深入浅出etcd/raft —— 0x02 etcd/raft总体设计

// Node represents a node in a raft cluster.
type Node interface {
	// Tick increments the internal logical clock for the Node by a single tick. Election
	// timeouts and heartbeat timeouts are in units of ticks.
	Tick()
	// Campaign causes the Node to transition to candidate state and start campaigning to become leader.
	Campaign(ctx context.Context) error
	// Propose proposes that data be appended to the log. Note that proposals can be lost without
	// notice, therefore it is user's job to ensure proposal retries.
	Propose(ctx context.Context, data []byte) error
	// ProposeConfChange proposes a configuration change. Like any proposal, the
	// configuration change may be dropped with or without an error being
	// returned. In particular, configuration changes are dropped unless the
	// leader has certainty that there is no prior unapplied configuration
	// change in its log.
	//
	// The method accepts either a pb.ConfChange (deprecated) or pb.ConfChangeV2
	// message. The latter allows arbitrary configuration changes via joint
	// consensus, notably including replacing a voter. Passing a ConfChangeV2
	// message is only allowed if all Nodes participating in the cluster run a
	// version of this library aware of the V2 API. See pb.ConfChangeV2 for
	// usage details and semantics.
	ProposeConfChange(ctx context.Context, cc pb.ConfChangeI) error

	// Step advances the state machine using the given message. ctx.Err() will be returned, if any.
	Step(ctx context.Context, msg pb.Message) error

	// Ready returns a channel that returns the current point-in-time state.
	// Users of the Node must call Advance after retrieving the state returned by Ready.
	//
	// NOTE: No committed entries from the next Ready may be applied until all committed entries
	// and snapshots from the previous one have finished.
	Ready() <-chan Ready

	// Advance notifies the Node that the application has saved progress up to the last Ready.
	// It prepares the node to return the next available Ready.
	//
	// The application should generally call Advance after it applies the entries in last Ready.
	//
	// However, as an optimization, the application may call Advance while it is applying the
	// commands. For example. when the last Ready contains a snapshot, the application might take
	// a long time to apply the snapshot data. To continue receiving Ready without blocking raft
	// progress, it can call Advance before finishing applying the last ready.
	Advance()
	// ApplyConfChange applies a config change (previously passed to
	// ProposeConfChange) to the node. This must be called whenever a config
	// change is observed in Ready.CommittedEntries, except when the app decides
	// to reject the configuration change (i.e. treats it as a noop instead), in
	// which case it must not be called.
	//
	// Returns an opaque non-nil ConfState protobuf which must be recorded in
	// snapshots.
	ApplyConfChange(cc pb.ConfChangeI) *pb.ConfState

	// TransferLeadership attempts to transfer leadership to the given transferee.
	TransferLeadership(ctx context.Context, lead, transferee uint64)

	// ReadIndex request a read state. The read state will be set in the ready.
	// Read state has a read index. Once the application advances further than the read
	// index, any linearizable read requests issued before the read request can be
	// processed safely. The read state will have the same rctx attached.
	ReadIndex(ctx context.Context, rctx []byte) error

	// Status returns the current status of the raft state machine.
	Status() Status
	// ReportUnreachable reports the given node is not reachable for the last send.
	ReportUnreachable(id uint64)
	// ReportSnapshot reports the status of the sent snapshot. The id is the raft ID of the follower
	// who is meant to receive the snapshot, and the status is SnapshotFinish or SnapshotFailure.
	// Calling ReportSnapshot with SnapshotFinish is a no-op. But, any failure in applying a
	// snapshot (for e.g., while streaming it from leader to follower), should be reported to the
	// leader with SnapshotFailure. When leader sends a snapshot to a follower, it pauses any raft
	// log probes until the follower can apply the snapshot and advance its state. If the follower
	// can't do that, for e.g., due to a crash, it could end up in a limbo, never getting any
	// updates from the leader. Therefore, it is crucial that the application ensures that any
	// failure in snapshot sending is caught and reported back to the leader; so it can resume raft
	// log probing in the follower.
	ReportSnapshot(id uint64, status SnapshotStatus)
	// Stop performs any necessary termination of the Node.
	Stop()
}

// Tick advances the internal logical clock by a single tick.
func (rn *RawNode) Tick() {
	rn.raft.tick()
}

type raft struct {
	// ... ...
	tick func()
	// ... ...
}

func (r *raft) becomeFollower(term uint64, lead uint64) {
	// ... ...
	r.tick = r.tickElection
	// ... ...
}

func (r *raft) becomeCandidate() {
	// ... ...
	r.tick = r.tickElection
	// ... ...
}

func (r *raft) becomePreCandidate() {
	// ... ...
	r.tick = r.tickElection
	// ... ...
}

func (r *raft) becomeLeader() {
	// ... ...
	r.tick = r.tickHeartbeat
	// ... ...
}

// tickElection is run by followers and candidates after r.electionTimeout.
func (r *raft) tickElection() {
	r.electionElapsed++

	if r.promotable() && r.pastElectionTimeout() {
		r.electionElapsed = 0
		r.Step(pb.Message{From: r.id, Type: pb.MsgHup})
	}
}

// Config contains the parameters to start a raft.
type Config struct {
	// ID is the identity of the local raft. ID cannot be 0.
	ID uint64

	// peers contains the IDs of all nodes (including self) in the raft cluster. It
	// should only be set when starting a new raft cluster. Restarting raft from
	// previous configuration will panic if peers is set. peer is private and only
	// used for testing right now.
	peers []uint64

	// learners contains the IDs of all learner nodes (including self if the
	// local node is a learner) in the raft cluster. learners only receives
	// entries from the leader node. It does not vote or promote itself.
	learners []uint64

	// ElectionTick is the number of Node.Tick invocations that must pass between
	// elections. That is, if a follower does not receive any message from the
	// leader of current term before ElectionTick has elapsed, it will become
	// candidate and start an election. ElectionTick must be greater than
	// HeartbeatTick. We suggest ElectionTick = 10 * HeartbeatTick to avoid
	// unnecessary leader switching.
	ElectionTick int
	// HeartbeatTick is the number of Node.Tick invocations that must pass between
	// heartbeats. That is, a leader sends heartbeat messages to maintain its
	// leadership every HeartbeatTick ticks.
	HeartbeatTick int

	// Storage is the storage for raft. raft generates entries and states to be
	// stored in storage. raft reads the persisted entries and states out of
	// Storage when it needs. raft reads out the previous state and configuration
	// out of storage when restarting.
	Storage Storage
	// Applied is the last applied index. It should only be set when restarting
	// raft. raft will not return entries to the application smaller or equal to
	// Applied. If Applied is unset when restarting, raft might return previous
	// applied entries. This is a very application dependent configuration.
	Applied uint64

	// MaxSizePerMsg limits the max byte size of each append message. Smaller
	// value lowers the raft recovery cost(initial probing and message lost
	// during normal operation). On the other side, it might affect the
	// throughput during normal replication. Note: math.MaxUint64 for unlimited,
	// 0 for at most one entry per message.
	MaxSizePerMsg uint64
	// MaxCommittedSizePerReady limits the size of the committed entries which
	// can be applied.
	MaxCommittedSizePerReady uint64
	// MaxUncommittedEntriesSize limits the aggregate byte size of the
	// uncommitted entries that may be appended to a leader's log. Once this
	// limit is exceeded, proposals will begin to return ErrProposalDropped
	// errors. Note: 0 for no limit.
	MaxUncommittedEntriesSize uint64
	// MaxInflightMsgs limits the max number of in-flight append messages during
	// optimistic replication phase. The application transportation layer usually
	// has its own sending buffer over TCP/UDP. Setting MaxInflightMsgs to avoid
	// overflowing that sending buffer. TODO (xiangli): feedback to application to
	// limit the proposal rate?
	MaxInflightMsgs int

	// CheckQuorum specifies if the leader should check quorum activity. Leader
	// steps down when quorum is not active for an electionTimeout.
	CheckQuorum bool

	// PreVote enables the Pre-Vote algorithm described in raft thesis section
	// 9.6. This prevents disruption when a node that has been partitioned away
	// rejoins the cluster.
	PreVote bool

	// ReadOnlyOption specifies how the read only request is processed.
	//
	// ReadOnlySafe guarantees the linearizability of the read only request by
	// communicating with the quorum. It is the default and suggested option.
	//
	// ReadOnlyLeaseBased ensures linearizability of the read only request by
	// relying on the leader lease. It can be affected by clock drift.
	// If the clock drift is unbounded, leader might keep the lease longer than it
	// should (clock can move backward/pause without any bound). ReadIndex is not safe
	// in that case.
	// CheckQuorum MUST be enabled if ReadOnlyOption is ReadOnlyLeaseBased.
	ReadOnlyOption ReadOnlyOption

	// Logger is the logger used for raft log. For multinode which can host
	// multiple raft group, each raft group can have its own logger
	Logger Logger

	// DisableProposalForwarding set to true means that followers will drop
	// proposals, rather than forwarding them to the leader. One use case for
	// this feature would be in a situation where the Raft leader is used to
	// compute the data of a proposal, for example, adding a timestamp from a
	// hybrid logical clock to data in a monotonically increasing way. Forwarding
	// should be disabled to prevent a follower with an inaccurate hybrid
	// logical clock from assigning the timestamp and then forwarding the data
	// to the leader.
	DisableProposalForwarding bool
}