evidence.EvidencePool, error) {
evidenceDB, err := dbProvider(&DBContext{"evidence", config})
if err != nil {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return nil, nil, err
}
evidenceLogger := logger.With("module", "evidence")
evidencePool := evidence.NewEvidencePool(stateDB, evidenceDB)
evidencePool.SetLogger(evidenceLogger)
evidenceReactor := evidence.NewEvidenceReactor(evidencePool)
evidenceReactor.SetLogger(evidenceLogger)
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return evidenceReactor, evidencePool, nil
}
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
func createBlockchainReactor(config *cfg.Config,
state sm.State,
blockExec *sm.BlockExecutor,
blockStore *store.BlockStore,
fastSync bool,
logger log.Logger) (bcReactor p2p.Reactor, err error) {
switch config.FastSync.Version {
case "v0":
bcReactor = bcv0.NewBlockchainReactor(state.Copy(), blockExec, blockStore, fastSync)
case "v1":
bcReactor = bcv1.NewBlockchainReactor(state.Copy(), blockExec, blockStore, fastSync)
default:
return nil, fmt.Errorf("unknown fastsync version %s", config.FastSync.Version)
}
bcReactor.SetLogger(logger.With("module", "blockchain"))
return bcReactor, nil
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
func createConsensusReactor(config *cfg.Config,
state sm.State,
blockExec *sm.BlockExecutor,
blockStore sm.BlockStore,
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
mempool *mempl.CListMempool,
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
evidencePool *evidence.EvidencePool,
privValidator types.PrivValidator,
csMetrics *cs.Metrics,
fastSync bool,
eventBus *types.EventBus,
consensusLogger log.Logger) (*consensus.ConsensusReactor, *consensus.ConsensusState) {
consensusState := cs.NewConsensusState(
config.Consensus,
state.Copy(),
blockExec,
blockStore,
mempool,
evidencePool,
cs.StateMetrics(csMetrics),
)
consensusState.SetLogger(consensusLogger)
if privValidator != nil {
consensusState.SetPrivValidator(privValidator)
}
consensusReactor := cs.NewConsensusReactor(consensusState, fastSync, cs.ReactorMetrics(csMetrics))
consensusReactor.SetLogger(consensusLogger)
// services which will be publishing and/or subscribing for messages (events)
// consensusReactor will set it on consensusState and blockExecutor
consensusReactor.SetEventBus(eventBus)
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return consensusReactor, consensusState
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
func createTransport(config *cfg.Config, nodeInfo p2p.NodeInfo, nodeKey *p2p.NodeKey, proxyApp proxy.AppConns) (*p2p.MultiplexTransport, []p2p.PeerFilterFunc) {
var (
mConnConfig = p2p.MConnConfig(config.P2P)
transport = p2p.NewMultiplexTransport(nodeInfo, *nodeKey, mConnConfig)
connFilters = []p2p.ConnFilterFunc{}
peerFilters = []p2p.PeerFilterFunc{}
)
if !config.P2P.AllowDuplicateIP {
connFilters = append(connFilters, p2p.ConnDuplicateIPFilter())
}
// Filter peers by addr or pubkey with an ABCI query.
// If the query return code is OK, add peer.
if config.FilterPeers {
connFilters = append(
connFilters,
// ABCI query for address filtering.
func(_ p2p.ConnSet, c net.Conn, _ []net.IP) error {
res, err := proxyApp.Query().QuerySync(abci.RequestQuery{
Path: fmt.Sprintf("/p2p/filter/addr/%s", c.RemoteAddr().String()),
})
if err != nil {
return err
}
if res.IsErr() {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return fmt.Errorf("error querying abci app: %v", res)
}
return nil
},
)
peerFilters = append(
peerFilters,
// ABCI query for ID filtering.
func(_ p2p.IPeerSet, p p2p.Peer) error {
res, err := proxyApp.Query().QuerySync(abci.RequestQuery{
Path: fmt.Sprintf("/p2p/filter/id/%s", p.ID()),
})
if err != nil {
return err
}
if res.IsErr() {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return fmt.Errorf("error querying abci app: %v", res)
}
return nil
},
)
}
p2p.MultiplexTransportConnFilters(connFilters...)(transport)
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return transport, peerFilters
}
func createSwitch(config *cfg.Config,
transport p2p.Transport,
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
p2pMetrics *p2p.Metrics,
peerFilters []p2p.PeerFilterFunc,
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
mempoolReactor *mempl.Reactor,
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
bcReactor p2p.Reactor,
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
consensusReactor *consensus.ConsensusReactor,
evidenceReactor *evidence.EvidenceReactor,
nodeInfo p2p.NodeInfo,
nodeKey *p2p.NodeKey,
p2pLogger log.Logger) *p2p.Switch {
sw := p2p.NewSwitch(
config.P2P,
transport,
p2p.WithMetrics(p2pMetrics),
p2p.SwitchPeerFilters(peerFilters...),
)
sw.SetLogger(p2pLogger)
sw.AddReactor("MEMPOOL", mempoolReactor)
sw.AddReactor("BLOCKCHAIN", bcReactor)
sw.AddReactor("CONSENSUS", consensusReactor)
sw.AddReactor("EVIDENCE", evidenceReactor)
sw.SetNodeInfo(nodeInfo)
sw.SetNodeKey(nodeKey)
p2pLogger.Info("P2P Node ID", "ID", nodeKey.ID(), "file", config.NodeKeyFile())
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
return sw
}
func createAddrBookAndSetOnSwitch(config *cfg.Config, sw *p2p.Switch,
p2pLogger log.Logger, nodeKey *p2p.NodeKey) (pex.AddrBook, error) {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
addrBook := pex.NewAddrBook(config.P2P.AddrBookFile(), config.P2P.AddrBookStrict)
addrBook.SetLogger(p2pLogger.With("book", config.P2P.AddrBookFile()))
// Add ourselves to addrbook to prevent dialing ourselves
if config.P2P.ExternalAddress != "" {
addr, err := p2p.NewNetAddressString(p2p.IDAddressString(nodeKey.ID(), config.P2P.ExternalAddress))
if err != nil {
return nil, errors.Wrap(err, "p2p.external_address is incorrect")
}
addrBook.AddOurAddress(addr)
}
if config.P2P.ListenAddress != "" {
addr, err := p2p.NewNetAddressString(p2p.IDAddressString(nodeKey.ID(), config.P2P.ListenAddress))
if err != nil {
return nil, errors.Wrap(err, "p2p.laddr is incorrect")
}
addrBook.AddOurAddress(addr)
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
sw.SetAddrBook(addrBook)
return addrBook, nil
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
}
func createPEXReactorAndAddToSwitch(addrBook pex.AddrBook, config *cfg.Config,
sw *p2p.Switch, logger log.Logger) *pex.PEXReactor {
// TODO persistent peers ? so we can have their DNS addrs saved
pexReactor := pex.NewPEXReactor(addrBook,
&pex.PEXReactorConfig{
Seeds: splitAndTrimEmpty(config.P2P.Seeds, ",", " "),
SeedMode: config.P2P.SeedMode,
// See consensus/reactor.go: blocksToContributeToBecomeGoodPeer 10000
// blocks assuming 10s blocks ~ 28 hours.
// TODO (melekes): make it dynamic based on the actual block latencies
// from the live network.
// https://github.com/tendermint/tendermint/issues/3523
SeedDisconnectWaitPeriod: 28 * time.Hour,
})
pexReactor.SetLogger(logger.With("module", "pex"))
sw.AddReactor("PEX", pexReactor)
return pexReactor
}
// NewNode returns a new, ready to go, Tendermint Node.
func NewNode(config *cfg.Config,
privValidator types.PrivValidator,
nodeKey *p2p.NodeKey,
clientCreator proxy.ClientCreator,
genesisDocProvider GenesisDocProvider,
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
dbProvider DBProvider,
metricsProvider MetricsProvider,
logger log.Logger,
options ...Option) (*Node, error) {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
blockStore, stateDB, err := initDBs(config, dbProvider)
if err != nil {
return nil, err
}
state, genDoc, err := LoadStateFromDBOrGenesisDocProvider(stateDB, genesisDocProvider)
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
if err != nil {
return nil, err
}
// Create the proxyApp and establish connections to the ABCI app (consensus, mempool, query).
proxyApp, err := createAndStartProxyAppConns(clientCreator, logger)
if err != nil {
return nil, err
}
// EventBus and IndexerService must be started before the handshake because
// we might need to index the txs of the replayed block as this might not have happened
// when the node stopped last time (i.e. the node stopped after it saved the block
// but before it indexed the txs, or, endblocker panicked)
eventBus, err := createAndStartEventBus(logger)
if err != nil {
return nil, err
}
// Transaction indexing
indexerService, txIndexer, err := createAndStartIndexerService(config, dbProvider, eventBus, logger)
if err != nil {
return nil, err
}
// Create the handshaker, which calls RequestInfo, sets the AppVersion on the state,
// and replays any blocks as necessary to sync tendermint with the app.
consensusLogger := logger.With("module", "consensus")
if err := doHandshake(stateDB, state, blockStore, genDoc, eventBus, proxyApp, consensusLogger); err != nil {
return nil, err
}
// Reload the state. It will have the Version.Consensus.App set by the
// Handshake, and may have other modifications as well (ie. depending on
// what happened during block replay).
state = sm.LoadState(stateDB)
// If an address is provided, listen on the socket for a connection from an
// external signing process.
if config.PrivValidatorListenAddr != "" {
// FIXME: we should start services inside OnStart
privValidator, err = createAndStartPrivValidatorSocketClient(config.PrivValidatorListenAddr, logger)
if err != nil {
return nil, errors.Wrap(err, "error with private validator socket client")
}
}
privval: refactor Remote signers (#3370) This PR is related to #3107 and a continuation of #3351 It is important to emphasise that in the privval original design, client/server and listening/dialing roles are inverted and do not follow a conventional interaction. Given two hosts A and B: Host A is listener/client Host B is dialer/server (contains the secret key) When A requires a signature, it needs to wait for B to dial in before it can issue a request. A only accepts a single connection and any failure leads to dropping the connection and waiting for B to reconnect. The original rationale behind this design was based on security. Host B only allows outbound connections to a list of whitelisted hosts. It is not possible to reach B unless B dials in. There are no listening/open ports in B. This PR results in the following changes: Refactors ping/heartbeat to avoid previously existing race conditions. Separates transport (dialer/listener) from signing (client/server) concerns to simplify workflow. Unifies and abstracts away the differences between unix and tcp sockets. A single signer endpoint implementation unifies connection handling code (read/write/close/connection obj) The signer request handler (server side) is customizable to increase testability. Updates and extends unit tests A high level overview of the classes is as follows: Transport (endpoints): The following classes take care of establishing a connection SignerDialerEndpoint SignerListeningEndpoint SignerEndpoint groups common functionality (read/write/timeouts/etc.) Signing (client/server): The following classes take care of exchanging request/responses SignerClient SignerServer This PR also closes #3601 Commits: * refactoring - work in progress * reworking unit tests * Encapsulating and fixing unit tests * Improve tests * Clean up * Fix/improve unit tests * clean up tests * Improving service endpoint * fixing unit test * fix linter issues * avoid invalid cache values (improve later?) * complete implementation * wip * improved connection loop * Improve reconnections + fixing unit tests * addressing comments * small formatting changes * clean up * Update node/node.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * check during initialization * dropping connecting when writing fails * removing break * use t.log instead * unifying and using cmn.GetFreePort() * review fixes * reordering and unifying drop connection * closing instead of signalling * refactored service loop * removed superfluous brackets * GetPubKey can return errors * Revert "GetPubKey can return errors" This reverts commit 68c06f19b4650389d7e5ab1659b318889028202c. * adding entry to changelog * Update CHANGELOG_PENDING.md Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * updating node.go * review fixes * fixes linter * fixing unit test * small fixes in comments * addressing review comments * addressing review comments 2 * reverting suggestion * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * do not expose brokenSignerDialerEndpoint * clean up logging * unifying methods shorten test time signer also drops * reenabling pings * improving testability + unit test * fixing go fmt + unit test * remove unused code * Addressing review comments * simplifying connection workflow * fix linter/go import issue * using base service quit * updating comment * Simplifying design + adjusting names * fixing linter issues * refactoring test harness + fixes * Addressing review comments * cleaning up * adding additional error check
2019-08-05 17:09:10 +02:00
pubKey := privValidator.GetPubKey()
if pubKey == nil {
// TODO: GetPubKey should return errors - https://github.com/tendermint/tendermint/issues/3602
return nil, errors.New("could not retrieve public key from private validator")
}
logNodeStartupInfo(state, pubKey, logger, consensusLogger)
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
// Decide whether to fast-sync or not
// We don't fast-sync when the only validator is us.
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
fastSync := config.FastSyncMode && !onlyValidatorIsUs(state, privValidator)
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
csMetrics, p2pMetrics, memplMetrics, smMetrics := metricsProvider(genDoc.ChainID)
// Make MempoolReactor
mempoolReactor, mempool := createMempoolAndMempoolReactor(config, proxyApp, state, memplMetrics, logger)
// Make Evidence Reactor
evidenceReactor, evidencePool, err := createEvidenceReactor(config, dbProvider, stateDB, logger)
if err != nil {
return nil, err
}
// make block executor for consensus and blockchain reactors to execute blocks
blockExec := sm.NewBlockExecutor(
stateDB,
logger.With("module", "state"),
proxyApp.Consensus(),
mempool,
evidencePool,
sm.BlockExecutorWithMetrics(smMetrics),
)
// Make BlockchainReactor
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
bcReactor, err := createBlockchainReactor(config, state, blockExec, blockStore, fastSync, logger)
if err != nil {
return nil, errors.Wrap(err, "could not create blockchain reactor")
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
// Make ConsensusReactor
consensusReactor, consensusState := createConsensusReactor(
config, state, blockExec, blockStore, mempool, evidencePool,
privValidator, csMetrics, fastSync, eventBus, consensusLogger,
)
nodeInfo, err := makeNodeInfo(config, nodeKey, txIndexer, genDoc, state)
if err != nil {
return nil, err
}
// Setup Transport.
transport, peerFilters := createTransport(config, nodeInfo, nodeKey, proxyApp)
// Setup Switch.
p2pLogger := logger.With("module", "p2p")
sw := createSwitch(
config, transport, p2pMetrics, peerFilters, mempoolReactor, bcReactor,
consensusReactor, evidenceReactor, nodeInfo, nodeKey, p2pLogger,
)
err = sw.AddPersistentPeers(splitAndTrimEmpty(config.P2P.PersistentPeers, ",", " "))
if err != nil {
return nil, errors.Wrap(err, "could not add peers from persistent_peers field")
}
addrBook, err := createAddrBookAndSetOnSwitch(config, sw, p2pLogger, nodeKey)
if err != nil {
return nil, errors.Wrap(err, "could not create addrbook")
}
// Optionally, start the pex reactor
//
// TODO:
//
// We need to set Seeds and PersistentPeers on the switch,
// since it needs to be able to use these (and their DNS names)
// even if the PEX is off. We can include the DNS name in the NetAddress,
// but it would still be nice to have a clear list of the current "PersistentPeers"
// somewhere that we can return with net_info.
//
// If PEX is on, it should handle dialing the seeds. Otherwise the switch does it.
// Note we currently use the addrBook regardless at least for AddOurAddress
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
var pexReactor *pex.PEXReactor
if config.P2P.PexReactor {
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
pexReactor = createPEXReactorAndAddToSwitch(addrBook, config, sw, logger)
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
if config.ProfListenAddress != "" {
go func() {
logger.Error("Profile server", "err", http.ListenAndServe(config.ProfListenAddress, nil))
}()
}
node := &Node{
config: config,
genesisDoc: genDoc,
privValidator: privValidator,
transport: transport,
sw: sw,
addrBook: addrBook,
nodeInfo: nodeInfo,
nodeKey: nodeKey,
stateDB: stateDB,
blockStore: blockStore,
bcReactor: bcReactor,
mempoolReactor: mempoolReactor,
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
mempool: mempool,
consensusState: consensusState,
consensusReactor: consensusReactor,
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
pexReactor: pexReactor,
evidencePool: evidencePool,
proxyApp: proxyApp,
txIndexer: txIndexer,
indexerService: indexerService,
eventBus: eventBus,
}
node.BaseService = *cmn.NewBaseService(logger, "Node", node)
for _, option := range options {
option(node)
}
return node, nil
}
// OnStart starts the Node. It implements cmn.Service.
func (n *Node) OnStart() error {
now := tmtime.Now()
genTime := n.genesisDoc.GenesisTime
if genTime.After(now) {
n.Logger.Info("Genesis time is in the future. Sleeping until then...", "genTime", genTime)
time.Sleep(genTime.Sub(now))
}
// Add private IDs to addrbook to block those peers being added
n.addrBook.AddPrivateIDs(splitAndTrimEmpty(n.config.P2P.PrivatePeerIDs, ",", " "))
// Start the RPC server before the P2P server
// so we can eg. receive txs for the first block
if n.config.RPC.ListenAddress != "" {
listeners, err := n.startRPC()
if err != nil {
return err
}
n.rpcListeners = listeners
}
if n.config.Instrumentation.Prometheus &&
n.config.Instrumentation.PrometheusListenAddr != "" {
n.prometheusSrv = n.startPrometheusServer(n.config.Instrumentation.PrometheusListenAddr)
}
// Start the transport.
addr, err := p2p.NewNetAddressString(p2p.IDAddressString(n.nodeKey.ID(), n.config.P2P.ListenAddress))
if err != nil {
return err
}
if err := n.transport.Listen(*addr); err != nil {
return err
}
n.isListening = true
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
if n.config.Mempool.WalEnabled() {
n.mempool.InitWAL() // no need to have the mempool wal during tests
}
// Start the switch (the P2P server).
err = n.sw.Start()
if err != nil {
return err
}
// Always connect to persistent peers
err = n.sw.DialPeersAsync(splitAndTrimEmpty(n.config.P2P.PersistentPeers, ",", " "))
if err != nil {
return errors.Wrap(err, "could not dial peers from persistent_peers field")
}
return nil
}
// OnStop stops the Node. It implements cmn.Service.
func (n *Node) OnStop() {
n.BaseService.OnStop()
n.Logger.Info("Stopping Node")
// first stop the non-reactor services
n.eventBus.Stop()
n.indexerService.Stop()
// now stop the reactors
n.sw.Stop()
// stop mempool WAL
if n.config.Mempool.WalEnabled() {
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
n.mempool.CloseWAL()
}
if err := n.transport.Close(); err != nil {
n.Logger.Error("Error closing transport", "err", err)
}
n.isListening = false
// finally stop the listeners / external services
for _, l := range n.rpcListeners {
n.Logger.Info("Closing rpc listener", "listener", l)
if err := l.Close(); err != nil {
n.Logger.Error("Error closing listener", "listener", l, "err", err)
}
}
if pvsc, ok := n.privValidator.(cmn.Service); ok {
pvsc.Stop()
}
if n.prometheusSrv != nil {
if err := n.prometheusSrv.Shutdown(context.Background()); err != nil {
// Error from closing listeners, or context timeout:
n.Logger.Error("Prometheus HTTP server Shutdown", "err", err)
}
}
}
// ConfigureRPC sets all variables in rpccore so they will serve
// rpc calls from this node
func (n *Node) ConfigureRPC() {
rpccore.SetStateDB(n.stateDB)
rpccore.SetBlockStore(n.blockStore)
rpccore.SetConsensusState(n.consensusState)
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
rpccore.SetMempool(n.mempool)
rpccore.SetEvidencePool(n.evidencePool)
rpccore.SetP2PPeers(n.sw)
rpccore.SetP2PTransport(n)
pubKey := n.privValidator.GetPubKey()
rpccore.SetPubKey(pubKey)
rpccore.SetGenesisDoc(n.genesisDoc)
rpccore.SetAddrBook(n.addrBook)
rpccore.SetProxyAppQuery(n.proxyApp.Query())
rpccore.SetTxIndexer(n.txIndexer)
rpccore.SetConsensusReactor(n.consensusReactor)
rpccore.SetEventBus(n.eventBus)
rpccore.SetLogger(n.Logger.With("module", "rpc"))
limit number of /subscribe clients and queries per client (#3269) * limit number of /subscribe clients and queries per client Add the following config variables (under [rpc] section): * max_subscription_clients * max_subscriptions_per_client * timeout_broadcast_tx_commit Fixes #2826 new HTTPClient interface for subscriptions finalize HTTPClient events interface remove EventSubscriber fix data race ``` WARNING: DATA RACE Read at 0x00c000a36060 by goroutine 129: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe.func1() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:168 +0x1f0 Previous write at 0x00c000a36060 by goroutine 132: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:191 +0x4e0 github.com/tendermint/tendermint/rpc/client.WaitForOneEvent() /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 Goroutine 129 (running) created at: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:164 +0x4b7 github.com/tendermint/tendermint/rpc/client.WaitForOneEvent() /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 Goroutine 132 (running) created at: testing.(*T).Run() /usr/local/go/src/testing/testing.go:878 +0x659 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:119 +0x186 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 ================== ``` lite client works (tested manually) godoc comments httpclient: do not close the out channel use TimeoutBroadcastTxCommit no timeout for unsubscribe but 1s Local (5s HTTP) timeout for resubscribe format code change Subscribe#out cap to 1 and replace config vars with RPCConfig TimeoutBroadcastTxCommit can't be greater than rpcserver.WriteTimeout rpc: Context as first parameter to all functions reformat code fixes after my own review fixes after Ethan's review add test stubs fix config.toml * fixes after manual testing - rpc: do not recommend to use BroadcastTxCommit because it's slow and wastes Tendermint resources (pubsub) - rpc: better error in Subscribe and BroadcastTxCommit - HTTPClient: do not resubscribe if err = ErrAlreadySubscribed * fixes after Ismail's review * Update rpc/grpc/grpc_test.go Co-Authored-By: melekes <anton.kalyaev@gmail.com>
2019-03-11 22:45:58 +04:00
rpccore.SetConfig(*n.config.RPC)
}
func (n *Node) startRPC() ([]net.Listener, error) {
n.ConfigureRPC()
listenAddrs := splitAndTrimEmpty(n.config.RPC.ListenAddress, ",", " ")
coreCodec := amino.NewCodec()
ctypes.RegisterAmino(coreCodec)
if n.config.RPC.Unsafe {
rpccore.AddUnsafeRoutes()
}
config := rpcserver.DefaultConfig()
config.MaxBodyBytes = n.config.RPC.MaxBodyBytes
config.MaxHeaderBytes = n.config.RPC.MaxHeaderBytes
config.MaxOpenConnections = n.config.RPC.MaxOpenConnections
// If necessary adjust global WriteTimeout to ensure it's greater than
// TimeoutBroadcastTxCommit.
// See https://github.com/tendermint/tendermint/issues/3435
if config.WriteTimeout <= n.config.RPC.TimeoutBroadcastTxCommit {
config.WriteTimeout = n.config.RPC.TimeoutBroadcastTxCommit + 1*time.Second
}
// we may expose the rpc over both a unix and tcp socket
listeners := make([]net.Listener, len(listenAddrs))
for i, listenAddr := range listenAddrs {
mux := http.NewServeMux()
rpcLogger := n.Logger.With("module", "rpc-server")
limit number of /subscribe clients and queries per client (#3269) * limit number of /subscribe clients and queries per client Add the following config variables (under [rpc] section): * max_subscription_clients * max_subscriptions_per_client * timeout_broadcast_tx_commit Fixes #2826 new HTTPClient interface for subscriptions finalize HTTPClient events interface remove EventSubscriber fix data race ``` WARNING: DATA RACE Read at 0x00c000a36060 by goroutine 129: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe.func1() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:168 +0x1f0 Previous write at 0x00c000a36060 by goroutine 132: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:191 +0x4e0 github.com/tendermint/tendermint/rpc/client.WaitForOneEvent() /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 Goroutine 129 (running) created at: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:164 +0x4b7 github.com/tendermint/tendermint/rpc/client.WaitForOneEvent() /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 Goroutine 132 (running) created at: testing.(*T).Run() /usr/local/go/src/testing/testing.go:878 +0x659 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:119 +0x186 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 ================== ``` lite client works (tested manually) godoc comments httpclient: do not close the out channel use TimeoutBroadcastTxCommit no timeout for unsubscribe but 1s Local (5s HTTP) timeout for resubscribe format code change Subscribe#out cap to 1 and replace config vars with RPCConfig TimeoutBroadcastTxCommit can't be greater than rpcserver.WriteTimeout rpc: Context as first parameter to all functions reformat code fixes after my own review fixes after Ethan's review add test stubs fix config.toml * fixes after manual testing - rpc: do not recommend to use BroadcastTxCommit because it's slow and wastes Tendermint resources (pubsub) - rpc: better error in Subscribe and BroadcastTxCommit - HTTPClient: do not resubscribe if err = ErrAlreadySubscribed * fixes after Ismail's review * Update rpc/grpc/grpc_test.go Co-Authored-By: melekes <anton.kalyaev@gmail.com>
2019-03-11 22:45:58 +04:00
wmLogger := rpcLogger.With("protocol", "websocket")
wm := rpcserver.NewWebsocketManager(rpccore.Routes, coreCodec,
rpcserver.OnDisconnect(func(remoteAddr string) {
err := n.eventBus.UnsubscribeAll(context.Background(), remoteAddr)
if err != nil && err != tmpubsub.ErrSubscriptionNotFound {
wmLogger.Error("Failed to unsubscribe addr from events", "addr", remoteAddr, "err", err)
}
}),
rpcserver.ReadLimit(config.MaxBodyBytes),
)
limit number of /subscribe clients and queries per client (#3269) * limit number of /subscribe clients and queries per client Add the following config variables (under [rpc] section): * max_subscription_clients * max_subscriptions_per_client * timeout_broadcast_tx_commit Fixes #2826 new HTTPClient interface for subscriptions finalize HTTPClient events interface remove EventSubscriber fix data race ``` WARNING: DATA RACE Read at 0x00c000a36060 by goroutine 129: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe.func1() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:168 +0x1f0 Previous write at 0x00c000a36060 by goroutine 132: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:191 +0x4e0 github.com/tendermint/tendermint/rpc/client.WaitForOneEvent() /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 Goroutine 129 (running) created at: github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe() /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:164 +0x4b7 github.com/tendermint/tendermint/rpc/client.WaitForOneEvent() /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 Goroutine 132 (running) created at: testing.(*T).Run() /usr/local/go/src/testing/testing.go:878 +0x659 github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync() /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:119 +0x186 testing.tRunner() /usr/local/go/src/testing/testing.go:827 +0x162 ================== ``` lite client works (tested manually) godoc comments httpclient: do not close the out channel use TimeoutBroadcastTxCommit no timeout for unsubscribe but 1s Local (5s HTTP) timeout for resubscribe format code change Subscribe#out cap to 1 and replace config vars with RPCConfig TimeoutBroadcastTxCommit can't be greater than rpcserver.WriteTimeout rpc: Context as first parameter to all functions reformat code fixes after my own review fixes after Ethan's review add test stubs fix config.toml * fixes after manual testing - rpc: do not recommend to use BroadcastTxCommit because it's slow and wastes Tendermint resources (pubsub) - rpc: better error in Subscribe and BroadcastTxCommit - HTTPClient: do not resubscribe if err = ErrAlreadySubscribed * fixes after Ismail's review * Update rpc/grpc/grpc_test.go Co-Authored-By: melekes <anton.kalyaev@gmail.com>
2019-03-11 22:45:58 +04:00
wm.SetLogger(wmLogger)
mux.HandleFunc("/websocket", wm.WebsocketHandler)
rpcserver.RegisterRPCFuncs(mux, rpccore.Routes, coreCodec, rpcLogger)
listener, err := rpcserver.Listen(
listenAddr,
config,
)
if err != nil {
return nil, err
}
2018-11-14 15:47:41 +03:00
var rootHandler http.Handler = mux
if n.config.RPC.IsCorsEnabled() {
corsMiddleware := cors.New(cors.Options{
AllowedOrigins: n.config.RPC.CORSAllowedOrigins,
AllowedMethods: n.config.RPC.CORSAllowedMethods,
AllowedHeaders: n.config.RPC.CORSAllowedHeaders,
})
rootHandler = corsMiddleware.Handler(mux)
}
if n.config.RPC.IsTLSEnabled() {
go rpcserver.StartHTTPAndTLSServer(
listener,
rootHandler,
n.config.RPC.CertFile(),
n.config.RPC.KeyFile(),
rpcLogger,
config,
)
} else {
go rpcserver.StartHTTPServer(
listener,
rootHandler,
rpcLogger,
config,
)
}
2018-11-14 15:47:41 +03:00
listeners[i] = listener
}
// we expose a simplified api over grpc for convenience to app devs
grpcListenAddr := n.config.RPC.GRPCListenAddress
if grpcListenAddr != "" {
config := rpcserver.DefaultConfig()
config.MaxOpenConnections = n.config.RPC.MaxOpenConnections
listener, err := rpcserver.Listen(grpcListenAddr, config)
if err != nil {
return nil, err
}
go grpccore.StartGRPCServer(listener)
listeners = append(listeners, listener)
}
return listeners, nil
}
// startPrometheusServer starts a Prometheus HTTP server, listening for metrics
// collectors on addr.
func (n *Node) startPrometheusServer(addr string) *http.Server {
srv := &http.Server{
Addr: addr,
Handler: promhttp.InstrumentMetricHandler(
prometheus.DefaultRegisterer, promhttp.HandlerFor(
prometheus.DefaultGatherer,
promhttp.HandlerOpts{MaxRequestsInFlight: n.config.Instrumentation.MaxOpenConnections},
),
),
}
go func() {
if err := srv.ListenAndServe(); err != http.ErrServerClosed {
// Error starting or closing listener:
n.Logger.Error("Prometheus HTTP server ListenAndServe", "err", err)
}
}()
return srv
}
// Switch returns the Node's Switch.
func (n *Node) Switch() *p2p.Switch {
return n.sw
}
// BlockStore returns the Node's BlockStore.
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
func (n *Node) BlockStore() *store.BlockStore {
return n.blockStore
}
// ConsensusState returns the Node's ConsensusState.
func (n *Node) ConsensusState() *cs.ConsensusState {
return n.consensusState
}
// ConsensusReactor returns the Node's ConsensusReactor.
func (n *Node) ConsensusReactor() *cs.ConsensusReactor {
return n.consensusReactor
}
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
// MempoolReactor returns the Node's mempool reactor.
func (n *Node) MempoolReactor() *mempl.Reactor {
return n.mempoolReactor
}
mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface
2019-05-04 10:41:31 +04:00
// Mempool returns the Node's mempool.
func (n *Node) Mempool() mempl.Mempool {
return n.mempool
}
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
// PEXReactor returns the Node's PEXReactor. It returns nil if PEX is disabled.
func (n *Node) PEXReactor() *pex.PEXReactor {
return n.pexReactor
}
// EvidencePool returns the Node's EvidencePool.
func (n *Node) EvidencePool() *evidence.EvidencePool {
return n.evidencePool
}
// EventBus returns the Node's EventBus.
func (n *Node) EventBus() *types.EventBus {
return n.eventBus
}
// PrivValidator returns the Node's PrivValidator.
// XXX: for convenience only!
func (n *Node) PrivValidator() types.PrivValidator {
return n.privValidator
}
// GenesisDoc returns the Node's GenesisDoc.
func (n *Node) GenesisDoc() *types.GenesisDoc {
return n.genesisDoc
}
// ProxyApp returns the Node's AppConns, representing its connections to the ABCI application.
func (n *Node) ProxyApp() proxy.AppConns {
return n.proxyApp
}
// Config returns the Node's config.
func (n *Node) Config() *cfg.Config {
return n.config
}
//------------------------------------------------------------------------------
func (n *Node) Listeners() []string {
return []string{
fmt.Sprintf("Listener(@%v)", n.config.P2P.ExternalAddress),
}
}
func (n *Node) IsListening() bool {
return n.isListening
}
// NodeInfo returns the Node's Info from the Switch.
func (n *Node) NodeInfo() p2p.NodeInfo {
return n.nodeInfo
}
func makeNodeInfo(
config *cfg.Config,
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
nodeKey *p2p.NodeKey,
txIndexer txindex.TxIndexer,
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
genDoc *types.GenesisDoc,
state sm.State,
) (p2p.NodeInfo, error) {
txIndexerStatus := "on"
if _, ok := txIndexer.(*null.TxIndex); ok {
txIndexerStatus = "off"
}
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
var bcChannel byte
switch config.FastSync.Version {
case "v0":
bcChannel = bcv0.BlockchainChannel
case "v1":
bcChannel = bcv1.BlockchainChannel
default:
return nil, fmt.Errorf("unknown fastsync version %s", config.FastSync.Version)
}
nodeInfo := p2p.DefaultNodeInfo{
node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil
2019-04-26 08:05:39 -04:00
ProtocolVersion: p2p.NewProtocolVersion(
version.P2PProtocol, // global
state.Version.Consensus.Block,
state.Version.Consensus.App,
),
ID_: nodeKey.ID(),
Network: genDoc.ChainID,
Version: version.TMCoreSemVer,
Channels: []byte{
blockchain: Reorg reactor (#3561) * go routines in blockchain reactor * Added reference to the go routine diagram * Initial commit * cleanup * Undo testing_logger change, committed by mistake * Fix the test loggers * pulled some fsm code into pool.go * added pool tests * changes to the design added block requests under peer moved the request trigger in the reactor poolRoutine, triggered now by a ticker in general moved everything required for making block requests smarter in the poolRoutine added a simple map of heights to keep track of what will need to be requested next added a few more tests * send errors to FSM in a different channel than blocks send errors (RemovePeer) from switch on a different channel than the one receiving blocks renamed channels added more pool tests * more pool tests * lint errors * more tests * more tests * switch fast sync to new implementation * fixed data race in tests * cleanup * finished fsm tests * address golangci comments :) * address golangci comments :) * Added timeout on next block needed to advance * updating docs and cleanup * fix issue in test from previous cleanup * cleanup * Added termination scenarios, tests and more cleanup * small fixes to adr, comments and cleanup * Fix bug in sendRequest() If we tried to send a request to a peer not present in the switch, a missing continue statement caused the request to be blackholed in a peer that was removed and never retried. While this bug was manifesting, the reactor kept asking for other blocks that would be stored and never consumed. Added the number of unconsumed blocks in the math for requesting blocks ahead of current processing height so eventually there will be no more blocks requested until the already received ones are consumed. * remove bpPeer's didTimeout field * Use distinct err codes for peer timeout and FSM timeouts * Don't allow peers to update with lower height * review comments from Ethan and Zarko * some cleanup, renaming, comments * Move block execution in separate goroutine * Remove pool's numPending * review comments * fix lint, remove old blockchain reactor and duplicates in fsm tests * small reorg around peer after review comments * add the reactor spec * verify block only once * review comments * change to int for max number of pending requests * cleanup and godoc * Add configuration flag fast sync version * golangci fixes * fix config template * move both reactor versions under blockchain * cleanup, golint, renaming stuff * updated documentation, fixed more golint warnings * integrate with behavior package * sync with master * gofmt * add changelog_pending entry * move to improvments * suggestion to changelog entry
2019-07-23 10:58:52 +02:00
bcChannel,
cs.StateChannel, cs.DataChannel, cs.VoteChannel, cs.VoteSetBitsChannel,
mempl.MempoolChannel,
evidence.EvidenceChannel,
},
Moniker: config.Moniker,
Other: p2p.DefaultNodeInfoOther{
TxIndex: txIndexerStatus,
RPCAddress: config.RPC.ListenAddress,
},
}
if config.P2P.PexReactor {
nodeInfo.Channels = append(nodeInfo.Channels, pex.PexChannel)
}
lAddr := config.P2P.ExternalAddress
if lAddr == "" {
lAddr = config.P2P.ListenAddress
}
nodeInfo.ListenAddr = lAddr
err := nodeInfo.Validate()
return nodeInfo, err
}
//------------------------------------------------------------------------------
var (
genesisDocKey = []byte("genesisDoc")
)
// LoadStateFromDBOrGenesisDocProvider attempts to load the state from the
// database, or creates one using the given genesisDocProvider and persists the
// result to the database. On success this also returns the genesis doc loaded
// through the given provider.
func LoadStateFromDBOrGenesisDocProvider(stateDB dbm.DB, genesisDocProvider GenesisDocProvider) (sm.State, *types.GenesisDoc, error) {
// Get genesis doc
genDoc, err := loadGenesisDoc(stateDB)
if err != nil {
genDoc, err = genesisDocProvider()
if err != nil {
return sm.State{}, nil, err
}
// save genesis doc to prevent a certain class of user errors (e.g. when it
// was changed, accidentally or not). Also good for audit trail.
saveGenesisDoc(stateDB, genDoc)
}
state, err := sm.LoadStateFromDBOrGenesisDoc(stateDB, genDoc)
if err != nil {
return sm.State{}, nil, err
}
return state, genDoc, nil
}
// panics if failed to unmarshal bytes
func loadGenesisDoc(db dbm.DB) (*types.GenesisDoc, error) {
b := db.Get(genesisDocKey)
if len(b) == 0 {
return nil, errors.New("Genesis doc not found")
}
var genDoc *types.GenesisDoc
err := cdc.UnmarshalJSON(b, &genDoc)
if err != nil {
panic(fmt.Sprintf("Failed to load genesis doc due to unmarshaling error: %v (bytes: %X)", err, b))
}
return genDoc, nil
}
// panics if failed to marshal the given genesis document
func saveGenesisDoc(db dbm.DB, genDoc *types.GenesisDoc) {
b, err := cdc.MarshalJSON(genDoc)
if err != nil {
panic(fmt.Sprintf("Failed to save genesis doc due to marshaling error: %v", err))
}
db.SetSync(genesisDocKey, b)
}
func createAndStartPrivValidatorSocketClient(
listenAddr string,
logger log.Logger,
) (types.PrivValidator, error) {
privval: refactor Remote signers (#3370) This PR is related to #3107 and a continuation of #3351 It is important to emphasise that in the privval original design, client/server and listening/dialing roles are inverted and do not follow a conventional interaction. Given two hosts A and B: Host A is listener/client Host B is dialer/server (contains the secret key) When A requires a signature, it needs to wait for B to dial in before it can issue a request. A only accepts a single connection and any failure leads to dropping the connection and waiting for B to reconnect. The original rationale behind this design was based on security. Host B only allows outbound connections to a list of whitelisted hosts. It is not possible to reach B unless B dials in. There are no listening/open ports in B. This PR results in the following changes: Refactors ping/heartbeat to avoid previously existing race conditions. Separates transport (dialer/listener) from signing (client/server) concerns to simplify workflow. Unifies and abstracts away the differences between unix and tcp sockets. A single signer endpoint implementation unifies connection handling code (read/write/close/connection obj) The signer request handler (server side) is customizable to increase testability. Updates and extends unit tests A high level overview of the classes is as follows: Transport (endpoints): The following classes take care of establishing a connection SignerDialerEndpoint SignerListeningEndpoint SignerEndpoint groups common functionality (read/write/timeouts/etc.) Signing (client/server): The following classes take care of exchanging request/responses SignerClient SignerServer This PR also closes #3601 Commits: * refactoring - work in progress * reworking unit tests * Encapsulating and fixing unit tests * Improve tests * Clean up * Fix/improve unit tests * clean up tests * Improving service endpoint * fixing unit test * fix linter issues * avoid invalid cache values (improve later?) * complete implementation * wip * improved connection loop * Improve reconnections + fixing unit tests * addressing comments * small formatting changes * clean up * Update node/node.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * check during initialization * dropping connecting when writing fails * removing break * use t.log instead * unifying and using cmn.GetFreePort() * review fixes * reordering and unifying drop connection * closing instead of signalling * refactored service loop * removed superfluous brackets * GetPubKey can return errors * Revert "GetPubKey can return errors" This reverts commit 68c06f19b4650389d7e5ab1659b318889028202c. * adding entry to changelog * Update CHANGELOG_PENDING.md Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * updating node.go * review fixes * fixes linter * fixing unit test * small fixes in comments * addressing review comments * addressing review comments 2 * reverting suggestion * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * do not expose brokenSignerDialerEndpoint * clean up logging * unifying methods shorten test time signer also drops * reenabling pings * improving testability + unit test * fixing go fmt + unit test * remove unused code * Addressing review comments * simplifying connection workflow * fix linter/go import issue * using base service quit * updating comment * Simplifying design + adjusting names * fixing linter issues * refactoring test harness + fixes * Addressing review comments * cleaning up * adding additional error check
2019-08-05 17:09:10 +02:00
pve, err := privval.NewSignerListener(listenAddr, logger)
Close and retry a RemoteSigner on err (#2923) * Close and recreate a RemoteSigner on err * Update changelog * Address Anton's comments / suggestions: - update changelog - restart TCPVal - shut down on `ErrUnexpectedResponse` * re-init remote signer client with fresh connection if Ping fails - add/update TODOs in secret connection - rename tcp.go -> tcp_client.go, same with ipc to clarify their purpose * account for `conn returned by waitConnection can be `nil` - also add TODO about RemoteSigner conn field * Tests for retrying: IPC / TCP - shorter info log on success - set conn and use it in tests to close conn * Tests for retrying: IPC / TCP - shorter info log on success - set conn and use it in tests to close conn - add rwmutex for conn field in IPC * comments and doc.go * fix ipc tests. fixes #2677 * use constants for tests * cleanup some error statements * fixes #2784, race in tests * remove print statement * minor fixes from review * update comment on sts spec * cosmetics * p2p/conn: add failing tests * p2p/conn: make SecretConnection thread safe * changelog * IPCVal signer refactor - use a .reset() method - don't use embedded RemoteSignerClient - guard RemoteSignerClient with mutex - drop the .conn - expose Close() on RemoteSignerClient * apply IPCVal refactor to TCPVal * remove mtx from RemoteSignerClient * consolidate IPCVal and TCPVal, fixes #3104 - done in tcp_client.go - now called SocketVal - takes a listener in the constructor - make tcpListener and unixListener contain all the differences * delete ipc files * introduce unix and tcp dialer for RemoteSigner * rename files - drop tcp_ prefix - rename priv_validator.go to file.go * bring back listener options * fix node * fix priv_val_server * fix node test * minor cleanup and comments
2019-01-13 20:31:31 +01:00
if err != nil {
privval: refactor Remote signers (#3370) This PR is related to #3107 and a continuation of #3351 It is important to emphasise that in the privval original design, client/server and listening/dialing roles are inverted and do not follow a conventional interaction. Given two hosts A and B: Host A is listener/client Host B is dialer/server (contains the secret key) When A requires a signature, it needs to wait for B to dial in before it can issue a request. A only accepts a single connection and any failure leads to dropping the connection and waiting for B to reconnect. The original rationale behind this design was based on security. Host B only allows outbound connections to a list of whitelisted hosts. It is not possible to reach B unless B dials in. There are no listening/open ports in B. This PR results in the following changes: Refactors ping/heartbeat to avoid previously existing race conditions. Separates transport (dialer/listener) from signing (client/server) concerns to simplify workflow. Unifies and abstracts away the differences between unix and tcp sockets. A single signer endpoint implementation unifies connection handling code (read/write/close/connection obj) The signer request handler (server side) is customizable to increase testability. Updates and extends unit tests A high level overview of the classes is as follows: Transport (endpoints): The following classes take care of establishing a connection SignerDialerEndpoint SignerListeningEndpoint SignerEndpoint groups common functionality (read/write/timeouts/etc.) Signing (client/server): The following classes take care of exchanging request/responses SignerClient SignerServer This PR also closes #3601 Commits: * refactoring - work in progress * reworking unit tests * Encapsulating and fixing unit tests * Improve tests * Clean up * Fix/improve unit tests * clean up tests * Improving service endpoint * fixing unit test * fix linter issues * avoid invalid cache values (improve later?) * complete implementation * wip * improved connection loop * Improve reconnections + fixing unit tests * addressing comments * small formatting changes * clean up * Update node/node.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * check during initialization * dropping connecting when writing fails * removing break * use t.log instead * unifying and using cmn.GetFreePort() * review fixes * reordering and unifying drop connection * closing instead of signalling * refactored service loop * removed superfluous brackets * GetPubKey can return errors * Revert "GetPubKey can return errors" This reverts commit 68c06f19b4650389d7e5ab1659b318889028202c. * adding entry to changelog * Update CHANGELOG_PENDING.md Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * updating node.go * review fixes * fixes linter * fixing unit test * small fixes in comments * addressing review comments * addressing review comments 2 * reverting suggestion * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * do not expose brokenSignerDialerEndpoint * clean up logging * unifying methods shorten test time signer also drops * reenabling pings * improving testability + unit test * fixing go fmt + unit test * remove unused code * Addressing review comments * simplifying connection workflow * fix linter/go import issue * using base service quit * updating comment * Simplifying design + adjusting names * fixing linter issues * refactoring test harness + fixes * Addressing review comments * cleaning up * adding additional error check
2019-08-05 17:09:10 +02:00
return nil, errors.Wrap(err, "failed to start private validator")
}
privval: refactor Remote signers (#3370) This PR is related to #3107 and a continuation of #3351 It is important to emphasise that in the privval original design, client/server and listening/dialing roles are inverted and do not follow a conventional interaction. Given two hosts A and B: Host A is listener/client Host B is dialer/server (contains the secret key) When A requires a signature, it needs to wait for B to dial in before it can issue a request. A only accepts a single connection and any failure leads to dropping the connection and waiting for B to reconnect. The original rationale behind this design was based on security. Host B only allows outbound connections to a list of whitelisted hosts. It is not possible to reach B unless B dials in. There are no listening/open ports in B. This PR results in the following changes: Refactors ping/heartbeat to avoid previously existing race conditions. Separates transport (dialer/listener) from signing (client/server) concerns to simplify workflow. Unifies and abstracts away the differences between unix and tcp sockets. A single signer endpoint implementation unifies connection handling code (read/write/close/connection obj) The signer request handler (server side) is customizable to increase testability. Updates and extends unit tests A high level overview of the classes is as follows: Transport (endpoints): The following classes take care of establishing a connection SignerDialerEndpoint SignerListeningEndpoint SignerEndpoint groups common functionality (read/write/timeouts/etc.) Signing (client/server): The following classes take care of exchanging request/responses SignerClient SignerServer This PR also closes #3601 Commits: * refactoring - work in progress * reworking unit tests * Encapsulating and fixing unit tests * Improve tests * Clean up * Fix/improve unit tests * clean up tests * Improving service endpoint * fixing unit test * fix linter issues * avoid invalid cache values (improve later?) * complete implementation * wip * improved connection loop * Improve reconnections + fixing unit tests * addressing comments * small formatting changes * clean up * Update node/node.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * check during initialization * dropping connecting when writing fails * removing break * use t.log instead * unifying and using cmn.GetFreePort() * review fixes * reordering and unifying drop connection * closing instead of signalling * refactored service loop * removed superfluous brackets * GetPubKey can return errors * Revert "GetPubKey can return errors" This reverts commit 68c06f19b4650389d7e5ab1659b318889028202c. * adding entry to changelog * Update CHANGELOG_PENDING.md Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_client.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_dialer_endpoint.go Co-Authored-By: jleni <juan.leni@zondax.ch> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: jleni <juan.leni@zondax.ch> * updating node.go * review fixes * fixes linter * fixing unit test * small fixes in comments * addressing review comments * addressing review comments 2 * reverting suggestion * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_client_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update privval/signer_listener_endpoint_test.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * do not expose brokenSignerDialerEndpoint * clean up logging * unifying methods shorten test time signer also drops * reenabling pings * improving testability + unit test * fixing go fmt + unit test * remove unused code * Addressing review comments * simplifying connection workflow * fix linter/go import issue * using base service quit * updating comment * Simplifying design + adjusting names * fixing linter issues * refactoring test harness + fixes * Addressing review comments * cleaning up * adding additional error check
2019-08-05 17:09:10 +02:00
pvsc, err := privval.NewSignerClient(pve)
if err != nil {
return nil, errors.Wrap(err, "failed to start private validator")
}
return pvsc, nil
}
// splitAndTrimEmpty slices s into all subslices separated by sep and returns a
// slice of the string s with all leading and trailing Unicode code points
// contained in cutset removed. If sep is empty, SplitAndTrim splits after each
// UTF-8 sequence. First part is equivalent to strings.SplitN with a count of
// -1. also filter out empty strings, only return non-empty strings.
func splitAndTrimEmpty(s, sep, cutset string) []string {
if s == "" {
return []string{}
}
spl := strings.Split(s, sep)
nonEmptyStrings := make([]string, 0, len(spl))
for i := 0; i < len(spl); i++ {
element := strings.Trim(spl[i], cutset)
if element != "" {
nonEmptyStrings = append(nonEmptyStrings, element)
}
}
return nonEmptyStrings
}