zookeeper 源码（二） session 和 处理事务请求
问题
- session 如何生成的？sessionId为什么不直接使用时间戳+单机名
- sessionid 关闭的时候的逻辑，sessionid 的维护是由各节点还是leader ?
会话相关
sessionid 生成
我们看一下session 管理类 SessionTrackerImpl。 它主要维护三个字段
//根据 sessionid 存放的 session
HashMap<Long, SessionImpl> sessionsById = new HashMap<Long, SessionImpl>();
//key 是时间 ，用于根据下次会话超时时间点来归纳会话，便于进行会话管理和超时检查，（分桶策略容器）
HashMap<Long, SessionSet> sessionSets = new HashMap<Long, SessionSet>();
//key 是 sessionId , value 是过期时间
ConcurrentHashMap<Long, Integer> sessionsWithTimeout;
其他的都是操作 session 的方法，我们看一下创建sessionId的过程,SessionImpl 类
public static long initializeNextSession(long id) {
long nextSid = 0;
nextSid = (Time.currentElapsedTime() << 24) >>> 8;
nextSid = nextSid | (id <<56);
return nextSid;
}
得出的 sessionid前8位确定所在的机器，后56位使用当前时间的毫秒数表示进行随机。
session 管理
使用的是分桶策略，如下图所示。
以时间戳为节点，每个桶装这该时间点（过期的时间）的 session集合，然后有一个线程那桶里的多个session 进行检查，加入过期时间被延长，那么session 进行迁移到其他的桶，否则将被清理。
创建会话执行事务过程
创建会话之前我们要先知道zk服务器是如何和客户端连接的。上一篇文章结尾处，我们知道了处理与 客户端的连接主要是由 NIOServerCnxnFactory 来负责的。而真正处理的逻辑就在 run 方法。
/**
* 默认允许连接 60个客户端，
* 这个（一个）线程会处理
* - 来自客户端的连接
* - 来自客户端的读
* - 来自客户端的写
*
*
*/
public void run() {
//只要没有断开，循环一直进行
//下面连接就是我们熟悉的 java NIO 的运用
while (!ss.socket().isClosed()) {
try {
//select 方法一直就阻塞
selector.select(1000);
Set<SelectionKey> selected;
//这里为什么要加锁呢？可能有多个线程
synchronized (this) {
selected = selector.selectedKeys();
}
ArrayList<SelectionKey> selectedList = new ArrayList<SelectionKey>(
selected);
Collections.shuffle(selectedList);
for (SelectionKey k : selectedList) {
if ((k.readyOps() & SelectionKey.OP_ACCEPT) != 0) {
SocketChannel sc = ((ServerSocketChannel) k
.channel()).accept();
InetAddress ia = sc.socket().getInetAddress();
int cnxncount = getClientCnxnCount(ia);
//客户端可以保存 60 个连接
if (maxClientCnxns > 0 && cnxncount >= maxClientCnxns){
LOG.warn("Too many connections from " + ia
+ " - max is " + maxClientCnxns );
sc.close();
} else {
LOG.info("Accepted socket connection from "
+ sc.socket().getRemoteSocketAddress());
sc.configureBlocking(false);
SelectionKey sk = sc.register(selector,
SelectionKey.OP_READ);
//创建连接，连接用 NIOServerCnxn这个类来维护
NIOServerCnxn cnxn = createConnection(sc, sk);
sk.attach(cnxn);
addCnxn(cnxn);
}
} else if ((k.readyOps() & (SelectionKey.OP_READ | SelectionKey.OP_WRITE)) != 0) {
//处理客户端读写操作，重点看 doIO 方法
NIOServerCnxn c = (NIOServerCnxn) k.attachment();
c.doIO(k);
} else {
if (LOG.isDebugEnabled()) {
LOG.debug("Unexpected ops in select "
+ k.readyOps());
}
}
}
selected.clear();
} catch (RuntimeException e) {
LOG.warn("Ignoring unexpected runtime exception", e);
} catch (Exception e) {
LOG.warn("Ignoring exception", e);
}
}
closeAll();
LOG.info("NIOServerCnxn factory exited run method");
}
我们再来看一下 doIO这个方法 ， 位于 NIOServerCnxn 内
/**
* Handles read/write IO on connection.
*/
void doIO(SelectionKey k) throws InterruptedException {
try {
if (isSocketOpen() == false) {
LOG.warn("trying to do i/o on a null socket for session:0x"
+ Long.toHexString(sessionId));
return;
}
//可读类型
if (k.isReadable()) {
int rc = sock.read(incomingBuffer);
if (rc < 0) {
throw new EndOfStreamException(
"Unable to read additional data from client sessionid 0x"
+ Long.toHexString(sessionId)
+ ", likely client has closed socket");
}
if (incomingBuffer.remaining() == 0) {
boolean isPayload;
//读取下一个请求
if (incomingBuffer == lenBuffer) { // start of next request
//翻转缓存区，可读
incomingBuffer.flip();
//读取lenBuffer的前四个字节，当读取的是内容长度时则为true，否则为false
isPayload = readLength(k);
//清空缓存
incomingBuffer.clear();
} else {
// continuation
isPayload = true;
}
// isPayload 为 true ,表示buffer 里面是负载
if (isPayload) { // not the case for 4letterword
readPayload();
}
else {
// four letter words take care
// need not do anything else
return;
}
}
}
//可写类型
if (k.isWritable()) {
// ZooLog.logTraceMessage(LOG,
// ZooLog.CLIENT_DATA_PACKET_TRACE_MASK
// "outgoingBuffers.size() = " +
// outgoingBuffers.size());
if (outgoingBuffers.size() > 0) {
// ZooLog.logTraceMessage(LOG,
// ZooLog.CLIENT_DATA_PACKET_TRACE_MASK,
// "sk " + k + " is valid: " +
// k.isValid());
/*
* This is going to reset the buffer position to 0 and the
* limit to the size of the buffer, so that we can fill it
* with data from the non-direct buffers that we need to
* send.
*/
ByteBuffer directBuffer = factory.directBuffer;
directBuffer.clear();
for (ByteBuffer b : outgoingBuffers) {
if (directBuffer.remaining() < b.remaining()) {
/*
* When we call put later, if the directBuffer is to
* small to hold everything, nothing will be copied,
* so we've got to slice the buffer if it's too big.
*/
b = (ByteBuffer) b.slice().limit(
directBuffer.remaining());
}
/*
* put() is going to modify the positions of both
* buffers, put we don't want to change the position of
* the source buffers (we'll do that after the send, if
* needed), so we save and reset the position after the
* copy
*/
int p = b.position();
directBuffer.put(b);
b.position(p);
if (directBuffer.remaining() == 0) {
break;
}
}
/*
* Do the flip: limit becomes position, position gets set to
* 0. This sets us up for the write.
*/
directBuffer.flip();
int sent = sock.write(directBuffer);
ByteBuffer bb;
// Remove the buffers that we have sent
while (outgoingBuffers.size() > 0) {
bb = outgoingBuffers.peek();
if (bb == ServerCnxnFactory.closeConn) {
throw new CloseRequestException("close requested");
}
int left = bb.remaining() - sent;
if (left > 0) {
/*
* We only partially sent this buffer, so we update
* the position and exit the loop.
*/
bb.position(bb.position() + sent);
break;
}
packetSent();
/* We've sent the whole buffer, so drop the buffer */
sent -= bb.remaining();
outgoingBuffers.remove();
}
// ZooLog.logTraceMessage(LOG,
// ZooLog.CLIENT_DATA_PACKET_TRACE_MASK, "after send,
// outgoingBuffers.size() = " + outgoingBuffers.size());
}
synchronized(this.factory){
if (outgoingBuffers.size() == 0) {
if (!initialized
&& (sk.interestOps() & SelectionKey.OP_READ) == 0) {
throw new CloseRequestException("responded to info probe");
}
sk.interestOps(sk.interestOps()
& (~SelectionKey.OP_WRITE));
} else {
sk.interestOps(sk.interestOps()
| SelectionKey.OP_WRITE);
}
}
}
} catch (CancelledKeyException e) {
LOG.warn("CancelledKeyException causing close of session 0x"
+ Long.toHexString(sessionId));
if (LOG.isDebugEnabled()) {
LOG.debug("CancelledKeyException stack trace", e);
}
close();
} catch (CloseRequestException e) {
// expecting close to log session closure
close();
} catch (EndOfStreamException e) {
LOG.warn(e.getMessage());
if (LOG.isDebugEnabled()) {
LOG.debug("EndOfStreamException stack trace", e);
}
// expecting close to log session closure
close();
} catch (IOException e) {
LOG.warn("Exception causing close of session 0x"
+ Long.toHexString(sessionId) + ": " + e.getMessage());
if (LOG.isDebugEnabled()) {
LOG.debug("IOException stack trace", e);
}
close();
}
}
/** Read the request payload (everything following the length prefix) */
private void readPayload() throws IOException, InterruptedException {
if (incomingBuffer.remaining() != 0) { // have we read length bytes?
int rc = sock.read(incomingBuffer); // sock is non-blocking, so ok
if (rc < 0) {
throw new EndOfStreamException(
"Unable to read additional data from client sessionid 0x"
+ Long.toHexString(sessionId)
+ ", likely client has closed socket");
}
}
if (incomingBuffer.remaining() == 0) { // have we read length bytes?
packetReceived();
incomingBuffer.flip();
//执行读取负载的逻辑
if (!initialized) {
//非初始化
readConnectRequest();
} else {
readRequest();
}
lenBuffer.clear();
incomingBuffer = lenBuffer;
}
}
其中我们看一下如何连接的，即上面的 readConnectRequest 方法
private void readConnectRequest() throws IOException, InterruptedException {
if (!isZKServerRunning()) {
throw new IOException("ZooKeeperServer not running");
}
zkServer.processConnectRequest(this, incomingBuffer);
initialized = true;
}
于是我们到了ZookeeperServer 的 processConnectRequest 方法
/**
* 封装成一个 connectRequest
* 提交请求到leader
*/
public void processConnectRequest(ServerCnxn cnxn, ByteBuffer incomingBuffer) throws IOException {
BinaryInputArchive bia = BinaryInputArchive.getArchive(new ByteBufferInputStream(incomingBuffer));
ConnectRequest connReq = new ConnectRequest();
connReq.deserialize(bia, "connect");
if (LOG.isDebugEnabled()) {
LOG.debug("Session establishment request from client "
+ cnxn.getRemoteSocketAddress()
+ " client's lastZxid is 0x"
+ Long.toHexString(connReq.getLastZxidSeen()));
}
boolean readOnly = false;
try {
readOnly = bia.readBool("readOnly");
cnxn.isOldClient = false;
} catch (IOException e) {
// this is ok -- just a packet from an old client which
// doesn't contain readOnly field
LOG.warn("Connection request from old client "
+ cnxn.getRemoteSocketAddress()
+ "; will be dropped if server is in r-o mode");
}
if (readOnly == false && this instanceof ReadOnlyZooKeeperServer) {
String msg = "Refusing session request for not-read-only client "
+ cnxn.getRemoteSocketAddress();
LOG.info(msg);
throw new CloseRequestException(msg);
}
if (connReq.getLastZxidSeen() > zkDb.dataTree.lastProcessedZxid) {
String msg = "Refusing session request for client "
+ cnxn.getRemoteSocketAddress()
+ " as it has seen zxid 0x"
+ Long.toHexString(connReq.getLastZxidSeen())
+ " our last zxid is 0x"
+ Long.toHexString(getZKDatabase().getDataTreeLastProcessedZxid())
+ " client must try another server";
LOG.info(msg);
throw new CloseRequestException(msg);
}
int sessionTimeout = connReq.getTimeOut();
byte passwd[] = connReq.getPasswd();
int minSessionTimeout = getMinSessionTimeout();
if (sessionTimeout < minSessionTimeout) {
sessionTimeout = minSessionTimeout;
}
int maxSessionTimeout = getMaxSessionTimeout();
if (sessionTimeout > maxSessionTimeout) {
sessionTimeout = maxSessionTimeout;
}
cnxn.setSessionTimeout(sessionTimeout);
// We don't want to receive any packets until we are sure that the
// session is setup
cnxn.disableRecv();
long sessionId = connReq.getSessionId();
if (sessionId != 0) {
long clientSessionId = connReq.getSessionId();
LOG.info("Client attempting to renew session 0x"
+ Long.toHexString(clientSessionId)
+ " at " + cnxn.getRemoteSocketAddress());
serverCnxnFactory.closeSession(sessionId);
cnxn.setSessionId(sessionId);
reopenSession(cnxn, sessionId, passwd, sessionTimeout);
} else {
LOG.info("Client attempting to establish new session at "
+ cnxn.getRemoteSocketAddress());
//真正执行请求的地方 : 创建 session,提交请求
createSession(cnxn, passwd, sessionTimeout);
}
}
long createSession(ServerCnxn cnxn, byte passwd[], int timeout) {
long sessionId = sessionTracker.createSession(timeout);
Random r = new Random(sessionId ^ superSecret);
r.nextBytes(passwd);
ByteBuffer to = ByteBuffer.allocate(4);
to.putInt(timeout);
cnxn.setSessionId(sessionId);
//提交请求
submitRequest(cnxn, sessionId, OpCode.createSession, 0, to, null);
return sessionId;
}
/**
* @param cnxn
* @param sessionId
* @param xid
* @param bb
*/
private void submitRequest(ServerCnxn cnxn, long sessionId, int type,
int xid, ByteBuffer bb, List<Id> authInfo) {
Request si = new Request(cnxn, sessionId, xid, type, bb, authInfo);
submitRequest(si);
}
public void submitRequest(Request si) {
if (firstProcessor == null) {
synchronized (this) {
try {
// Since all requests are passed to the request
// processor it should wait for setting up the request
// processor chain. The state will be updated to RUNNING
// after the setup.
// 在 startup 方法中直到初始化完成才得以接受请求
while (state == State.INITIAL) {
wait(1000);
}
} catch (InterruptedException e) {
LOG.warn("Unexpected interruption", e);
}
if (firstProcessor == null || state != State.RUNNING) {
throw new RuntimeException("Not started");
}
}
}
try {
//判断 session 是否还存活
touch(si.cnxn);
boolean validpacket = Request.isValid(si.type);
if (validpacket) {
//firstProcessor 开始执行
firstProcessor.processRequest(si);
if (si.cnxn != null) {
incInProcess();
}
} else {
LOG.warn("Received packet at server of unknown type " + si.type);
new UnimplementedRequestProcessor().processRequest(si);
}
} catch (MissingSessionException e) {
if (LOG.isDebugEnabled()) {
LOG.debug("Dropping request: " + e.getMessage());
}
} catch (RequestProcessorException e) {
LOG.error("Unable to process request:" + e.getMessage(), e);
}
}
清楚了服务器与客户端的连接，知道了最终会到执行任务链上。
我们可以连接任意一台 zk服务器，进行事务请求，当follower接受到请求后它就会转发给leader ,而follower 和 leader 都是使用责任链进行处理来自客户端的请求的。
我们先来看一下follower 和 leader 责任链的流程。
请求处理流程
会话创建请求
我们通过下面的图片先来了解整个过程
概括起来就是 ：
- NIOServerCnxn 接受请求
- 协商 sessionTimeout ,创建 connectRequest
- 创建会话，生成 sessionId,注册会话和激活会话
- 交给 leader 的 PrepRequestProcessor
- 创建请求事务体 createSessionTxn
- 交给 ProposalRequestProcessor,接下来就会进入三个子流程
子流程如下 ： - Sync流程，follower做好日志记录，同时返回 ACK 给leader
- Proposal流程，生成proposal ,广播提议，获得半数票后，请求加入到 toBeApplied 队列，广播commit 信息
- Commit流程，将请求交付给CommitProcessor 处理器，等待上阶段的投票结果，提交请求，交付给下一个处理器 : FinalRequestProcessor
- 到此三个子流程走完后，到了最后的阶段，事务应用，之前我们的议案只是应用在日志中，并没有在内存中生效，这阶段需要将请求应用在内存中。
这个就是整个会话创建请求的处理过程，当客户端发出事务请求时也是像处理会话创建一样的流程。
总结
这一篇主要讲了zk 创建session的策略和管理的逻辑，同时介绍了处理事务的过程。
zookeeper 源码（二） session 和 处理事务请求的更多相关文章
- zookeeper源码 — 二、集群启动—leader选举
上一篇介绍了zookeeper的单机启动,集群模式下启动和单机启动有相似的地方,但是也有各自的特点.集群模式的配置方式和单机模式也是不一样的,这一篇主要包含以下内容: 概念介绍:角色,服务器状态 服务 ...
- ZooKeeper源码阅读——client（二）
原创技术文章,转载请注明:转自http://newliferen.github.io/ 如何连接ZooKeeper集群 要想了解ZooKeeper客户端实现原理,首先需要关注一下客户端的使用方式, ...
- Zookeeper 源码（二）序列化组件 Jute
Zookeeper 源码(二)序列化组件 Jute 一.序列化组件 Jute 对于一个网络通信,首先需要解决的就是对数据的序列化和反序列化处理,在 ZooKeeper 中,使用了Jute 这一序列化组 ...
- zookeeper源码分析之五服务端(集群leader)处理请求流程
leader的实现类为LeaderZooKeeperServer,它间接继承自标准ZookeeperServer.它规定了请求到达leader时需要经历的路径: PrepRequestProcesso ...
- zookeeper源码分析之四服务端(单机)处理请求流程
上文: zookeeper源码分析之一服务端启动过程 中,我们介绍了zookeeper服务器的启动过程,其中单机是ZookeeperServer启动,集群使用QuorumPeer启动,那么这次我们分析 ...
- Zookeeper 源码（七）请求处理
Zookeeper 源码(七)请求处理 以单机启动为例讲解 Zookeeper 是如何处理请求的.先回顾一下单机时的请求处理链. // 单机包含 3 个请求链:PrepRequestProcessor ...
- Zookeeper 源码（六）Leader-Follower-Observer
Zookeeper 源码(六)Leader-Follower-Observer 上一节介绍了 Leader 选举的全过程,本节讲解一下 Leader-Follower-Observer 服务器的三种角 ...
- Zookeeper 源码（五）Leader 选举
Zookeeper 源码(五)Leader 选举 前面学习了 Zookeeper 服务端的相关细节,其中对于集群启动而言,很重要的一部分就是 Leader 选举,接着就开始深入学习 Leader 选举 ...
- Zookeeper 源码（四）Zookeeper 服务端源码
Zookeeper 源码(四)Zookeeper 服务端源码 Zookeeper 服务端的启动入口为 QuorumPeerMain public static void main(String[] a ...
- Zookeeper 源码（三）Zookeeper 客户端源码
Zookeeper 源码(三)Zookeeper 客户端源码 Zookeeper 客户端主要有以下几个重要的组件.客户端会话创建可以分为三个阶段:一是初始化阶段.二是会话创建阶段.三是响应处理阶段. ...
随机推荐
- VS2017连接MySQL数据库
vs默认无法直接连接mysql,需要我们自己配置环境. 1.下载mysql-installer-community-8.0.18.0.msi 下载地址:https://dev.mysql.com/do ...
- uevent机制
uevent, user space event. 内核与用户空间的一种通信机制,基于netlink机制,主要用于设备驱动模型,例如热插拔. 1.调用/sbin/mdev的流程分析 在驱动程序中经常出 ...
- HITCON-Training-Writeup
HITCON-Training-Writeup 原文链接M4x@10.0.0.55 项目地址M4x's github,欢迎star~ 更新时间5月16 复习一下二进制基础,写写HITCON-Train ...
- LitElement(三)Styles样式
1.为组件添加样式 import { LitElement, css, html } from 'lit-element'; class MyElement extends LitElement { ...
- 三分钟快速上手TensorFlow 2.0 （中）——常用模块和模型的部署
本文学习笔记参照来源:https://tf.wiki/zh/basic/basic.html 前文:三分钟快速上手TensorFlow 2.0 (上)——前置基础.模型建立与可视化 tf.train. ...
- Python_内置函数和匿名函数
楔子 在讲新知识之前,我们先来复习复习函数的基础知识. 问:函数怎么调用? 函数名() 如果你们这么说...那你们就对了!好了记住这个事儿别给忘记了,咱们继续谈下一话题... 来你们在自己的环境里打印 ...
- Java之字符串替换replace（）
replace(char oldChar, char newChar)返回一个新的字符串,它是通过用 newChar 替换此字符串中出现的所有 oldChar 而生成的 import java.uti ...
- (转)json格式转换成javaBean对象的方法
把json格式转换成javaBean才可以.于是查了一下资料,网上最多的资料就是下面的这种方式: Java code? 1 2 3 4 5 6 7 8 9 String str = "[{\ ...
- LCA 倍增算法模板
. #include <cstring> #include <cstdio> #include <cstdlib> #include <algorithm&g ...
- Apache Kafka（十）Partitions与Replication Factor 调整准则
Partitions与Replication Factor调整准则 Partition 数目与Replication Factor是在创建一个topic时非常重要的两个参数,这两个参数的取值会直接影响 ...