// send persists state to stable storage and then sends to its mailbox.
func (r *raft) send(m pb.Message) {
m.From = r.id
// do not attach term to MsgProp
// proposals are a way to forward to the leader and
// should be treated as local message.
if m.Type != pb.MsgProp {
m.Term = r.Term
}
r.msgs = append(r.msgs, m)
}
func stepFollower(r *raft, m pb.Message) {
switch m.Type {
case pb.MsgProp:
if r.lead == None {
r.logger.Infof("%x no leader at term %d; dropping proposal", r.id, r.Term)
return
}
m.To = r.lead
r.send(m)
case pb.MsgApp:
r.elapsed = 0
r.lead = m.From
r.handleAppendEntries(m)
case pb.MsgHeartbeat:
r.elapsed = 0
r.lead = m.From
r.handleHeartbeat(m)
case pb.MsgSnap:
r.elapsed = 0
r.handleSnapshot(m)
case pb.MsgVote:
if (r.Vote == None || r.Vote == m.From) && r.raftLog.isUpToDate(m.Index, m.LogTerm) {
r.elapsed = 0
r.logger.Infof("%x [logterm: %d, index: %d, vote: %x] voted for %x [logterm: %d, index: %d] at term %d",
r.id, r.raftLog.lastTerm(), r.raftLog.lastIndex(), r.Vote, m.From, m.LogTerm, m.Index, r.Term)
r.Vote = m.From
r.send(pb.Message{To: m.From, Type: pb.MsgVoteResp})
} else {
r.logger.Infof("%x [logterm: %d, index: %d, vote: %x] rejected vote from %x [logterm: %d, index: %d] at term %d",
r.id, r.raftLog.lastTerm(), r.raftLog.lastIndex(), r.Vote, m.From, m.LogTerm, m.Index, r.Term)
r.send(pb.Message{To: m.From, Type: pb.MsgVoteResp, Reject: true})
}
}
}
// Encode
func (e *Encoder) Encode(m raftpb.Message) error {
var buf [4]byte
b := buf[:4]
s := m.Size()
if s == 0 {
return ErrEmptyMessage
}
binary.BigEndian.PutUint32(b, uint32(s))
e.w.Write(b)
e.off += 4
b, err := m.Marshal()
if err != nil {
return err
}
e.w.Write(b)
e.off += s
return nil
}
// sendAppend sends RRPC, with entries to the given peer.
func (r *raft) sendAppend(to uint64) {
pr := r.prs[to]
if pr.isPaused() {
return
}
m := pb.Message{}
m.To = to
term, errt := r.raftLog.term(pr.Next - 1)
ents, erre := r.raftLog.entries(pr.Next, r.maxMsgSize)
if errt != nil || erre != nil { // send snapshot if we failed to get term or entries
m.Type = pb.MsgSnap
snapshot, err := r.raftLog.snapshot()
if err != nil {
panic(err) // TODO(bdarnell)
}
if IsEmptySnap(snapshot) {
panic("need non-empty snapshot")
}
m.Snapshot = snapshot
sindex, sterm := snapshot.Metadata.Index, snapshot.Metadata.Term
r.logger.Debugf("%x [firstindex: %d, commit: %d] sent snapshot[index: %d, term: %d] to %x [%s]",
r.id, r.raftLog.firstIndex(), r.Commit, sindex, sterm, to, pr)
pr.becomeSnapshot(sindex)
r.logger.Debugf("%x paused sending replication messages to %x [%s]", r.id, to, pr)
} else {
m.Type = pb.MsgApp
m.Index = pr.Next - 1
m.LogTerm = term
m.Entries = ents
m.Commit = r.raftLog.committed
if n := len(m.Entries); n != 0 {
switch pr.State {
// optimistically increase the next when in ProgressStateReplicate
case ProgressStateReplicate:
last := m.Entries[n-1].Index
pr.optimisticUpdate(last)
pr.ins.add(last)
case ProgressStateProbe:
pr.pause()
default:
r.logger.Panicf("%x is sending append in unhandled state %s", r.id, pr.State)
}
}
}
r.send(m)
}
// Decode decodes the next raft message. If there is no data to read it returns
// io.EOF.
func (d *Decoder) Decode(m *raftpb.Message) error {
var buf [4]byte
b := buf[:4]
n, err := d.r.Read(b)
if err != nil {
return err
}
if n != 4 {
return ErrNoEnoughData
}
s := int(binary.BigEndian.Uint32(b))
b = make([]byte, s)
n, err = io.ReadFull(d.r, b)
if err != nil {
return err
}
if n != s {
return ErrNoEnoughData
}
m.Unmarshal(b)
return nil
}
