07、ZooKeeper ZookeeperServer
继承层次
在之前分析zookeeper各个角色初始化的过程中，看到了每个角色都会创建一个相应的ZookeeperServer
ZookeeperServer代表服务器
QuorumZookeeperServer代表参与选举的服务器
LearnerZookeeperServer代表非Leader服务器
LeaderZookeeperServer代表Leader节点服务器
FollowerZookeeperServer代表Follower节点服务器
ObserverZookeeperServer代表Observer节点服务器
ReadOnlyZookeeperServer代表只读节点服务器
ZookeeperServer
public static interface SessionExpirer {
// 过期指定的session
void expire(Session session);
// 获取当前节点的id
long getServerId();
}
// 获取当前server的一些统计信息
public interface Provider {
public long getOutstandingRequests();
public long getLastProcessedZxid();
public String getState();
public int getNumAliveConnections();
public long getDataDirSize();
public long getLogDirSize();
}
public class ZooKeeperServer implements SessionExpirer, ServerStats.Provider {
}
因为ZookeeperServer实现了SessionExpirer和ServerStatus.Provider接口，因此具有过期会话和查询当前服务器信息的能力
内部类
ChangeRecord
// ChangeRecord
static class ChangeRecord {
ChangeRecord(long zxid, String path, StatPersisted stat, int childCount,
List<ACL> acl) {
this.zxid = zxid;
this.path = path;
this.stat = stat;
this.childCount = childCount;
this.acl = acl;
}
// 事务id
long zxid;
// 记录的路径
String path;
StatPersisted stat; /* Make sure to create a new object when changing */
int childCount;
List<ACL> acl; /* Make sure to create a new object when changing */
// 拷贝
ChangeRecord duplicate(long zxid) {
StatPersisted stat = new StatPersisted();
if (this.stat != null) {
DataTree.copyStatPersisted(this.stat, stat);
}
return new ChangeRecord(zxid, path, stat, childCount,
acl == null ? new ArrayList<ACL>() : new ArrayList<ACL>(acl));
}
}
ChangeRecord用来在PrepRequestProcessor和FinalRequestProcessor之间传递信息
MissingSessionException
// MissingSessionException
public static class MissingSessionException extends IOException {
private static final long serialVersionUID = 7467414635467261007L;
public MissingSessionException(String msg) {
super(msg);
}
}
MissingSessionException代表事务丢失
State
// State
protected enum State {
INITIAL, RUNNING, SHUTDOWN, ERROR
}
State代表当前状态
主要属性
// jmx bean
protected ZooKeeperServerBean jmxServerBean;
protected DataTreeBean jmxDataTreeBean;
// 默认的心跳时间
public static final int DEFAULT_TICK_TIME = 3000;
protected int tickTime = DEFAULT_TICK_TIME;
// 最小session过期时间
/** value of -1 indicates unset, use default */
protected int minSessionTimeout = -1;
// 最大session过期时间
/** value of -1 indicates unset, use default */
protected int maxSessionTimeout = -1;
// 事务跟踪器
protected SessionTracker sessionTracker;
// 事务日志
private FileTxnSnapLog txnLogFactory = null;
// zookeeper内部数据库
private ZKDatabase zkDb;
// 请求处理器
protected RequestProcessor firstProcessor;
// 当前状态
protected volatile State state = State.INITIAL;
private final AtomicInteger requestsInProcess = new AtomicInteger(0);
// 未处理changeRecord
final Deque<ChangeRecord> outstandingChanges = new ArrayDeque<>();
// this data structure must be accessed under the outstandingChanges lock
// 未处理changeRecord和其path之间的映射
final HashMap<String, ChangeRecord> outstandingChangesForPath =
new HashMap<String, ChangeRecord>();
// 连接管理工厂
protected ServerCnxnFactory serverCnxnFactory;
protected ServerCnxnFactory secureServerCnxnFactory;
// 服务器统计信息
private final ServerStats serverStats;
// 当有重要的线程异常时会回调这个listener
private final ZooKeeperServerListener listener;
构造方法
默认构造方法
public ZooKeeperServer() {
serverStats = new ServerStats(this);
listener = new ZooKeeperServerListenerImpl(this);
}
// ZooKeeperServerListenerImpl
class ZooKeeperServerListenerImpl implements ZooKeeperServerListener {
private static final Logger LOG = LoggerFactory
.getLogger(ZooKeeperServerListenerImpl.class);
private final ZooKeeperServer zkServer;
ZooKeeperServerListenerImpl(ZooKeeperServer zkServer) {
this.zkServer = zkServer;
}
@Override
public void notifyStopping(String threadName, int exitCode) {
LOG.info("Thread {} exits, error code {}", threadName, exitCode);
zkServer.setState(State.ERROR);
}
}
ZooKeeperServerListenerImpl默认的ZooKeeperServerListener的实现类，当重要线程出现致命错误时，会通知该类，该类会将当前的服务器状态设置为失败
ZooKeeperServer(FileTxnSnapLog txnLogFactory, int tickTime, int minSessionTimeout, int maxSessionTimeout, ZKDatabase zkDb)
public ZooKeeperServer(FileTxnSnapLog txnLogFactory, int tickTime,
int minSessionTimeout, int maxSessionTimeout, ZKDatabase zkDb) {
serverStats = new ServerStats(this);
// 设置事务日志
this.txnLogFactory = txnLogFactory;
this.txnLogFactory.setServerStats(this.serverStats);
this.zkDb = zkDb;
this.tickTime = tickTime;
// 如果为-1，会设置为2倍的心跳时间
setMinSessionTimeout(minSessionTimeout);
// 如果为-1，会设置为20倍的心跳时间
setMaxSessionTimeout(maxSessionTimeout);
listener = new ZooKeeperServerListenerImpl(this);
LOG.info("Created server with tickTime " + tickTime
+ " minSessionTimeout " + getMinSessionTimeout()
+ " maxSessionTimeout " + getMaxSessionTimeout()
+ " datadir " + txnLogFactory.getDataDir()
+ " snapdir " + txnLogFactory.getSnapDir());
}
ZooKeeperServer(FileTxnSnapLog txnLogFactory, int tickTime)
// ZooKeeperServer
public ZooKeeperServer(FileTxnSnapLog txnLogFactory, int tickTime)
throws IOException {
// 指定心跳时间
// 最小事务过期时间和最大事务过期时间设置为默认值
this(txnLogFactory, tickTime, -1, -1, new ZKDatabase(txnLogFactory));
}
ZooKeeperServer(FileTxnSnapLog txnLogFactory)
public ZooKeeperServer(FileTxnSnapLog txnLogFactory)
throws IOException
{
// 使用默认心跳时间
// 最小事务过期时间和最大事务过期时间设置为默认值
this(txnLogFactory, DEFAULT_TICK_TIME, -1, -1, new ZKDatabase(txnLogFactory));
}
ZooKeeperServer(File snapDir, File logDir, int tickTime)
public ZooKeeperServer(File snapDir, File logDir, int tickTime)
throws IOException {
// 使用指定的快照目录和事务日志目录创建文件快照日志
this( new FileTxnSnapLog(snapDir, logDir),
tickTime);
}
重要方法
startup
public synchronized void startup() {
// 1. 创建SessionTracker
if (sessionTracker == null) {
createSessionTracker();
}
// 2. 启动SessionTracker
startSessionTracker();
// 3. 设置请求处理链
setupRequestProcessors();
// 4. 注册jmx
registerJMX();
// 5. 设置当前状态为running
setState(State.RUNNING);
notifyAll();
}
protected void createSessionTracker() {
sessionTracker = new SessionTrackerImpl(this, zkDb.getSessionWithTimeOuts(),
tickTime, createSessionTrackerServerId, getZooKeeperServerListener());
}
setupRequestProcessor
setupRequestProcessor用来设置请求处理链
// ZookeeperServer
protected void setupRequestProcessors() {
RequestProcessor finalProcessor = new FinalRequestProcessor(this);
RequestProcessor syncProcessor = new SyncRequestProcessor(this,
finalProcessor);
((SyncRequestProcessor)syncProcessor).start();
firstProcessor = new PrepRequestProcessor(this, syncProcessor);
((PrepRequestProcessor)firstProcessor).start();
}
loadData
public void loadData() throws IOException, InterruptedException {
/*
* When a new leader starts executing Leader#lead, it
* invokes this method. The database, however, has been
* initialized before running leader election so that
* the server could pick its zxid for its initial vote.
* It does it by invoking QuorumPeer#getLastLoggedZxid.
* Consequently, we don't need to initialize it once more
* and avoid the penalty of loading it a second time. Not
* reloading it is particularly important for applications
* that host a large database.
*
* The following if block checks whether the database has
* been initialized or not. Note that this method is
* invoked by at least one other method:
* ZooKeeperServer#startdata.
*
* See ZOOKEEPER-1642 for more detail.
*/
// 1. 设置事务id
// 判断内存数据库是否初始化过
if(zkDb.isInitialized()){
// 初始化过，使用上一次处理过的事务id
setZxid(zkDb.getDataTreeLastProcessedZxid());
}
else {
// 没有初始化过，使用快照文件和事务日志文件来恢复内存数据库
setZxid(zkDb.loadDataBase());
}
// Clean up dead sessions
// 2. 杀死过期的事务
LinkedList<Long> deadSessions = new LinkedList<Long>();
for (Long session : zkDb.getSessions()) {
if (zkDb.getSessionWithTimeOuts().get(session) == null) {
deadSessions.add(session);
}
}
for (long session : deadSessions) {
// XXX: Is lastProcessedZxid really the best thing to use?
// 清除事务
killSession(session, zkDb.getDataTreeLastProcessedZxid());
}
// Make a clean snapshot
// 3. 进行快照
takeSnapshot();
}
// ZookeeperServer
protected void killSession(long sessionId, long zxid) {
// 1. 修改内存数据库，将当前session创建的临时节点删除
zkDb.killSession(sessionId, zxid);
if (LOG.isTraceEnabled()) {
ZooTrace.logTraceMessage(LOG, ZooTrace.SESSION_TRACE_MASK,
"ZooKeeperServer --- killSession: 0x"
+ Long.toHexString(sessionId));
}
// 从sessionTracker中移除当前事务
if (sessionTracker != null) {
sessionTracker.removeSession(sessionId);
}
}
submitRequest
public void submitRequest(Request si) {
if (firstProcessor == null) {
synchronized (this) {
try {
// 当设置完请求处理链之后，会将状态设置为running，这里会循环等待设置的完成
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 {
// 更新事务过期时间
touch(si.cnxn);
// 验证请求类型是否合法
boolean validpacket = Request.isValid(si.type);
if (validpacket) {
// 将请求交给请求处理链处理
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);
}
}
LeaderZookeeperServer
Leader节点会使用LeaderZookeeperServer作为自己的服务器实现
重要方法
setupRequestProcessor
// LeaderZookeeperServer
protected void setupRequestProcessors() {
RequestProcessor finalProcessor = new FinalRequestProcessor(this);
RequestProcessor toBeAppliedProcessor = new Leader.ToBeAppliedRequestProcessor(finalProcessor, getLeader());
commitProcessor = new CommitProcessor(toBeAppliedProcessor,
Long.toString(getServerId()), false,
getZooKeeperServerListener());
commitProcessor.start();
ProposalRequestProcessor proposalProcessor = new ProposalRequestProcessor(this,
commitProcessor);
proposalProcessor.initialize();
prepRequestProcessor = new PrepRequestProcessor(this, proposalProcessor);
prepRequestProcessor.start();
firstProcessor = new LeaderRequestProcessor(this, prepRequestProcessor);
setupContainerManager();
}
FollowerZookeeperServer
Follower节点使用的是FollowerZookeeperServer作为自己的服务器实现
主要属性
// 待同步请求
ConcurrentLinkedQueue<Request> pendingSyncs;
// 待处理的事务
LinkedBlockingQueue<Request> pendingTxns = new LinkedBlockingQueue<Request>();
重要方法
setupRequestProcessors
protected void setupRequestProcessors() {
RequestProcessor finalProcessor = new FinalRequestProcessor(this);
commitProcessor = new CommitProcessor(finalProcessor,
Long.toString(getServerId()), true, getZooKeeperServerListener());
commitProcessor.start();
firstProcessor = new FollowerRequestProcessor(this, commitProcessor);
((FollowerRequestProcessor) firstProcessor).start();
syncProcessor = new SyncRequestProcessor(this,
new SendAckRequestProcessor((Learner)getFollower()));
syncProcessor.start();
}
logRequest
public void logRequest(TxnHeader hdr, Record txn) {
Request request = new Request(hdr.getClientId(), hdr.getCxid(), hdr.getType(), hdr, txn, hdr.getZxid());
if ((request.zxid & 0xffffffffL) != 0) {
pendingTxns.add(request);
}
// 将请求交给syncProcessor处理，SyncProcessor会将请求日志落盘
syncProcessor.processRequest(request);
}
当Follower接收到Leader端发来的proposal请求时，会调用logRequest来记录事务日志
sync
// LeaderZookeeperServer
synchronized public void sync(){
// 当前没有待同步请求
if(pendingSyncs.size() ==0){
LOG.warn("Not expecting a sync.");
return;
}
// 取出待同步请求，交给commitProcessor进行处理
Request r = pendingSyncs.remove();
commitProcessor.commit(r);
}
当客户端向follower发送sync请求，想要follower和leader数据进行同步，FollowerRequestProcessor处理到该请求，会调用sync方法
public void run() {
// 省略
switch (request.type) {
case OpCode.sync:
// 将同步请求添加到队列中，并向leader发起同步请求
zks.pendingSyncs.add(request);
zks.getFollower().request(request);
break;
}
commit
// FollowerZookeeperServer
public void commit(long zxid) {
if (pendingTxns.size() == 0) {
LOG.warn("Committing " + Long.toHexString(zxid)
+ " without seeing txn");
return;
}
// 1. 匹配当前提交的请求和当前未处理的最老的请求
long firstElementZxid = pendingTxns.element().zxid;
if (firstElementZxid != zxid) {
LOG.error("Committing zxid 0x" + Long.toHexString(zxid)
+ " but next pending txn 0x"
+ Long.toHexString(firstElementZxid));
System.exit(12);
}
// 2. 将请求交给CommitProcessor
Request request = pendingTxns.remove();
commitProcessor.commit(request);
}
版权声明：「DDKK.COM 弟弟快看，程序员编程资料站」本站文章，版权归原作者所有
正文到此结束
热门推荐
相关文章
- 暂无相关文章