zookeeper(4) 网络

　　zookeeper底层通过NIO进行网络传输，如果对NIO不是很熟悉，先参见java NIO。下面我们来逐步实现基于NIO的zookeeper实现，首先我们要定义应用层的通信协议。

传输协议

　　请求协议格式：

	字段	长度	说明
	len	4	请求长度，解决半包粘包
请求头	xid	4	请求在客户端的id，用于唯一标识和保证响应顺序
	type	4	请求类型code
请求体	request		不同的请求类型，有不同的请求体结构

　　响应协议格式：

	字段	长度	解释
	len	4	响应长度，解决半包粘包
响应头	xid	4	包序号，唯一标识一个包，通过该标识找到对应的客户端Packet对象
	zxid	8	服务端事务处理id
	err	4	返回状态code
响应体	response		不同的响应类型有不同的响应体

　　请求和响应体类型：

List<String> children

请求code	请求头类型	说明	请求格式	字段说明	响应格式	字段说明
0	notification
1	create	创建节点	String path	路径	String path	路径
			byte[] data	节点值
			List<ACL> acl	acl值
			int flags	节点类型
2	delete	删除节点	String path	路径	无
			int version	版本
3	exists	是否存在指定节点	String path	路径	Stat stat	节点状态
			boolean watch	是否有watch
4	getData	获取节点数据	String path	路径	Stat stat	节点状态
			boolean watch	是否有watch	byte[] data	节点数据
5	setData	设置节点数据	String path	路径	Stat stat	节点状态
			byte[] data	数据
			int version	版本
6	getACL	获取节点权限	String path	路径	List<ACL> acl	权限
					Stat stat	节点状态
7	setACL	设置节点权限	String path	路径	Stat stat	节点状态
			List<ACL> acl	权限
			int version	版本
8	getChildren	获取子节点	String path	路径	List<String> children	子节点名
			boolean watch	是否有watch
9	sync	同步路径	String path	路径	String path	路径
11	ping
12	getChildren2
100	auth
101	setWatches
-10	createSession
-11	closeSession
－1	error

　　1.将请求封装成Packet对象放入outgoingQueue队列中。

　　2.有一个收发送线程会从outgoingQueue队列中取出一个可发送的Packet对象，并发送序列化信息，然后把该Packet放入到pendingQueue队列中。

　　3.收发线程会接收服务端响应，反序列号出结果数据，然后在pendingQueue中找到对应的Packet，设置结果，将Packet放入到waitingEvents队列中。

　　4.有一个事件线程，会不断从waitingEvents队列中取出一个Packet，并调用响应的callback方法。

发送packet包

Packet queuePacket(RequestHeader h, ReplyHeader r, Record request,
            Record response, AsyncCallback cb, String clientPath,
            String serverPath, Object ctx, WatchRegistration watchRegistration)
    {
        
        Packet packet = null;
        synchronized (outgoingQueue) {
            //设置一个全局唯一的id，作为数据包的id
            if (h.getType() != OpCode.ping && h.getType() != OpCode.auth) {
                h.setXid(getXid());
            }
            //将请求头，请求体，返回结果，watcher等封装成数据包。
            packet = new Packet(h, r, request, response, null,
                    watchRegistration);
            packet.cb = cb;
            packet.ctx = ctx;
            packet.clientPath = clientPath;
            packet.serverPath = serverPath;
            //将数据包添加到outgoing队列中。
            outgoingQueue.add(packet);
        }
        sendThread.wakeup();
        return packet;
    }

发送线程主流程（ClientCnxn.SendThread.run）：

class SendThread extends Thread {
        SelectionKey sockKey;
        private final Selector selector = Selector.open();
        public void run() {
            while (zooKeeper.state.isAlive()) {
                //建立连接
                startConnect();
                //获取注册通道
                selector.select(1000);
                Set<SelectionKey> selected;
                synchronized (this) {
                    selected = selector.selectedKeys();
                }
                for (SelectionKey k : selected) {
                    SocketChannel sc = ((SocketChannel) k.channel());
                    if ((k.readyOps() & SelectionKey.OP_CONNECT) != 0) {
                        //建立连接
                        if (sc.finishConnect()) {
                            primeConnection(k);
                        }
                        //读写数据
                    } else if ((k.readyOps() & (SelectionKey.OP_READ | SelectionKey.OP_WRITE)) != 0) {
                        doIO();
                    }
                }
            }
            try {
                selector.close();
            } catch (IOException e) {
                LOG.warn("Ignoring exception during selector close", e);
            }
        }

      //通过nio建立连接
        private void startConnect() throws IOException {
            //从服务器列表中获取一台服务器地址
            InetSocketAddress addr = serverAddrs.get(nextAddrToTry);
            nextAddrToTry++;
            if (nextAddrToTry == serverAddrs.size()) {
                nextAddrToTry = 0;
            }
            //通过nio注册
            SocketChannel sock;
            sock = SocketChannel.open();
            sock.configureBlocking(false);
            sock.socket().setSoLinger(false, -1);
            sock.socket().setTcpNoDelay(true);
            try {
                sockKey = sock.register(selector, SelectionKey.OP_CONNECT);
            } catch (IOException e) {
                sock.close();
                throw e;
            }
            //初始化缓存
            lenBuffer.clear();
            incomingBuffer = lenBuffer;
        }
}

处理读写主流程,主要是nio操作（ClientCnxn.SendThread.doIO）：

boolean doIO() throws InterruptedException, IOException {
            boolean packetReceived = false;
            //获取socketchannel
            SocketChannel sock = (SocketChannel) sockKey.channel();
            //如果可读
            if (sockKey.isReadable()) {
                //读取数据到缓存中
                int rc = sock.read(incomingBuffer);
                //直到缓存读满
                if (!incomingBuffer.hasRemaining()) {
                    //重置缓存
                    incomingBuffer.flip();
                    //如果读取的是长度信息，读取长度信息，并且分配相应缓存
                    if (incomingBuffer == lenBuffer) {
                        int len = incomingBuffer.getInt();
                        incomingBuffer = ByteBuffer.allocate(len);
                    } else if (!initialized) {
                      //如果是connect命令的返回值，获取session，timeout等相关信息
                        readConnectResult();
                        enableRead();
                        lenBuffer.clear();
                        //重置缓存
                        incomingBuffer = lenBuffer;
                        initialized = true;
                    } else {
                        //读取数据内容
                        readResponse();
                        //重置缓存
                        lenBuffer.clear();
                        incomingBuffer = lenBuffer;
                        packetReceived = true;
                    }
                }
            }
            //如果是写
            if (sockKey.isWritable()) {
                synchronized (outgoingQueue) {
                    if (!outgoingQueue.isEmpty()) {
                        //从outgoingQueue队列中拿数据包写入通道
                        ByteBuffer pbb = outgoingQueue.getFirst().bb;
                        sock.write(pbb);
                        if (!pbb.hasRemaining()) {
                            sentCount++;
                            Packet p = outgoingQueue.removeFirst();
                            if (p.header != null
                                    && p.header.getType() != OpCode.ping
                                    && p.header.getType() != OpCode.auth) {
                                pendingQueue.add(p);
                            }
                        }
                    }
                }
            }
            if (outgoingQueue.isEmpty()) {
                disableWrite();
            } else {
                enableWrite();
            }
            return packetReceived;
        }

处理返回结果，xid＝－2为ping命令的返回结果；xid＝-4为auth命令；xid＝－1为服务器发送的notification；其他命令返回结果。

void readResponse() throws IOException {
            //对返回数据进行反序列化
            ByteBufferInputStream bbis = new ByteBufferInputStream(
                    incomingBuffer);
            BinaryInputArchive bbia = BinaryInputArchive.getArchive(bbis);
            ReplyHeader replyHdr = new ReplyHeader();
            replyHdr.deserialize(bbia, "header");
            //根据返回头信息，封装想要的事件，放入事件队列中，交给eventthread处理
//向消息队列放入验证失败消息
            if (replyHdr.getXid() == -4) {
                 // -4 is the xid for AuthPacket               
                if(replyHdr.getErr() == KeeperException.Code.AUTHFAILED.intValue()) {
                    zooKeeper.state = States.AUTH_FAILED;                    
                    eventThread.queueEvent( new WatchedEvent(Watcher.Event.EventType.None, 
                            Watcher.Event.KeeperState.AuthFailed, null) );                                        
                }
                return;
            }
//
            if (replyHdr.getXid() == -1) {
                WatcherEvent event = new WatcherEvent();
                event.deserialize(bbia, "response");
                if (chrootPath != null) {
                    String serverPath = event.getPath();
                    if(serverPath.compareTo(chrootPath)==0)
                        event.setPath("/");
                    else
                        event.setPath(serverPath.substring(chrootPath.length()));
                }
                WatchedEvent we = new WatchedEvent(event);
                eventThread.queueEvent( we );
                return;
            }
//反序列化返回结果
            Packet packet = null;
            synchronized (pendingQueue) {
                packet = pendingQueue.remove();
            }
            try {
                packet.replyHeader.setXid(replyHdr.getXid());
                packet.replyHeader.setErr(replyHdr.getErr());
                packet.replyHeader.setZxid(replyHdr.getZxid());
                if (replyHdr.getZxid() > 0) {
                    lastZxid = replyHdr.getZxid();
                }
                if (packet.response != null && replyHdr.getErr() == 0) {
                    packet.response.deserialize(bbia, "response");
                }
            } finally {
                finishPacket(packet);
            }
        }

private void finishPacket(Packet p) {
        if (p.watchRegistration != null) {
            p.watchRegistration.register(p.replyHeader.getErr());
        }

        if (p.cb == null) {
            synchronized (p) {
                p.finished = true;
                p.notifyAll();
            }
        } else {
            p.finished = true;
            eventThread.queuePacket(p);
        }
    }

事件线程主要是处理回调函数（ClientCnxn.EventThread.run）：

public void run() {
           try {
              isRunning = true;
              while (true) {
                  //从队列中获取事件
                 Object event = waitingEvents.take();
                 if (event == eventOfDeath) {
                    wasKilled = true;
                 } else {
                     //处理事件
                    processEvent(event);
                 }
                 if (wasKilled)
                    synchronized (waitingEvents) {
                       if (waitingEvents.isEmpty()) {
                          isRunning = false;
                          break;
                       }
                    }
              }
           } catch (InterruptedException e) {
              LOG.error("Event thread exiting due to interruption", e);
           }
            LOG.info("EventThread shut down");
        }

处理回调函数和watcher

private void processEvent(Object event) {
          try {
              //处理watcher
              if (event instanceof WatcherSetEventPair) {
                  WatcherSetEventPair pair = (WatcherSetEventPair) event;
                  for (Watcher watcher : pair.watchers) {
                      try {
                          watcher.process(pair.event);
                      } catch (Throwable t) {
                          LOG.error("Error while calling watcher ", t);
                      }
                  }
              } else {
                  //
                  Packet p = (Packet) event;
                  int rc = 0;
                  String clientPath = p.clientPath;
                  if (p.replyHeader.getErr() != 0) {
                      rc = p.replyHeader.getErr();
                  }
                  if (p.response instanceof ExistsResponse
                          || p.response instanceof SetDataResponse
                          || p.response instanceof SetACLResponse) {
                      StatCallback cb = (StatCallback) p.cb;
                      if (rc == 0) {
                          if (p.response instanceof ExistsResponse) {
                              cb.processResult(rc, clientPath, p.ctx,
                                      ((ExistsResponse) p.response)
                                              .getStat());
                          } else if (p.response instanceof SetDataResponse) {
                              cb.processResult(rc, clientPath, p.ctx,
                                      ((SetDataResponse) p.response)
                                              .getStat());
                          } else if (p.response instanceof SetACLResponse) {
                              cb.processResult(rc, clientPath, p.ctx,
                                      ((SetACLResponse) p.response)
                                              .getStat());
                          }
                      } else {
                          cb.processResult(rc, clientPath, p.ctx, null);
                      }
                  } else if (p.response instanceof GetDataResponse) {
                      DataCallback cb = (DataCallback) p.cb;
                      GetDataResponse rsp = (GetDataResponse) p.response;
                      if (rc == 0) {
                          cb.processResult(rc, clientPath, p.ctx, rsp
                                  .getData(), rsp.getStat());
                      } else {
                          cb.processResult(rc, clientPath, p.ctx, null,
                                  null);
                      }
                  } else if (p.response instanceof GetACLResponse) {
                      ACLCallback cb = (ACLCallback) p.cb;
                      GetACLResponse rsp = (GetACLResponse) p.response;
                      if (rc == 0) {
                          cb.processResult(rc, clientPath, p.ctx, rsp
                                  .getAcl(), rsp.getStat());
                      } else {
                          cb.processResult(rc, clientPath, p.ctx, null,
                                  null);
                      }
                  } else if (p.response instanceof GetChildrenResponse) {
                      ChildrenCallback cb = (ChildrenCallback) p.cb;
                      GetChildrenResponse rsp = (GetChildrenResponse) p.response;
                      if (rc == 0) {
                          cb.processResult(rc, clientPath, p.ctx, rsp
                                  .getChildren());
                      } else {
                          cb.processResult(rc, clientPath, p.ctx, null);
                      }
                  } else if (p.response instanceof GetChildren2Response) {
                      Children2Callback cb = (Children2Callback) p.cb;
                      GetChildren2Response rsp = (GetChildren2Response) p.response;
                      if (rc == 0) {
                          cb.processResult(rc, clientPath, p.ctx, rsp
                                  .getChildren(), rsp.getStat());
                      } else {
                          cb.processResult(rc, clientPath, p.ctx, null, null);
                      }
                  } else if (p.response instanceof CreateResponse) {
                      StringCallback cb = (StringCallback) p.cb;
                      CreateResponse rsp = (CreateResponse) p.response;
                      if (rc == 0) {
                          cb.processResult(rc, clientPath, p.ctx,
                                  (chrootPath == null
                                          ? rsp.getPath()
                                          : rsp.getPath()
                                    .substring(chrootPath.length())));
                      } else {
                          cb.processResult(rc, clientPath, p.ctx, null);
                      }
                  } else if (p.cb instanceof VoidCallback) {
                      VoidCallback cb = (VoidCallback) p.cb;
                      cb.processResult(rc, clientPath, p.ctx);
                  }
              }
          } catch (Throwable t) {
              LOG.error("Caught unexpected throwable", t);
          }
       }

Packet结构

　　包，ClientCnxn内部管理请求内容的模块。由以下几个模块组成：

　　1.RequestHeader header 请求头

　　2.Record request 请求内容

　　3.ByteBuffer bb 实际需要发送的请求内容。

　　4.ReplyHeader replyHeader 响应头

　　5.Record response 响应内容

　　6.String clientPath

　　7.String serverPath 

　　8.boolean finished

　　9.AsyncCallback cb

　　10.Object ctx 

　　11.WatchRegistration watchRegistration 

Packet(RequestHeader header, ReplyHeader replyHeader, Record record,
                Record response, ByteBuffer bb,
                WatchRegistration watchRegistration) {
            this.header = header;
            this.replyHeader = replyHeader;
            this.request = record;
            this.response = response;
            if (bb != null) {
                this.bb = bb;
            } else {
                try {
                    ByteArrayOutputStream baos = new ByteArrayOutputStream();
                    BinaryOutputArchive boa = BinaryOutputArchive
                            .getArchive(baos);
                    boa.writeInt(-1, "len"); // We'll fill this in later
                    header.serialize(boa, "header");
                    if (record != null) {
                        record.serialize(boa, "request");
                    }
                    baos.close();
                    this.bb = ByteBuffer.wrap(baos.toByteArray());
                    this.bb.putInt(this.bb.capacity() - 4);
                    this.bb.rewind();
                } catch (IOException e) {
                    LOG.warn("Ignoring unexpected exception", e);
                }
            }
            this.watchRegistration = watchRegistration;
        }