lotus/lib/consensus/raft/consensus.go
2022-11-25 11:43:26 +00:00

513 lines
14 KiB
Go

// Package raft implements a Consensus component for IPFS Cluster which uses
// Raft (go-libp2p-raft).
package consensus
import (
"bytes"
"context"
"errors"
"fmt"
"sort"
"time"
"github.com/google/uuid"
"golang.org/x/exp/slices"
addr "github.com/filecoin-project/go-address"
"github.com/filecoin-project/lotus/api"
"github.com/filecoin-project/lotus/chain/messagepool"
"github.com/filecoin-project/lotus/chain/types"
"github.com/filecoin-project/lotus/lib/addrutil"
"github.com/filecoin-project/lotus/node/repo"
//ds "github.com/ipfs/go-datastore"
logging "github.com/ipfs/go-log/v2"
consensus "github.com/libp2p/go-libp2p-consensus"
rpc "github.com/libp2p/go-libp2p-gorpc"
libp2praft "github.com/libp2p/go-libp2p-raft"
host "github.com/libp2p/go-libp2p/core/host"
peer "github.com/libp2p/go-libp2p/core/peer"
)
var logger = logging.Logger("raft")
type RaftState struct {
NonceMap api.NonceMapType
MsgUuids api.MsgUuidMapType
// TODO: add comment explaining why this is needed
// We need a reference to the messagepool in the raft state in order to
// sync messages that have been sent by the leader node
// Miner calls StateWaitMsg after MpoolPushMessage to check if the message has
// landed on chain. This check requires the message be stored in the local chainstore
// If a leadernode goes down after sending a message to the chain and is replaced by
// another node, the other node needs to have this message in its chainstore for the
// above check to succeed.
// This is because the miner only stores signed CIDs but the message received from in a
// block will be unsigned (for BLS). Hence, the process relies on the node to store the
// signed message which holds a copy of the unsigned message to properly perform all the
// needed checks
Mpool *messagepool.MessagePool
}
func newRaftState(mpool *messagepool.MessagePool) *RaftState {
return &RaftState{
NonceMap: make(map[addr.Address]uint64),
MsgUuids: make(map[uuid.UUID]*types.SignedMessage),
Mpool: mpool,
}
}
type ConsensusOp struct {
Nonce uint64 `codec:"nonce,omitempty"`
Uuid uuid.UUID `codec:"uuid,omitempty"`
Addr addr.Address `codec:"addr,omitempty"`
SignedMsg *types.SignedMessage `codec:"signedMsg,omitempty"`
}
func (c ConsensusOp) ApplyTo(state consensus.State) (consensus.State, error) {
s := state.(*RaftState)
s.NonceMap[c.Addr] = c.Nonce
if c.SignedMsg != nil {
// Deep copy to tmp
var buffer bytes.Buffer
err := c.SignedMsg.MarshalCBOR(&buffer)
if err != nil {
return nil, err
}
tmp, err := types.DecodeSignedMessage(buffer.Bytes())
if err != nil {
return nil, err
}
s.MsgUuids[c.Uuid] = tmp
_, err = s.Mpool.Push(context.TODO(), tmp, false)
// Since this is only meant to keep messages in sync, ignore any error which
// shows the message already exists in the mpool
if err != nil && !api.ErrorIsIn(err, []error{messagepool.ErrExistingNonce}) {
return nil, err
}
}
return s, nil
}
var _ consensus.Op = &ConsensusOp{}
// Consensus handles the work of keeping a shared-state between
// the peers of a Lotus Cluster, as well as modifying that state and
// applying any updates in a thread-safe manner.
type Consensus struct {
ctx context.Context
cancel func()
config *ClusterRaftConfig
host host.Host
consensus consensus.OpLogConsensus
actor consensus.Actor
raft *raftWrapper
state *RaftState
RpcClient *rpc.Client
rpcReady chan struct{}
readyCh chan struct{}
peerSet []peer.ID
repo repo.LockedRepo
}
// NewConsensus builds a new ClusterConsensus component using Raft.
//
// Raft saves state snapshots regularly and persists log data in a bolt
// datastore. Therefore, unless memory usage is a concern, it is recommended
// to use an in-memory go-datastore as store parameter.
//
// The staging parameter controls if the Raft peer should start in
// staging mode (used when joining a new Raft peerset with other peers).
func NewConsensus(host host.Host, cfg *ClusterRaftConfig, mpool *messagepool.MessagePool, repo repo.LockedRepo, staging bool) (*Consensus, error) {
err := ValidateConfig(cfg)
if err != nil {
return nil, err
}
ctx, cancel := context.WithCancel(context.Background())
logger.Debug("starting Consensus and waiting for a leader...")
state := newRaftState(mpool)
consensus := libp2praft.NewOpLog(state, &ConsensusOp{})
raft, err := newRaftWrapper(host, cfg, consensus.FSM(), repo, staging)
if err != nil {
logger.Error("error creating raft: ", err)
cancel()
return nil, err
}
actor := libp2praft.NewActor(raft.raft)
consensus.SetActor(actor)
peers := []peer.ID{}
addrInfos, err := addrutil.ParseAddresses(ctx, cfg.InitPeerset)
if err != nil {
logger.Error("error parsing addresses: ", err)
cancel()
return nil, err
}
for _, addrInfo := range addrInfos {
peers = append(peers, addrInfo.ID)
// Add peer to address book
host.Peerstore().AddAddrs(addrInfo.ID, addrInfo.Addrs, time.Hour*100)
}
cc := &Consensus{
ctx: ctx,
cancel: cancel,
config: cfg,
host: host,
consensus: consensus,
actor: actor,
state: state,
raft: raft,
peerSet: peers,
rpcReady: make(chan struct{}, 1),
readyCh: make(chan struct{}, 1),
repo: repo,
}
go cc.finishBootstrap()
return cc, nil
}
// TODO: Merge with NewConsensus and remove the rpcReady chan
func NewConsensusWithRPCClient(staging bool) func(host host.Host,
cfg *ClusterRaftConfig,
rpcClient *rpc.Client,
mpool *messagepool.MessagePool,
repo repo.LockedRepo,
) (*Consensus, error) {
return func(host host.Host, cfg *ClusterRaftConfig, rpcClient *rpc.Client, mpool *messagepool.MessagePool, repo repo.LockedRepo) (*Consensus, error) {
cc, err := NewConsensus(host, cfg, mpool, repo, staging)
if err != nil {
return nil, err
}
cc.RpcClient = rpcClient
cc.rpcReady <- struct{}{}
return cc, nil
}
}
// WaitForSync waits for a leader and for the state to be up to date, then returns.
func (cc *Consensus) WaitForSync(ctx context.Context) error {
leaderCtx, cancel := context.WithTimeout(ctx, cc.config.WaitForLeaderTimeout)
defer cancel()
// 1 - wait for leader
// 2 - wait until we are a Voter
// 3 - wait until last index is applied
// From raft docs:
// once a staging server receives enough log entries to be sufficiently
// caught up to the leader's log, the leader will invoke a membership
// change to change the Staging server to a Voter
// Thus, waiting to be a Voter is a guarantee that we have a reasonable
// up to date state. Otherwise, we might return too early (see
// https://github.com/ipfs-cluster/ipfs-cluster/issues/378)
_, err := cc.raft.WaitForLeader(leaderCtx)
if err != nil {
return errors.New("error waiting for leader: " + err.Error())
}
err = cc.raft.WaitForVoter(ctx)
if err != nil {
return errors.New("error waiting to become a Voter: " + err.Error())
}
err = cc.raft.WaitForUpdates(ctx)
if err != nil {
return errors.New("error waiting for consensus updates: " + err.Error())
}
return nil
}
// waits until there is a consensus leader and syncs the state
// to the tracker. If errors happen, this will return and never
// signal the component as Ready.
func (cc *Consensus) finishBootstrap() {
// wait until we have RPC to perform any actions.
select {
case <-cc.ctx.Done():
return
case <-cc.rpcReady:
}
// Sometimes bootstrap is a no-Op. It only applies when
// no state exists and staging=false.
_, err := cc.raft.Bootstrap()
if err != nil {
return
}
logger.Debugf("Bootstrap finished")
err = cc.WaitForSync(cc.ctx)
if err != nil {
return
}
logger.Debug("Raft state is now up to date")
logger.Debug("consensus ready")
cc.readyCh <- struct{}{}
}
// Shutdown stops the component so it will not process any
// more updates. The underlying consensus is permanently
// shutdown, along with the libp2p transport.
func (cc *Consensus) Shutdown(ctx context.Context) error {
logger.Info("stopping Consensus component")
// Raft Shutdown
err := cc.raft.Shutdown(ctx)
if err != nil {
logger.Error(err)
}
cc.cancel()
close(cc.rpcReady)
return nil
}
// Ready returns a channel which is signaled when the Consensus
// algorithm has finished bootstrapping and is ready to use
func (cc *Consensus) Ready(ctx context.Context) <-chan struct{} {
return cc.readyCh
}
// IsTrustedPeer returns true. In Raft we trust all peers.
func (cc *Consensus) IsTrustedPeer(ctx context.Context, p peer.ID) bool {
return slices.Contains(cc.peerSet, p)
}
// Trust is a no-Op.
func (cc *Consensus) Trust(ctx context.Context, pid peer.ID) error { return nil }
// Distrust is a no-Op.
func (cc *Consensus) Distrust(ctx context.Context, pid peer.ID) error { return nil }
// returns true if the operation was redirected to the leader
// note that if the leader just dissappeared, the rpc call will
// fail because we haven't heard that it's gone.
func (cc *Consensus) RedirectToLeader(method string, arg interface{}, ret interface{}) (bool, error) {
ctx := cc.ctx
var finalErr error
// Retry redirects
for i := 0; i <= cc.config.CommitRetries; i++ {
logger.Debugf("redirect try %d", i)
leader, err := cc.Leader(ctx)
// No leader, wait for one
if err != nil {
logger.Warn("there seems to be no leader. Waiting for one")
rctx, cancel := context.WithTimeout(ctx, cc.config.WaitForLeaderTimeout)
defer cancel()
pidstr, err := cc.raft.WaitForLeader(rctx)
// means we timed out waiting for a leader
// we don't retry in this case
if err != nil {
return false, fmt.Errorf("timed out waiting for leader: %s", err)
}
leader, err = peer.Decode(pidstr)
if err != nil {
return false, err
}
}
logger.Infof("leader: %s, curr host: %s, peerSet: %s", leader, cc.host.ID(), cc.peerSet)
// We are the leader. Do not redirect
if leader == cc.host.ID() {
return false, nil
}
logger.Debugf("redirecting %s to leader: %s", method, leader.Pretty())
finalErr = cc.RpcClient.CallContext(
ctx,
leader,
"Consensus",
method,
arg,
ret,
)
if finalErr != nil {
logger.Errorf("retrying to redirect request to leader: %s", finalErr)
time.Sleep(2 * cc.config.RaftConfig.HeartbeatTimeout)
continue
}
break
}
// We tried to redirect, but something happened
return true, finalErr
}
// commit submits a cc.consensus commit. It retries upon failures.
func (cc *Consensus) Commit(ctx context.Context, op *ConsensusOp) error {
var finalErr error
for i := 0; i <= cc.config.CommitRetries; i++ {
logger.Debugf("attempt #%d: committing %+v", i, op)
// this means we are retrying
if finalErr != nil {
logger.Errorf("retrying upon failed commit (retry %d): %s ",
i, finalErr)
}
// Being here means we are the LEADER. We can commit.
// now commit the changes to our state
_, finalErr = cc.consensus.CommitOp(op)
if finalErr != nil {
goto RETRY
}
RETRY:
time.Sleep(cc.config.CommitRetryDelay)
}
return finalErr
}
// AddPeer adds a new peer to participate in this consensus. It will
// forward the operation to the leader if this is not it.
func (cc *Consensus) AddPeer(ctx context.Context, pid peer.ID) error {
var finalErr error
for i := 0; i <= cc.config.CommitRetries; i++ {
logger.Debugf("attempt #%d: AddPeer %s", i, pid.Pretty())
if finalErr != nil {
logger.Errorf("retrying to add peer. Attempt #%d failed: %s", i, finalErr)
}
ok, err := cc.RedirectToLeader("AddPeer", pid, struct{}{})
if err != nil || ok {
return err
}
// Being here means we are the leader and can commit
finalErr = cc.raft.AddPeer(ctx, pid)
if finalErr != nil {
time.Sleep(cc.config.CommitRetryDelay)
continue
}
logger.Infof("peer added to Raft: %s", pid.Pretty())
break
}
return finalErr
}
// RmPeer removes a peer from this consensus. It will
// forward the operation to the leader if this is not it.
func (cc *Consensus) RmPeer(ctx context.Context, pid peer.ID) error {
var finalErr error
for i := 0; i <= cc.config.CommitRetries; i++ {
logger.Debugf("attempt #%d: RmPeer %s", i, pid.Pretty())
if finalErr != nil {
logger.Errorf("retrying to remove peer. Attempt #%d failed: %s", i, finalErr)
}
ok, err := cc.RedirectToLeader("RmPeer", pid, struct{}{})
if err != nil || ok {
return err
}
// Being here means we are the leader and can commit
finalErr = cc.raft.RemovePeer(ctx, pid.String())
if finalErr != nil {
time.Sleep(cc.config.CommitRetryDelay)
continue
}
logger.Infof("peer removed from Raft: %s", pid.Pretty())
break
}
return finalErr
}
// RaftState retrieves the current consensus RaftState. It may error if no RaftState has
// been agreed upon or the state is not consistent. The returned RaftState is the
// last agreed-upon RaftState known by this node. No writes are allowed, as all
// writes to the shared state should happen through the Consensus component
// methods.
func (cc *Consensus) State(ctx context.Context) (*RaftState, error) {
st, err := cc.consensus.GetLogHead()
if err == libp2praft.ErrNoState {
return newRaftState(nil), nil
}
if err != nil {
return nil, err
}
state, ok := st.(*RaftState)
if !ok {
return nil, errors.New("wrong state type")
}
return state, nil
}
// Leader returns the peerID of the Leader of the
// cluster. It returns an error when there is no leader.
func (cc *Consensus) Leader(ctx context.Context) (peer.ID, error) {
// Note the hard-dependency on raft here...
raftactor := cc.actor.(*libp2praft.Actor)
return raftactor.Leader()
}
// Clean removes the Raft persisted state.
func (cc *Consensus) Clean(ctx context.Context) error {
//return CleanupRaft(cc.config)
return nil
}
//Rollback replaces the current agreed-upon
//state with the state provided. Only the consensus leader
//can perform this operation.
//func (cc *Consensus) Rollback(state RaftState) error {
// // This is unused. It *might* be used for upgrades.
// // There is rather untested magic in libp2p-raft's FSM()
// // to make this possible.
// return cc.consensus.Rollback(state)
//}
// Peers return the current list of peers in the consensus.
// The list will be sorted alphabetically.
func (cc *Consensus) Peers(ctx context.Context) ([]peer.ID, error) {
peers := []peer.ID{}
raftPeers, err := cc.raft.Peers(ctx)
if err != nil {
return nil, fmt.Errorf("cannot retrieve list of peers: %s", err)
}
sort.Strings(raftPeers)
for _, p := range raftPeers {
id, err := peer.Decode(p)
if err != nil {
panic("could not decode peer")
}
peers = append(peers, id)
}
return peers, nil
}
func (cc *Consensus) IsLeader(ctx context.Context) bool {
leader, _ := cc.Leader(ctx)
return leader == cc.host.ID()
}