Phoenix 之kerberos 环境下 JDBC 操作Demo
一、操作环境
ambari 2.4.3 + hdp 2.5.3 + hbase 1.1.2 + phoenix4.7 + kerberos + centos6.9
二、代码如下：
package ycb.service;
import java.io.IOException;
import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.PreparedStatement;
import java.sql.ResultSet;
import java.sql.ResultSetMetaData;
import java.sql.SQLException;
import java.sql.Statement;
import java.util.ArrayList;
import org.apache.hadoop.security.UserGroupInformation;
import org.apache.log4j.Logger;
import ycb.utils.FileUtils;
/**
* @author YeChunBo
* @time 2018年3月28日
*
* 类说明 :测试phoenix JDBC简单样例（已开启kerberos安全认证机制）
*/
public class PhoenixDemo {
private static Logger log = Logger.getLogger(PhoenixDemo.class);
/**
* 获取连接
*/
public Connection getConn() {
// 连接hadoop环境，进行 Kerberos认证
org.apache.hadoop.conf.Configuration conf = new org.apache.hadoop.conf.Configuration();
conf.set("hadoop.security.authentication", "Kerberos");
// linux 环境会默认读取/etc/krb.cnf文件，win不指定会默认读取C:/Windows/krb5.ini
if (System.getProperty("os.name").toLowerCase().startsWith("win")) {
System.setProperty("java.security.krb5.conf", "C:/Windows/krbconf/pro/krb5.ini");
}
UserGroupInformation.setConfiguration(conf);
Connection conn = null;
try {
UserGroupInformation.loginUserFromKeytab("test/[email protected]", "./conf/test.keytab");
Class.forName("org.apache.phoenix.jdbc.PhoenixDriver");
String url = "jdbc:phoenix:bigdata04,bigdata05,bigdata06:2181"; // 这里配置zookeeper server的地址,可单个也可多个
conn = DriverManager.getConnection(url);
} catch (ClassNotFoundException e) {
log.error(e.getMessage());
e.printStackTrace();
} catch (SQLException e1) {
log.error(e1.getMessage());
e1.printStackTrace();
} catch (IOException e2) {
log.error(e2.getMessage());
e2.printStackTrace();
}
return conn;
}
/**
* 通过phoenix 创建表、插入数据、索引、查询数据
*
* @param phoenix demo
*/
public void phoenixDemo() {
Connection conn = getConn();
ResultSet rs = null;
Statement stmt = null;
try {
stmt = conn.createStatement();
stmt.execute("drop table if exists test");
stmt.execute("create table test (mykey integer not null primary key, mycolumn varchar)");
stmt.execute("create index test_idx on test(mycolumn)");
for (int i = 0; i < 100; i++) {
stmt.executeUpdate("upsert into test values ("+ i +",'The num is "+ i +"')");
}
conn.commit();
PreparedStatement statement = conn.prepareStatement("select mykey from test where mycolumn='The num is 88'");
rs = statement.executeQuery();
while (rs.next()) {
log.info("-------------The num is ---------------" + rs.getInt(1));
}
} catch (SQLException e) {
log.error(e.getMessage());
e.printStackTrace();
} finally {
closeRes(conn,stmt,rs);
}
}
/**
* 通过Phoenix 将hbase数据导入到CSV
*
* @param querySql
* @param fileName
*/
public void hbaseData2CSV(String querySql, String fileName) {
Connection conn = null;
ResultSet rs = null;
Statement statement = null;
try {
conn = getConn();
statement = conn.createStatement();
int count = 0;
ArrayList dataList = new ArrayList();
StringBuffer dataLineBuf = new StringBuffer();
rs = statement.executeQuery(querySql);
ResultSetMetaData m = rs.getMetaData();
int columns = m.getColumnCount();
while (rs.next()) {
dataLineBuf.setLength(0);
count = count + 1;
for (int i = 1; i <= columns; i++) {
dataLineBuf.append(rs.getString(i)).append(",");
}
dataList.add(dataLineBuf.toString());
}
// 将查询出来的数据输出到CSV文件中
FileUtils.arrayList2CSVFile(dataList, fileName);
} catch (Exception e) {
log.error(e.getMessage());
e.printStackTrace();
} finally {
closeRes(conn,statement,rs);
}
}
/**
* 关闭资源
* @param conn
* @param statement
* @param rs
*/
public void closeRes(Connection conn, Statement statement, ResultSet rs) {
try {
if (conn != null) {
conn.close();
}
if (statement != null)
statement.close();
if (rs != null)
rs.close();
} catch (Exception e) {
e.printStackTrace();
}
}
public static void main(String[] args) {
PhoenixDemo phoenixDemo = new PhoenixDemo();
phoenixDemo.phoenixDemo();
System.out.println("----------------------------------------------");
phoenixDemo.hbaseData2CSV("select * from test limit 100", "/tmp/phoenixTest/test.csv");
}
}
运行结果如下所示：
[INFO ] 2018-03-29 18:07:24,360 method:ycb.service.PhoenixDemo.phoenixDemo(PhoenixDemo.java:85)
-------------The num is ---------------88
[root@bigdata07 phoenixTest]# more test.csv
0,The num is 0,
1,The num is 1,
10,The num is 10,
11,The num is 11,
12,The num is 12,三、测试多线程获取hbase中对应的数据并测试效率
测试样例代码如下：
package ycb.thread;
import ycb.service.PhoenixDemo;
/**
* @author YeChunBo
* @time 2018年3月28日
*
* 类说明:测试phoenix 进行多线程读取hbase数据并输入到csv中
*/
public class PhoenixThread implements Runnable {
public String threadName;
PhoenixDemo phoenixDemo = new PhoenixDemo();
public PhoenixThread(String name, PhoenixDemo phoenixDemo) {
threadName = name;
this.phoenixDemo = phoenixDemo;
}
public void run() {
phoenixDemo.hbaseData2CSV("SELECT * FROM \"t_hbase1\" limit 1000000",
"/tmp/phoenixTest/t_hbase1_" + threadName + ".csv");
}
public static void main(String[] args) {
new Thread(new PhoenixThread("ThreadA", new PhoenixDemo())).start();
new Thread(new PhoenixThread("ThreadB", new PhoenixDemo())).start();
new Thread(new PhoenixThread("ThreadC", new PhoenixDemo())).start();
}
}
测试结果如下所示，100万数据从hbase中读取到csv不到10秒全部写完：
-rw-r--r--. 1 root root 1896494 Mar 29 17:49 t_hbase1_ThreadA.csv
-rw-r--r--. 1 root root 1896494 Mar 29 17:49 t_hbase1_ThreadB.csv
-rw-r--r--. 1 root root 1896494 Mar 29 17:49 t_hbase1_ThreadC.csv
[root@bigdata20 phoenixTest]# more t_hbase1_ThreadA.csv
1,1,696083Ԫ,http://blog.csdn.net/s20082043/article/details/8534092,hello world avz7qgu77wog3r6c5qw8426b4ape432523974591we9t5u314356hzy1kxj7x8g39a2l9tl7734mbxn3oa2192kaq9381,
10,10,391968Ԫ,http://blog.csdn.net/s20082043/article/details/3138741,hello world avz7qgu77wog3r6c5qw8426b4ape432523974591we9t5u314356hzy1kxj7x8g39a2l9tl7734mbxn3oa2192kaq93810,
100,100,528644Ԫ,http://blog.csdn.net/s20082043/article/details/3381080,hello world avz7qgu77wog3r6c5qw8426b4ape432523974591we9t5u314356hzy1kxj7x8g39a2l9tl7734mbxn3oa2192kaq938100,
1000,1000,328084Ԫ,http://blog.csdn.net/s20082043/article/details/7711796,hello world avz7qgu77wog3r6c5qw8426b4ape432523974591we9t5u314356hzy1kxj7x8g39a2l9tl7734mbxn3oa2192kaq9381000,
10000,10000,582678Ԫ,http://blog.csdn.net/s20082043/article/details/7523690,hello world avz7qgu77wog3r6c5qw8426b4ape432523974591we9t5u314356hzy1kxj7x8g39a2l9tl7734mbxn3oa2192kaq93810000,
100000,100000,201643Ԫ,http://blog.csdn.net/s20082043/article/details/6580911,hello world avz7qgu77wog3r6c5qw8426b4ape432523974591we9t5u314356hzy1kxj7x8g39a2l9tl7734mbxn3oa2192kaq938100000,
1000000,1000000,null,null,null,
附写文件的代码如下：
package ycb.utils;
import java.io.BufferedWriter;
import java.io.File;
import java.io.FileOutputStream;
import java.io.OutputStreamWriter;
import java.util.ArrayList;
import java.util.Iterator;
import org.apache.log4j.Logger;
public class FileUtils {
private static Logger log = Logger.getLogger(FileUtils.class);
/**
* 将ArrayList集合中的数据输出到文件中
* @param userMap 数据集合
* @param fileName 文件名：全路径
* @return
*/
public static boolean arrayList2CSVFile(ArrayList dataList,String fileName ){
boolean flag = true;
File file = new File(fileName);
OutputStreamWriter out = null;
BufferedWriter bfw = null;
try {
out = new OutputStreamWriter(new FileOutputStream(file),"GBK");
bfw = new BufferedWriter(out);
Iterator it = dataList.iterator();
String strLine = null;
int count = 0;
while(it.hasNext()){
strLine = it.next();
bfw.write(strLine.toString());
bfw.newLine();
count = count + 1;
if (count % 1000 == 0) {
log.info("线程：" + Thread.currentThread().getName() + "成功插入【" + count +"】");
bfw.flush();
}
}
bfw.flush();
} catch (Exception e) {
System.out.println("将数据写到文件时出现异常： " + e.getMessage());
flag = false;
e.printStackTrace();
}
return flag;
}
}
附Eclipse中代码结构如下：
测试后发现：只需要hbase-site.xml配置文件跟hdp对应版本的phoenix client jar包便可。
五、遇到的坑
- 1、创建二级索引时报如下错
[INFO ] 2018-03-29 17:56:27,617 method:org.apache.hadoop.hbase.client.HBaseAdmin$CreateTableFuture.postOperationResult(HBaseAdmin.java:779)
Created TEST
[ERROR] 2018-03-29 17:56:27,642 method:ycb.service.PhoenixDemo.phoenixDemo(PhoenixDemo.java:88)
ERROR 1029 (42Y88): Mutable secondary indexes must have the hbase.regionserver.wal.codec property set to org.apache.hadoop.hbase.regionserver.wal.IndexedWALEditCodec in the hbase-sites.xml of every region server. tableName=TEST_IDX
java.sql.SQLException: ERROR 1029 (42Y88): Mutable secondary indexes must have the hbase.regionserver.wal.codec property set to org.apache.hadoop.hbase.regionserver.wal.IndexedWALEditCodec in the hbase-sites.xml of every region server. tableName=TEST_IDX
at org.apache.phoenix.exception.SQLExceptionCode$Factory$1.newException(SQLExceptionCode.java:441)
at org.apache.phoenix.exception.SQLExceptionInfo.buildException(SQLExceptionInfo.java:145)
at org.apache.phoenix.schema.MetaDataClient.createIndex(MetaDataClient.java:1195)
at org.apache.phoenix.compile.CreateIndexCompiler$1.execute(CreateIndexCompiler.java:85)
at org.apache.phoenix.jdbc.PhoenixStatement$2.call(PhoenixStatement.java:343)
at org.apache.phoenix.jdbc.PhoenixStatement$2.call(PhoenixStatement.java:331)
at org.apache.phoenix.call.CallRunner.run(CallRunner.java:53)
at org.apache.phoenix.jdbc.PhoenixStatement.executeMutation(PhoenixStatement.java:329)
at org.apache.phoenix.jdbc.PhoenixStatement.execute(PhoenixStatement.java:1440)
at ycb.service.PhoenixDemo.phoenixDemo(PhoenixDemo.java:77)
at ycb.service.PhoenixDemo.main(PhoenixDemo.java:162)根据报错日志配置hbase对应的属性便可
change hbase.regionserver.wal.codec from
org.apache.hadoop.hbase.regionserver.wal.WALCellCodec
to org.apache.hadoop.hbase.regionserver.wal.IndexedWALEditCodec
具体界面如下所示：
修改完上述属性需要重启Hbase，同样需要重新进入phoenix命令才会生效。
-2、没有对应的操作权限
[ERROR] 2018-03-29 18:03:30,178 method:ycb.service.PhoenixDemo.phoenixDemo(PhoenixDemo.java:88)
org.apache.hadoop.hbase.security.AccessDeniedException: Insufficient permissions for user 'keke_user/[email protected]' (action=create)
at org.apache.ranger.authorization.hbase.AuthorizationSession.publishResults(AuthorizationSession.java:261)
at org.apache.ranger.authorization.hbase.RangerAuthorizationCoprocessor.authorizeAccess(RangerAuthorizationCoprocessor.java:595)
at org.apache.ranger.authorization.hbase.RangerAuthorizationCoprocessor.requirePermission(RangerAuthorizationCoprocessor.java:664)
at org.apache.ranger.authorization.hbase.RangerAuthorizationCoprocessor.preCreateTable(RangerAuthorizationCoprocessor.java:769)
at org.apache.ranger.authorization.hbase.RangerAuthorizationCoprocessor.preCreateTable(RangerAuthorizationCoprocessor.java:496)
at org.apache.hadoop.hbase.master.MasterCoprocessorHost$11.call(MasterCoprocessorHost.java:222)
at org.apache.hadoop.hbase.master.MasterCoprocessorHost.execOperation(MasterCoprocessorHost.java:1146)
at org.apache.hadoop.hbase.master.MasterCoprocessorHost.preCreateTable(MasterCoprocessorHost.java:218)
at org.apache.hadoop.hbase.master.HMaster.createTable(HMaster.java:1603)
at org.apache.hadoop.hbase.master.MasterRpcServices.createTable(MasterRpcServices.java:462)
at org.apache.hadoop.hbase.protobuf.generated.MasterProtos$MasterService$2.callBlockingMethod(MasterProtos.java:57204)
at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:2127)
at org.apache.hadoop.hbase.ipc.CallRunner.run(CallRunner.java:107)
at org.apache.hadoop.hbase.ipc.RpcExecutor.consumerLoop(RpcExecutor.java:133)
at org.apache.hadoop.hbase.ipc.RpcExecutor$1.run(RpcExecutor.java:108)
at java.lang.Thread.run(Thread.java:745)
org.apache.phoenix.exception.PhoenixIOException: org.apache.hadoop.hbase.security.AccessDeniedException: Insufficient permissions for user 'keke_user/[email protected]' (action=create)根据日志对该kerberos用户进行授权便可，笔者是通过Ranger进行用户权限管理的。修改界面如下所示：
其中 TEST为将要创建的表，TEST_IDX 为要创建的索引名。
3、如果需要通过Eclipse进行远程访问Hbase数据，则首先需要该Hbase环境支持直接通过Eclipse进行远程访问。
Hbase支持远程访问的条件如下所示：
1、hbase 所有对应的master 和 regionserver对应的端口需要对外可访问，默认是16000,16020。
2、因为hbase client 会通过Zookeeper来访问hbase数据，所以也需要将所有zookeeper对应的访问端口开放，默认是2181。
3、同时需要配置win对应的host文件，具体路径如下：C:\Windows\System32\drivers\etc的hosts。也就是在该文件中添加集群中对应的ip及映射名。
笔者在第三步时自己坑了自己一把：
windows中hosts配置如下：
10.189.75.4 bigdata04 bigdata04.bigdata.com.cn
10.189.75.5 bigdata05 bigdata05.bigdata.com.cn
10.189.75.6 bigdata06 bigdata06.bigdata.com.cn
10.189.75.7 bigdata07 bigdata07.bigdata.com.cn
10.189.75.8 bigdata08 bigdata08.bigdata.com.cn
10.189.75.9 bigdata09 bigdata09.bigdata.com.cn
10.189.75.41 bigdata41 bigdata41.bigdata.com.cneclispe执行出现的错误如下所示：
[INFO ] 2018-06-08 16:02:53,064 method:org.apache.phoenix.shaded.org.apache.zookeeper.ZooKeeper.(ZooKeeper.java:438)
Initiating client connection, connectString=bigdata04:2181,bigdata05:2181,bigdata06:2181 sessionTimeout=180000 watcher=org.apache.hadoop.hbase.zookeeper.PendingWatcher@78ffe6dc
[INFO ] 2018-06-08 16:02:53,088 method:org.apache.phoenix.shaded.org.apache.zookeeper.ClientCnxn$SendThread.logStartConnect(ClientCnxn.java:1019)
Opening socket connection to server bigdata06/10.189.75.6:2181. Will not attempt to authenticate using SASL (unknown error)
[INFO ] 2018-06-08 16:02:53,092 method:org.apache.phoenix.shaded.org.apache.zookeeper.ClientCnxn$SendThread.primeConnection(ClientCnxn.java:864)
Socket connection established to bigdata06/10.189.75.6:2181, initiating session
[INFO ] 2018-06-08 16:02:53,102 method:org.apache.phoenix.shaded.org.apache.zookeeper.ClientCnxn$SendThread.onConnected(ClientCnxn.java:1279)
Session establishment complete on server bigdata06/10.189.75.6:2181, sessionid = 0x362f5ba9b76402c, negotiated timeout = 40000
[WARN ] 2018-06-08 16:02:53,685 method:org.apache.hadoop.hdfs.shortcircuit.DomainSocketFactory.(DomainSocketFactory.java:117)
The short-circuit local reads feature cannot be used because UNIX Domain sockets are not available on Windows.
[INFO ] 2018-06-08 16:02:53,728 method:org.apache.phoenix.metrics.Metrics.initialize(Metrics.java:44)
Initializing metrics system: phoenix
[WARN ] 2018-06-08 16:02:53,781 method:org.apache.hadoop.metrics2.impl.MetricsConfig.loadFirst(MetricsConfig.java:125)
Cannot locate configuration: tried hadoop-metrics2-phoenix.properties,hadoop-metrics2.properties
[INFO ] 2018-06-08 16:02:53,832 method:org.apache.hadoop.metrics2.impl.MetricsSystemImpl.startTimer(MetricsSystemImpl.java:376)
Scheduled snapshot period at 10 second(s).
[INFO ] 2018-06-08 16:02:53,833 method:org.apache.hadoop.metrics2.impl.MetricsSystemImpl.start(MetricsSystemImpl.java:192)
phoenix metrics system started
[WARN ] 2018-06-08 16:03:12,467 method:org.apache.hadoop.hbase.ipc.RpcClientImpl$Connection$1.run(RpcClientImpl.java:664)
Couldn't setup connection for keke_user/[email protected] to hbase/[email protected]
[WARN ] 2018-06-08 16:03:30,162 method:org.apache.hadoop.hbase.ipc.RpcClientImpl$Connection$1.run(RpcClientImpl.java:664)
Couldn't setup connection for keke_user/[email protected] to hbase/[email protected] 根据上述错误（Couldn’t setup connection for keke_user/[email protected] to hbase/[email protected]）发现实际上去连接时取的是hosts中ip后面对应的第一个映射名，但笔者在kerberos中配置机器是对应ip第二个映射名的，所以需要将hosts中的内容修改成如下所示：
10.189.75.4 bigdata04.bigdata.com.cn bigdata04
10.189.75.5 bigdata05.bigdata.com.cn bigdata05
10.189.75.6 bigdata06.bigdata.com.cn bigdata06
10.189.75.7 bigdata07.bigdata.com.cn bigdata07
10.189.75.8 bigdata08.bigdata.com.cn bigdata08
10.189.75.9 bigdata09.bigdata.com.cn bigdata09
10.189.75.41 bigdata41.bigdata.com.cn bigdata41