一、开发环境配置
第一:打开Zookeeper服务端,参考之前的博客。
第二:配置开发环境环境可以有两种方式:
① 直接下载相关的依赖Jar包,然后在IDE中添加依赖。
② 建立maven项目,使用maven进行依赖管理。
现在开发步骤(① 手动添加依赖至IDE)如下:
步骤一:点击http://hadoop.apache.org/zookeeper/下载对应版本的Jar包,包括(jar、javadoc.jar、sources.jar),笔者对应下载的Zookeeper3.4.9版本。
步骤二:打开IDE(笔者使用eclispe),新建名为zookeeperApp1的java项目,并添加相应的jar包。
步骤三:新建包、Java类进行测试。
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooKeeper;
public class ZookeeperApp1Test {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static void main(String[] args) throws Exception {
ZooKeeper zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher(){
@Override
public void process(WatchedEvent event) {
System.out.println("Receive watched event : " + event);
if (KeeperState.SyncConnected == event.getState()) {
connectedSemaphore.countDown();
}
}}
);
System.out.println(zookeeper.getState());
try {
connectedSemaphore.await();
} catch (InterruptedException e) {
e.printStackTrace();
}
System.out.println("Zookeeper session established");
}
}
运行结果如下:
现在开发步骤(② 使用maven管理依赖)如下:
步骤一:新建maven项目.
步骤二:配置pom.xml文件如下:
4.0.0
com.npf
zookeeperApp2
0.0.1-SNAPSHOT
jar
zookeeperApp2
http://maven.apache.org
UTF-8
junit
junit
3.8.1
test
org.apache.zookeeper
zookeeper
3.4.9
log4j
log4j
1.2.16
org.slf4j
slf4j-api
1.6.1
jdk-1.7
true
1.7
1.7
1.7
1.7
步骤三:新建java类进行测试:
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooKeeper;
public class ZookeeperApp2Test {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static void main(String[] args) throws Exception {
ZooKeeper zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher(){
@Override
public void process(WatchedEvent event) {
System.out.println("Receive watched event : " + event);
if (KeeperState.SyncConnected == event.getState()) {
connectedSemaphore.countDown();
}
}}
);
System.out.println(zookeeper.getState());
try {
connectedSemaphore.await();
} catch (InterruptedException e) {
e.printStackTrace();
}
System.out.println("Zookeeper session established");
}
}
运行结果如下:
二、操作示例
3.1 创建节点
创建节点有异步和同步两种方式。无论是异步或者同步,Zookeeper都不支持递归调用,即无法在父节点不存在的情况下创建一个子节点,如在/zk-ephemeral节点不存在的情况下创建/zk-ephemeral/ch1节点;并且如果一个节点已经存在,那么创建同名节点时,会抛出NodeExistsException异常。
① 同步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class ZookeeperCreateNodeAPITest {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static void main(String[] args) throws Exception {
ZooKeeper zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher(){
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
connectedSemaphore.countDown();
}
}}
);
System.out.println(zookeeper.getState());
connectedSemaphore.await();
String path1 = zookeeper.create("/zk-test-ephemeral-", "datatest".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL);
System.out.println("Success create znode: " + path1);
String path2 = zookeeper.create("/zk-test-ephemeral-", "datatest".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL_SEQUENTIAL);
System.out.println("Success create znode: " + path2);
}
}
运行结果如下:
结果表明已经成功创建了临时节点和临时顺序节点,在创建顺序节点时,系统会在后面自动增加一串数字。
② 异步方式
使用异步方式与同步方式的区别在于节点的创建过程(包括网络通信和服务端的节点创建过程)是异步的,在同步接口调用过程中,开发者需要关注接口抛出异常的可能,但是在异步接口中,接口本身不会抛出异常,所有异常都会在回调函数中通过Result Code来体现。
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.AsyncCallback;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class ZookeeperCreateNodeAPITest2 {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static void main(String[] args) throws Exception {
ZooKeeper zookeeper = new ZooKeeper("npfdev1:2181", 5000,
new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
connectedSemaphore.countDown();
}
}
});
System.out.println(zookeeper.getState());
connectedSemaphore.await();
zookeeper.create("/zk-test-ephemeral-", "datatest".getBytes(),
Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL,
new IStringCallback(), "I am context. ");
zookeeper.create("/zk-test-ephemeral-", "datatest".getBytes(),
Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL,
new IStringCallback(), "I am context. ");
zookeeper.create("/zk-test-ephemeral-", "datatest".getBytes(),
Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL_SEQUENTIAL,
new IStringCallback(), "I am context. ");
Thread.sleep(Integer.MAX_VALUE);
}
}
class IStringCallback implements AsyncCallback.StringCallback {
public void processResult(int rc, String path, Object ctx, String name) {
System.out.println("Create path result: [" + rc + ", " + path + ", "
+ ctx + ", real path name: " + name);
}
}
运行结果如下:
结果表明已经成功使用异步方式创建了相应节点。
3.2 删除节点
只允许删除叶子节点,即一个节点如果有子节点,那么该节点将无法直接删除,必须先删掉其所有子节点。同样也有同步和异步两种方式。
① 同步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class ZookeeperDeleteNodeAPITest {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static void main(String[] args) throws Exception {
String path = "/zk-book";
ZooKeeper zookeeper = new ZooKeeper("npfdev1:2181", 5000,
new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
}
}
}
});
connectedSemaphore.await();
zookeeper.create(path, "testdata".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
System.out.println("success create znode: " + path);
zookeeper.create(path + "/c1", "testdata".getBytes(),
Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT);
System.out.println("success create znode: " + path + "/c1");
try {
zookeeper.delete(path, -1);
} catch (Exception e) {
System.out.println("fail to delete znode: " + path);
}
zookeeper.delete(path + "/c1", -1);
System.out.println("success delete znode: " + path + "/c1");
zookeeper.delete(path, -1);
System.out.println("success delete znode: " + path);
Thread.sleep(Integer.MAX_VALUE);
}
}
运行结果如下:
结果表明若节点有子节点,则无法将其删除,必须先删除其所有子节点。
② 异步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.AsyncCallback;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class ZookeeperDeleteNodeAPITest2 {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static void main(String[] args) throws Exception {
String path = "/zk-book";
ZooKeeper zookeeper = new ZooKeeper("npfdev1:2181", 5000,
new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
}
}
}
});
connectedSemaphore.await();
zookeeper.create(path, "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
System.out.println("success create znode: " + path);
zookeeper.create(path + "/c1", "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
System.out.println("success create znode: " + path + "/c1");
zookeeper.delete(path, -1, new IVoidCallback(), null);
zookeeper.delete(path + "/c1", -1, new IVoidCallback(), null);
zookeeper.delete(path, -1, new IVoidCallback(), null);
Thread.sleep(Integer.MAX_VALUE);
}
}
class IVoidCallback implements AsyncCallback.VoidCallback {
@Override
public void processResult(int rc, String path, Object ctx) {
System.out.println(rc + ", " + path + ", " + ctx);
}
}
运行结果如下:
结果结果表明第一次删除/zk-book的时异常,ResultCode为-111。
3.3 子节点获取
读取节点的子节点列表,同样可以使用同步和异步的方式进行操作。
① 同步方式
package com.npf.test;
import java.util.List;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class ZookeeperGetNodeAPITest {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static ZooKeeper zookeeper = null;
public static void main(String[] args) throws Exception {
String path = "/zk-book-1";
zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
} else if (event.getType() == EventType.NodeChildrenChanged) {
try {
System.out.println("ReGet Child:"
+ zookeeper.getChildren(event.getPath(),
true));
} catch (Exception e) {
}
}
}
}
});
connectedSemaphore.await();
zookeeper.create(path, "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
System.out.println("success create znode: " + path);
zookeeper.create(path + "/c1", "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path + "/c1");
List childrenList = zookeeper.getChildren(path, true);
System.out.println(childrenList);
zookeeper.create(path + "/c2", "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path + "/c2");
Thread.sleep(1000);
zookeeper.create(path + "/c3", "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path + "/c3");
Thread.sleep(Integer.MAX_VALUE);
}
}
运行结果如下:
值得注意的是,Watcher通知是一次性的,即一旦触发一次通知后,该Watcher就失效了,因此客户端需要反复注册Watcher,即程序中在process里面又注册了Watcher,否则,将无法获取c3节点的创建而导致子节点变化的事件。
② 异步方式
package com.npf.test;
import java.util.List;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.AsyncCallback;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
import org.apache.zookeeper.data.Stat;
public class ZookeeperGetNodeAPITest2 {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static ZooKeeper zookeeper = null;
public static void main(String[] args) throws Exception {
String path = "/zk-book-2";
zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
} else if (event.getType() == EventType.NodeChildrenChanged) {
try {
System.out.println("ReGet Child:"
+ zookeeper.getChildren(event.getPath(),
true));
} catch (Exception e) {
}
}
}
}
});
connectedSemaphore.await();
zookeeper.create(path, "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
System.out.println("success create znode: " + path);
zookeeper.create(path + "/c1", "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path + "/c1");
zookeeper.getChildren(path, true, new IChildren2Callback(), null);
zookeeper.create(path + "/c2", "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path + "/c2");
Thread.sleep(Integer.MAX_VALUE);
}
}
class IChildren2Callback implements AsyncCallback.Children2Callback {
@Override
public void processResult(int rc, String path, Object ctx,
List children, Stat stat) {
System.out.println("Get Children znode result: [response code: " + rc
+ ", param path: " + path + ", ctx: " + ctx
+ ", children list: " + children + ", stat: " + stat);
}
}
运行结果如下:
结果表示通过异步的方式可以获取子节点信息。
3.4 数据节点获取
对于节点的数据获取,同样存在同步和异步两种方式。
① 同步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
import org.apache.zookeeper.data.Stat;
public class ZookeeperGetDataAPITest {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static ZooKeeper zookeeper = null;
private static Stat stat = new Stat();
public static void main(String[] args) throws Exception {
String path = "/zk-book-4";
zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
} else if (event.getType() == EventType.NodeDataChanged) {
try {
System.out.println("the data of znode "
+ event.getPath()
+ " is : "
+ new String(zookeeper.getData(
event.getPath(), true, stat)));
System.out.println("czxID: " + stat.getCzxid()
+ ", mzxID: " + stat.getMzxid()
+ ", version: " + stat.getVersion());
} catch (Exception e) {
}
}
}
}
});
connectedSemaphore.await();
zookeeper.create(path, "123".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path);
System.out.println("the data of znode " + path + " is : "
+ new String(zookeeper.getData(path, true, stat)));
System.out.println("czxID: " + stat.getCzxid() + ", mzxID: "
+ stat.getMzxid() + ", version: " + stat.getVersion());
zookeeper.setData(path, "1234".getBytes(), -1);
Thread.sleep(Integer.MAX_VALUE);
}
}
运行结果如下:
结果表明可以使用getData函数获取节点的数据。
② 异步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.AsyncCallback;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
import org.apache.zookeeper.data.Stat;
public class ZookeeperGetDataAPITest2 {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static ZooKeeper zookeeper = null;
public static void main(String[] args) throws Exception {
String path = "/zk-book-5";
zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
} else if (event.getType() == EventType.NodeDataChanged) {
try {
zookeeper.getData(event.getPath(), true,
new IDataCallback(), null);
} catch (Exception e) {
}
}
}
}
});
connectedSemaphore.await();
zookeeper.create(path, "123".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path);
zookeeper.getData(path, true, new IDataCallback(), null);
zookeeper.setData(path, "1234".getBytes(), -1);
Thread.sleep(Integer.MAX_VALUE);
}
}
class IDataCallback implements AsyncCallback.DataCallback {
public void processResult(int rc, String path, Object ctx, byte[] data,
Stat stat) {
System.out.println("rc: " + rc + ", path: " + path + ", data: "
+ new String(data));
System.out.println("czxID: " + stat.getCzxid() + ", mzxID: "
+ stat.getMzxid() + ", version: " + stat.getVersion());
}
}
结果表明采用异步方式同样可方便获取节点的数据。
3.5 更新数据
在更新数据时,setData方法存在一个version参数,其用于指定节点的数据版本,表明本次更新操作是针对指定的数据版本进行的,但是,在getData方法中,并没有提供根据指定数据版本来获取数据的接口,那么,这里为何要指定数据更新版本呢,这里方便理解,可以等效于CAS(compare and swap),对于值V,每次更新之前都会比较其值是否是预期值A,只有符合预期,才会将V原子化地更新到新值B。Zookeeper的setData接口中的version参数可以对应预期值,表明是针对哪个数据版本进行更新,假如一个客户端试图进行更新操作,它会携带上次获取到的version值进行更新,而如果这段时间内,Zookeeper服务器上该节点的数据已经被其他客户端更新,那么其数据版本也会相应更新,而客户端携带的version将无法匹配,无法更新成功,因此可以有效地避免分布式更新的并发问题。
① 同步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
import org.apache.zookeeper.data.Stat;
public class ZookeeperSetDataAPITest {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static ZooKeeper zookeeper = null;
public static void main(String[] args) throws Exception {
String path = "/zk-book-6";
zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
}
}
}
});
connectedSemaphore.await();
zookeeper.create(path, "123".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path);
zookeeper.getData(path, true, null);
Stat stat = zookeeper.setData(path, "456".getBytes(), -1);
System.out.println("czxID: " + stat.getCzxid() + ", mzxID: "
+ stat.getMzxid() + ", version: " + stat.getVersion());
Stat stat2 = zookeeper.setData(path, "456".getBytes(),
stat.getVersion());
System.out.println("czxID: " + stat2.getCzxid() + ", mzxID: "
+ stat2.getMzxid() + ", version: " + stat2.getVersion());
try {
zookeeper.setData(path, "456".getBytes(), stat.getVersion());
} catch (KeeperException e) {
System.out.println("Error: " + e.code() + "," + e.getMessage());
}
Thread.sleep(Integer.MAX_VALUE);
}
}
结果表明由于携带的数据版本不正确,而无法成功更新节点。其中,setData中的version参数设置-1含义为客户端需要基于数据的最新版本进行更新操作。
② 异步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.AsyncCallback;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
import org.apache.zookeeper.data.Stat;
public class ZookeeperSetDataAPITest2 {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
public static ZooKeeper zookeeper = null;
public static void main(String[] args) throws Exception {
String path = "/zk-book-7";
zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher() {
@Override
public void process(WatchedEvent event) {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
}
}
}
});
connectedSemaphore.await();
zookeeper.create(path, "123".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL);
System.out.println("success create znode: " + path);
zookeeper
.setData(path, "456".getBytes(), -1, new IStatCallback(), null);
Thread.sleep(Integer.MAX_VALUE);
}
}
class IStatCallback implements AsyncCallback.StatCallback {
public void processResult(int rc, String path, Object ctx, Stat stat) {
System.out.println("rc: " + rc + ", path: " + path + ", stat: " + stat);
}
}
rc(ResultCode)为0,表明成功更新节点数据。
3.6 检测节点是否存在
在调用接口时注册Watcher的话,还可以对节点是否存在进行监听,一旦节点被创建、被删除、数据更新,都会通知客户端。
① 同步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class ZookeeperNodeExistAPITest {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
static ZooKeeper zookeeper = null;
public static void main(String[] args) throws Exception {
String path = "/zk-book";
zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher() {
@Override
public void process(WatchedEvent event) {
try {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
} else if (EventType.NodeCreated == event.getType()) {
System.out.println("success create znode: "
+ event.getPath());
zookeeper.exists(event.getPath(), true);
} else if (EventType.NodeDeleted == event.getType()) {
System.out.println("success delete znode: "
+ event.getPath());
zookeeper.exists(event.getPath(), true);
} else if (EventType.NodeDataChanged == event.getType()) {
System.out.println("data changed of znode: "
+ event.getPath());
zookeeper.exists(event.getPath(), true);
}
}
} catch (Exception e) {
}
}
});
connectedSemaphore.await();
zookeeper.exists(path, true);
zookeeper.create(path, "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
zookeeper.setData(path, "123".getBytes(), -1);
zookeeper.create(path + "/c1", "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
System.out.println("success create znode: " + path + "/c1");
zookeeper.delete(path + "/c1", -1);
zookeeper.delete(path, -1);
Thread.sleep(Integer.MAX_VALUE);
}
}
运行结果如下:
结果表明:
· 无论节点是否存在,都可以通过exists接口注册Watcher。
· 注册的Watcher,对节点创建、删除、数据更新事件进行监听。
· 对于指定节点的子节点的各种变化,不会通知客户端。
② 异步方式
package com.npf.test;
import java.util.concurrent.CountDownLatch;
import org.apache.zookeeper.AsyncCallback;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.Watcher.Event.EventType;
import org.apache.zookeeper.Watcher.Event.KeeperState;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
import org.apache.zookeeper.data.Stat;
public class ZookeeperNodeExistAPITest2 {
private static CountDownLatch connectedSemaphore = new CountDownLatch(1);
static ZooKeeper zookeeper = null;
public static void main(String[] args) throws Exception {
String path = "/zk-book";
zookeeper = new ZooKeeper("npfdev1:2181", 5000, new Watcher() {
@Override
public void process(WatchedEvent event) {
try {
if (KeeperState.SyncConnected == event.getState()) {
if (EventType.None == event.getType()
&& null == event.getPath()) {
connectedSemaphore.countDown();
} else if (EventType.NodeCreated == event.getType()) {
System.out.println("success create znode: "
+ event.getPath());
zookeeper.exists(event.getPath(), true,
new IIStatCallback(), null);
} else if (EventType.NodeDeleted == event.getType()) {
System.out.println("success delete znode: "
+ event.getPath());
zookeeper.exists(event.getPath(), true,
new IIStatCallback(), null);
} else if (EventType.NodeDataChanged == event.getType()) {
System.out.println("data changed of znode: "
+ event.getPath());
zookeeper.exists(event.getPath(), true,
new IIStatCallback(), null);
}
}
} catch (Exception e) {
}
}
});
connectedSemaphore.await();
zookeeper.exists(path, true, new IIStatCallback(), null);
zookeeper.create(path, "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
zookeeper.setData(path, "123".getBytes(), -1);
zookeeper.create(path + "/c1", "".getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.PERSISTENT);
System.out.println("success create znode: " + path + "/c1");
zookeeper.delete(path + "/c1", -1);
zookeeper.delete(path, -1);
Thread.sleep(Integer.MAX_VALUE);
}
}
class IIStatCallback implements AsyncCallback.StatCallback {
@Override
public void processResult(int rc, String path, Object ctx, Stat stat) {
System.out.println("rc: " + rc + ", path: " + path + ", stat: " + stat);
}
}
运行结果如下:
结果表明当节点不存在时,其rc(ResultCode)为-101。
3.7 权限控制
通过设置Zookeeper服务器上数据节点的ACL控制,就可以对其客户端对该数据节点的访问权限:如果符合ACL控制,则可以进行访问,否则无法访问。
① 使用无权限信息的Zookeeper会话访问含权限信息的数据节点
package com.npf.test;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class AuthSample_Get {
final static String PATH = "/zk-book-auth_test";
public static void main(String[] args) throws Exception {
ZooKeeper zookeeper1 = new ZooKeeper("npfdev1:2181", 5000, null);
zookeeper1.addAuthInfo("digest", "foo:true".getBytes());
zookeeper1.create(PATH, "init".getBytes(), Ids.CREATOR_ALL_ACL, CreateMode.EPHEMERAL);
System.out.println("success create znode: " + PATH);
ZooKeeper zookeeper2 = new ZooKeeper("npfdev1:2181", 5000, null);
zookeeper2.getData(PATH, false, null);
}
}
运行结果如下:
表示权限不够,不能进行操作。
② 删除带权限控制的节点
package com.npf.test;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class AuthSample_Delete {
final static String PATH = "/zk-book-auth_test";
final static String PATH2 = "/zk-book-auth_test/child";
public static void main(String[] args) throws Exception {
ZooKeeper zookeeper1 = new ZooKeeper("npfdev1:2181", 5000, null);
zookeeper1.addAuthInfo("digest", "foo:true".getBytes());
zookeeper1.create(PATH, "init".getBytes(), Ids.CREATOR_ALL_ACL,
CreateMode.PERSISTENT);
zookeeper1.create(PATH2, "init".getBytes(), Ids.CREATOR_ALL_ACL,
CreateMode.EPHEMERAL);
try {
ZooKeeper zookeeper2 = new ZooKeeper("npfdev1:2181", 5000, null);
zookeeper2.delete(PATH2, -1);
} catch (Exception e) {
System.out.println("fail to delete: " + e.getMessage());
}
ZooKeeper zookeeper3 = new ZooKeeper("npfdev1:2181", 5000, null);
zookeeper3.addAuthInfo("digest", "foo:true".getBytes());
zookeeper3.delete(PATH2, -1);
System.out.println("success delete znode: " + PATH2);
ZooKeeper zookeeper4 = new ZooKeeper("npfdev1:2181", 5000, null);
zookeeper4.delete(PATH, -1);
System.out.println("success delete znode: " + PATH);
}
}
运行结果如下:
结果表明若没有权限,则无法删除节点。