mirrors/etcd
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

rafthttp: extract pipeline from peer

Browse Source
This commit is contained in:
Yicheng Qin
2015-02-05 09:40:46 -08:00
parent 2af33fd494
commit d8a9e11e22
3 changed files with 214 additions and 158 deletions
Download Patch File Download Diff File Expand all files Collapse all files

145
rafthttp/peer.go
View File

@ -15,28 +15,17 @@
package rafthttp
import (
"bytes"
"errors"
"fmt"
"log"
"net/http"
"sync"
"time"
"github.com/coreos/etcd/etcdserver/stats"
"github.com/coreos/etcd/pkg/pbutil"
"github.com/coreos/etcd/pkg/types"
"github.com/coreos/etcd/raft/raftpb"
)
const (
connPerSender = 4
// senderBufSize is the size of sender buffer, which helps hold the
// temporary network latency.
// The size ensures that sender does not drop messages when the network
// is out of work for less than 1 second in good path.
senderBufSize = 64
appRespBatchMs = 50
propBatchMs = 10
@ -50,50 +39,35 @@ type peer struct {
id types.ID
cid types.ID
tr http.RoundTripper
r Raft
fs *stats.FollowerStats
errorc chan error
tr http.RoundTripper
// the url this sender post to
u string
r Raft
fs *stats.FollowerStats
batcher *Batcher
propBatcher *ProposalBatcher
q chan *raftpb.Message
stream *stream
pipeline *pipeline
stream *stream
// wait for the handling routines
wg sync.WaitGroup
// the url this sender post to
u string
// if the last send was successful, the sender is active.
// Or it is inactive
active bool
errored error
paused bool
stopped bool
}
func NewPeer(tr http.RoundTripper, u string, id types.ID, cid types.ID, r Raft, fs *stats.FollowerStats, errorc chan error) *peer {
p := &peer{
return &peer{
id: id,
active: true,
cid: cid,
tr: tr,
u: u,
cid: cid,
r: r,
fs: fs,
pipeline: newPipeline(tr, u, id, cid, fs, errorc),
stream: &stream{},
errorc: errorc,
batcher: NewBatcher(100, appRespBatchMs*time.Millisecond),
propBatcher: NewProposalBatcher(100, propBatchMs*time.Millisecond),
q: make(chan *raftpb.Message, senderBufSize),
}
p.wg.Add(connPerSender)
for i := 0; i < connPerSender; i++ {
go p.handle()
}
return p
}
func (p *peer) Update(u string) {
@ -104,6 +78,7 @@ func (p *peer) Update(u string) {
panic("peer: update a stopped peer")
}
p.u = u
p.pipeline.update(u)
}
// Send sends the data to the remote node. It is always non-blocking.
@ -134,14 +109,14 @@ func (p *peer) Send(m raftpb.Message) error {
p.propBatcher.Batch(m)
case canBatch(m) && p.stream.isOpen():
if !p.batcher.ShouldBatch(time.Now()) {
err = p.send(m)
err = p.pipeline.send(m)
}
case canUseStream(m):
if ok := p.stream.write(m); !ok {
err = p.send(m)
err = p.pipeline.send(m)
}
default:
err = p.send(m)
err = p.pipeline.send(m)
}
// send out batched MsgProp if needed
// TODO: it is triggered by all outcoming send now, and it needs
@ -150,111 +125,23 @@ func (p *peer) Send(m raftpb.Message) error {
if !p.propBatcher.IsEmpty() {
t := time.Now()
if !p.propBatcher.ShouldBatch(t) {
p.send(p.propBatcher.Message)
p.pipeline.send(p.propBatcher.Message)
p.propBatcher.Reset(t)
}
}
return err
}
func (p *peer) send(m raftpb.Message) error {
// TODO: don't block. we should be able to have 1000s
// of messages out at a time.
select {
case p.q <- &m:
return nil
default:
log.Printf("sender: dropping %s because maximal number %d of sender buffer entries to %s has been reached",
m.Type, senderBufSize, p.u)
return fmt.Errorf("reach maximal serving")
}
}
// Stop performs any necessary finalization and terminates the peer
// elegantly.
func (p *peer) Stop() {
close(p.q)
p.wg.Wait()
p.Lock()
defer p.Unlock()
p.pipeline.stop()
p.stream.stop()
p.stopped = true
}
func (p *peer) handle() {
defer p.wg.Done()
for m := range p.q {
start := time.Now()
err := p.post(pbutil.MustMarshal(m))
end := time.Now()
p.Lock()
if err != nil {
if p.errored == nil || p.errored.Error() != err.Error() {
log.Printf("sender: error posting to %s: %v", p.id, err)
p.errored = err
}
if p.active {
log.Printf("sender: the connection with %s became inactive", p.id)
p.active = false
}
if m.Type == raftpb.MsgApp {
p.fs.Fail()
}
} else {
if !p.active {
log.Printf("sender: the connection with %s became active", p.id)
p.active = true
p.errored = nil
}
if m.Type == raftpb.MsgApp {
p.fs.Succ(end.Sub(start))
}
}
p.Unlock()
}
}
// post POSTs a data payload to a url. Returns nil if the POST succeeds,
// error on any failure.
func (p *peer) post(data []byte) error {
p.Lock()
req, err := http.NewRequest("POST", p.u, bytes.NewBuffer(data))
p.Unlock()
if err != nil {
return err
}
req.Header.Set("Content-Type", "application/protobuf")
req.Header.Set("X-Etcd-Cluster-ID", p.cid.String())
resp, err := p.tr.RoundTrip(req)
if err != nil {
return err
}
resp.Body.Close()
switch resp.StatusCode {
case http.StatusPreconditionFailed:
err := fmt.Errorf("conflicting cluster ID with the target cluster (%s != %s)", resp.Header.Get("X-Etcd-Cluster-ID"), p.cid)
select {
case p.errorc <- err:
default:
}
return nil
case http.StatusForbidden:
err := fmt.Errorf("the member has been permanently removed from the cluster")
select {
case p.errorc <- err:
default:
}
return nil
case http.StatusNoContent:
return nil
default:
return fmt.Errorf("unexpected http status %s while posting to %q", http.StatusText(resp.StatusCode), req.URL.String())
}
}
// attachStream attaches a streamSever to the peer.
func (p *peer) attachStream(sw *streamWriter) error {
p.Lock()

169
rafthttp/pipeline.go Normal file
View File

@ -0,0 +1,169 @@
// Copyright 2015 CoreOS, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package rafthttp
import (
"bytes"
"fmt"
"log"
"net/http"
"sync"
"time"
"github.com/coreos/etcd/etcdserver/stats"
"github.com/coreos/etcd/pkg/pbutil"
"github.com/coreos/etcd/pkg/types"
"github.com/coreos/etcd/raft/raftpb"
)
const (
connPerPipeline = 4
// pipelineBufSize is the size of pipeline buffer, which helps hold the
// temporary network latency.
// The size ensures that pipeline does not drop messages when the network
// is out of work for less than 1 second in good path.
pipelineBufSize = 64
)
type pipeline struct {
id types.ID
cid types.ID
tr http.RoundTripper
// the url this pipeline sends to
u string
fs *stats.FollowerStats
errorc chan error
q chan *raftpb.Message
// wait for the handling routines
wg sync.WaitGroup
sync.Mutex
// if the last send was successful, the pipeline is active.
// Or it is inactive
active bool
errored error
}
func newPipeline(tr http.RoundTripper, u string, id, cid types.ID, fs *stats.FollowerStats, errorc chan error) *pipeline {
p := &pipeline{
id: id,
cid: cid,
tr: tr,
u: u,
fs: fs,
errorc: errorc,
q: make(chan *raftpb.Message, pipelineBufSize),
active: true,
}
p.wg.Add(connPerPipeline)
for i := 0; i < connPerPipeline; i++ {
go p.handle()
}
return p
}
func (p *pipeline) update(u string) { p.u = u }
func (p *pipeline) send(m raftpb.Message) error {
// TODO: don't block. we should be able to have 1000s
// of messages out at a time.
select {
case p.q <- &m:
return nil
default:
log.Printf("pipeline: dropping %s because maximal number %d of pipeline buffer entries to %s has been reached",
m.Type, pipelineBufSize, p.u)
return fmt.Errorf("reach maximal serving")
}
}
func (p *pipeline) stop() {
close(p.q)
p.wg.Wait()
}
func (p *pipeline) handle() {
defer p.wg.Done()
for m := range p.q {
start := time.Now()
err := p.pipeline(pbutil.MustMarshal(m))
end := time.Now()
p.Lock()
if err != nil {
if p.errored == nil || p.errored.Error() != err.Error() {
log.Printf("pipeline: error posting to %s: %v", p.id, err)
p.errored = err
}
if p.active {
log.Printf("pipeline: the connection with %s became inactive", p.id)
p.active = false
}
if m.Type == raftpb.MsgApp {
p.fs.Fail()
}
} else {
if !p.active {
log.Printf("pipeline: the connection with %s became active", p.id)
p.active = true
p.errored = nil
}
if m.Type == raftpb.MsgApp {
p.fs.Succ(end.Sub(start))
}
}
p.Unlock()
}
}
// post POSTs a data payload to a url. Returns nil if the POST succeeds,
// error on any failure.
func (p *pipeline) pipeline(data []byte) error {
p.Lock()
req, err := http.NewRequest("POST", p.u, bytes.NewBuffer(data))
p.Unlock()
if err != nil {
return err
}
req.Header.Set("Content-Type", "application/protobuf")
req.Header.Set("X-Etcd-Cluster-ID", p.cid.String())
resp, err := p.tr.RoundTrip(req)
if err != nil {
return err
}
resp.Body.Close()
switch resp.StatusCode {
case http.StatusPreconditionFailed:
err := fmt.Errorf("conflicting cluster ID with the target cluster (%s != %s)", resp.Header.Get("X-Etcd-Cluster-ID"), p.cid)
select {
case p.errorc <- err:
default:
}
return nil
case http.StatusForbidden:
err := fmt.Errorf("the member has been permanently removed from the cluster")
select {
case p.errorc <- err:
default:
}
return nil
case http.StatusNoContent:
return nil
default:
return fmt.Errorf("unexpected http status %s while posting to %q", http.StatusText(resp.StatusCode), req.URL.String())
}
}

58
rafthttp/peer_test.go → rafthttp/pipeline_test.go
View File

@ -27,17 +27,17 @@ import (
"github.com/coreos/etcd/raft/raftpb"
)
// TestSenderSend tests that send func could post data using roundtripper
// TestPipelineSend tests that pipeline could send data using roundtripper
// and increase success count in stats.
func TestSenderSend(t *testing.T) {
func TestPipelineSend(t *testing.T) {
tr := &roundTripperRecorder{}
fs := &stats.FollowerStats{}
p := NewPeer(tr, "http://10.0.0.1", types.ID(1), types.ID(1), &nopProcessor{}, fs, nil)
p := newPipeline(tr, "http://10.0.0.1", types.ID(1), types.ID(1), fs, nil)
if err := p.Send(raftpb.Message{Type: raftpb.MsgApp}); err != nil {
if err := p.send(raftpb.Message{Type: raftpb.MsgApp}); err != nil {
t.Fatalf("unexpect send error: %v", err)
}
p.Stop()
p.stop()
if tr.Request() == nil {
t.Errorf("sender fails to post the data")
@ -49,15 +49,15 @@ func TestSenderSend(t *testing.T) {
}
}
func TestSenderExceedMaximalServing(t *testing.T) {
func TestPipelineExceedMaximalServing(t *testing.T) {
tr := newRoundTripperBlocker()
fs := &stats.FollowerStats{}
p := NewPeer(tr, "http://10.0.0.1", types.ID(1), types.ID(1), &nopProcessor{}, fs, nil)
p := newPipeline(tr, "http://10.0.0.1", types.ID(1), types.ID(1), fs, nil)
// keep the sender busy and make the buffer full
// nothing can go out as we block the sender
for i := 0; i < connPerSender+senderBufSize; i++ {
if err := p.Send(raftpb.Message{}); err != nil {
for i := 0; i < connPerPipeline+pipelineBufSize; i++ {
if err := p.send(raftpb.Message{}); err != nil {
t.Errorf("send err = %v, want nil", err)
}
// force the sender to grab data
@ -65,7 +65,7 @@ func TestSenderExceedMaximalServing(t *testing.T) {
}
// try to send a data when we are sure the buffer is full
if err := p.Send(raftpb.Message{}); err == nil {
if err := p.send(raftpb.Message{}); err == nil {
t.Errorf("unexpect send success")
}
@ -74,22 +74,22 @@ func TestSenderExceedMaximalServing(t *testing.T) {
testutil.ForceGosched()
// It could send new data after previous ones succeed
if err := p.Send(raftpb.Message{}); err != nil {
if err := p.send(raftpb.Message{}); err != nil {
t.Errorf("send err = %v, want nil", err)
}
p.Stop()
p.stop()
}
// TestSenderSendFailed tests that when send func meets the post error,
// TestPipelineSendFailed tests that when send func meets the post error,
// it increases fail count in stats.
func TestSenderSendFailed(t *testing.T) {
func TestPipelineSendFailed(t *testing.T) {
fs := &stats.FollowerStats{}
p := NewPeer(newRespRoundTripper(0, errors.New("blah")), "http://10.0.0.1", types.ID(1), types.ID(1), &nopProcessor{}, fs, nil)
p := newPipeline(newRespRoundTripper(0, errors.New("blah")), "http://10.0.0.1", types.ID(1), types.ID(1), fs, nil)
if err := p.Send(raftpb.Message{Type: raftpb.MsgApp}); err != nil {
if err := p.send(raftpb.Message{Type: raftpb.MsgApp}); err != nil {
t.Fatalf("unexpect Send error: %v", err)
}
p.Stop()
p.stop()
fs.Lock()
defer fs.Unlock()
@ -98,13 +98,13 @@ func TestSenderSendFailed(t *testing.T) {
}
}
func TestSenderPost(t *testing.T) {
func TestPipelinePost(t *testing.T) {
tr := &roundTripperRecorder{}
p := NewPeer(tr, "http://10.0.0.1", types.ID(1), types.ID(1), &nopProcessor{}, nil, nil)
if err := p.post([]byte("some data")); err != nil {
p := newPipeline(tr, "http://10.0.0.1", types.ID(1), types.ID(1), nil, nil)
if err := p.pipeline([]byte("some data")); err != nil {
t.Fatalf("unexpect post error: %v", err)
}
p.Stop()
p.stop()
if g := tr.Request().Method; g != "POST" {
t.Errorf("method = %s, want %s", g, "POST")
@ -127,7 +127,7 @@ func TestSenderPost(t *testing.T) {
}
}
func TestSenderPostBad(t *testing.T) {
func TestPipelinePostBad(t *testing.T) {
tests := []struct {
u string
code int
@ -142,9 +142,9 @@ func TestSenderPostBad(t *testing.T) {
{"http://10.0.0.1", http.StatusCreated, nil},
}
for i, tt := range tests {
p := NewPeer(newRespRoundTripper(tt.code, tt.err), tt.u, types.ID(1), types.ID(1), &nopProcessor{}, nil, make(chan error))
err := p.post([]byte("some data"))
p.Stop()
p := newPipeline(newRespRoundTripper(tt.code, tt.err), tt.u, types.ID(1), types.ID(1), nil, make(chan error))
err := p.pipeline([]byte("some data"))
p.stop()
if err == nil {
t.Errorf("#%d: err = nil, want not nil", i)
@ -152,7 +152,7 @@ func TestSenderPostBad(t *testing.T) {
}
}
func TestPeerPostErrorc(t *testing.T) {
func TestPipelinePostErrorc(t *testing.T) {
tests := []struct {
u string
code int
@ -163,9 +163,9 @@ func TestPeerPostErrorc(t *testing.T) {
}
for i, tt := range tests {
errorc := make(chan error, 1)
p := NewPeer(newRespRoundTripper(tt.code, tt.err), tt.u, types.ID(1), types.ID(1), &nopProcessor{}, nil, errorc)
p.post([]byte("some data"))
p.Stop()
p := newPipeline(newRespRoundTripper(tt.code, tt.err), tt.u, types.ID(1), types.ID(1), nil, errorc)
p.pipeline([]byte("some data"))
p.stop()
select {
case <-errorc:
default: