E-MapReduce:Phoenix

Step 2: Write and package the Java application

• Method 1: Use a precompiled JAR file (for quick verification)

  hbase-phoenix-kerberos-1.0-SNAPSHOT.jar

• Method 2: Manually compile and package the application (recommended for production)

  You can refer to the following code sample to manually compile and package the application.

  Core code sample (PhoenixKerberosDemo.java)

  import java.sql.Connection;
  import java.sql.DriverManager;
  import java.sql.PreparedStatement;
  import java.sql.ResultSet;
  import java.sql.SQLException;
  import java.sql.Statement;
  
  /**
   * A client that connects to a Kerberos-secured Phoenix cluster using JDBC.
   * All connection parameters are provided through a complete JDBC URL passed from the command line.
   */
  public class PhoenixKerberosDemo {
  
      /**
       * The main entry point for the application.
       *
       * @param args The command-line arguments. The program expects one argument: the full Phoenix JDBC URL.
       */
      public static void main(String[] args) {
          // --- 1. Validate command-line input: Expect one argument, the JDBC URL ---
          if (args.length != 1) {
              System.err.println("ERROR: Invalid number of arguments.");
              System.err.println("Usage: java PhoenixKerberosDemo \"<full_jdbc_url>\"");
              System.err.println("Example: \"jdbc:phoenix:zk1,zk2:2181:/hbase:user@REALM.COM:/path/to/user.keytab\"");
              System.exit(1); // Exit with an error code
          }
  
          String jdbcUrl = args[0];
  
          System.out.println("Attempting to connect to Phoenix...");
          System.out.println("Using JDBC URL: " + jdbcUrl);
  
          try {
              // --- 2. Load the Phoenix driver ---
              Class.forName("org.apache.phoenix.jdbc.PhoenixDriver");
          } catch (ClassNotFoundException e) {
              System.err.println("FATAL ERROR: Phoenix JDBC driver not found in the classpath.");
              e.printStackTrace();
              System.exit(1);
          }
  
          // --- 3. Use a try-with-resources statement to establish a connection and execute SQL. This syntax automatically closes resources. ---
          try (Connection con = DriverManager.getConnection(jdbcUrl);
               Statement stmt = con.createStatement()) {
  
              System.out.println("Connection established successfully.");
  
              final String tableName = "TEST";
              System.out.println("Creating table '" + tableName + "'...");
              stmt.executeUpdate("CREATE TABLE IF NOT EXISTS " + tableName + " (mykey INTEGER NOT NULL PRIMARY KEY, mycolumn VARCHAR)");
              con.commit();
  
              System.out.println("Upserting data...");
              stmt.executeUpdate("UPSERT INTO " + tableName + " VALUES (1, 'Hello')");
              stmt.executeUpdate("UPSERT INTO " + tableName + " VALUES (2, 'World-Kerberos!')");
              con.commit();
              System.out.println("Data upserted successfully.");
  
              String sql = "SELECT * FROM " + tableName;
              System.out.println("Querying for results with: " + sql);
  
              try (PreparedStatement statement = con.prepareStatement(sql);
                   ResultSet rset = statement.executeQuery()) {
  
                  System.out.println("Query results:");
                  while (rset.next()) {
                      System.out.println(rset.getInt("mykey") + " -> " + rset.getString("mycolumn"));
                  }
              }
  
              System.out.println("Cleaning up the test table...");
              stmt.executeUpdate("DROP TABLE IF EXISTS " + tableName);
              con.commit();
  
          } catch (SQLException e) {
              // Catch SQL exceptions and provide helpful troubleshooting tips
              System.err.println("\n--- FAILED TO EXECUTE DATABASE OPERATION ---");
              System.err.println("Please check the following:");
              System.err.println("1. The JDBC URL is correct (format, principal, keytab path).");
              System.err.println("2. Network connectivity to ZooKeeper and HBase.");
              System.err.println("3. The keytab file exists and has correct read permissions.");
              System.err.println("4. The principal has sufficient permissions on HBase tables and namespaces.");
              e.printStackTrace();
          }
  
          System.out.println("\nExecution finished.");
      }
  }
  

  Maven configuration (pom.xml)

  <?xml version="1.0" encoding="UTF-8"?>
  <project xmlns="http://maven.apache.org/POM/4.0.0"
           xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
           xsi:schemaLocation="http://maven.apache.org/POM/4.0.0
                               http://maven.apache.org/maven-v4_0_0.xsd">
      <modelVersion>4.0.0</modelVersion>
      <groupId>com.aliyun.emr.doctor</groupId>
      <artifactId>hbase-phoenix-kerberos</artifactId>
      <version>1.0-SNAPSHOT</version>
      <name>Archetype - hbase-phoenix-kerberos</name>
      <url>http://maven.apache.org</url>
      <properties>
          <phoenix.version>5.2.1</phoenix.version>
      </properties>
      <dependencies>
          <dependency>
              <groupId>org.apache.phoenix</groupId>
              <artifactId>phoenix-core</artifactId>
              <version>${phoenix.version}</version>
              <scope>provided</scope>
          </dependency>
      </dependencies>
      <build>
          <plugins>
              <!--  Java Compiler  -->
              <plugin>
                  <groupId>org.apache.maven.plugins</groupId>
                  <artifactId>maven-compiler-plugin</artifactId>
                  <version>3.1</version>
                  <configuration>
                      <source>1.8</source>
                      <target>1.8</target>
                  </configuration>
              </plugin>
              <plugin>
                  <groupId>org.apache.maven.plugins</groupId>
                  <artifactId>maven-shade-plugin</artifactId>
                  <version>3.2.4</version>
                  <executions>
                      <execution>
                          <phase>package</phase>
                          <goals>
                              <goal>shade</goal>
                          </goals>
                          <configuration>
                              <filters>
                                  <filter>
                                      <artifact>*:*</artifact>
                                      <excludes>
                                          <exclude>META-INF/*.SF</exclude>
                                          <exclude>META-INF/*.DSA</exclude>
                                          <exclude>META-INF/*.RSA</exclude>
                                      </excludes>
                                  </filter>
                              </filters>
                          </configuration>
                      </execution>
                  </executions>
              </plugin>
          </plugins>
      </build>
  </project>

Step 3: Run the application

1. Select a runtime environment.

   When you run the application, you can choose one of the following two runtime environments:

   • Run on a cluster node (Recommended)

     • Cluster nodes have all the necessary Hadoop, HBase, and Phoenix dependency libraries pre-installed. No extra configuration is required. The network connectivity is pre-configured, which makes the environment complete and stable.

     • Scenarios: This method is suitable for quick verification, testing, development, and debugging.

   • Run outside the cluster

     To run the program outside the cluster, ensure that the following conditions are met:

     • Network connectivity: Ensure that the machine where the program runs can connect to the ZooKeeper, HBase Master, and RegionServer nodes of the cluster.

     • Kerberos configuration: Copy the cluster's krb5.conf file and the generated keytab file to the machine where the program runs.

     • Dependency management: The classpath for the run command must include all required Hadoop, HBase, and Phoenix client dependency JAR packages. This process is often more complex than running on a cluster node. You can use a tool such as Maven or Gradle to manage dependencies.

2. Run the script.

   The following kerberos-phoenix.sh script integrates all necessary configurations for easy modification and execution.

   #!/bin/bash
   
   # ======================= 1. User Configuration (Modify based on your environment) =======================
   
   # Directory where Hadoop and HBase configuration files are located
   HADOOP_CONF_DIR="/etc/taihao-apps/hadoop-conf"
   HBASE_CONF_DIR="/etc/taihao-apps/hbase-conf"
   
   # Path to the Phoenix client JAR file. Using a symbolic link is a best practice to handle version changes.
   # First, use 'ls -l /opt/apps/PHOENIX/phoenix-current/' to confirm this file exists. This path may need to be modified for different versions.
   PHOENIX_JAR="/opt/apps/PHOENIX/phoenix-current/phoenix-client-lite-hbase-2.6.jar"
   
   # Your application's JAR file name.
   YOUR_JAR_FILE="hbase-phoenix-kerberos-1.0-SNAPSHOT.jar"
   
   # Path to the Kerberos configuration file.
   KRB5_CONF_PATH="/etc/krb5.conf"
   
   # --- [Core] JDBC URL Configuration ---
   # Format: jdbc:phoenix:[ZK Address]:[ZK Port]:[HBase ZNode]:[Principal]:[Absolute Keytab Path]
   # Replace the ZK address, REALM, and Keytab path below with your actual information.
   ZK_QUORUM="master-1-1" # If there are multiple ZooKeeper nodes, separate them with commas, such as "zk1,zk2,zk3"
   ZK_PORT="2181"
   HBASE_ZNODE="/hbase" # For a secure cluster, it might be /hbase-secure
   PRINCIPAL="phoenix_client@EMR.C-4FC5FDDE3759****.COM" # Replace with your Principal
   KEYTAB_PATH="/tmp/phoenix_client.keytab" # Absolute path to the Keytab file
   
   JDBC_URL="jdbc:phoenix:${ZK_QUORUM}:${ZK_PORT}:${HBASE_ZNODE}:${PRINCIPAL}:${KEYTAB_PATH}"
   # =================================================================================
   
   # ======================= 2. Execution Area (Usually no modification needed) =================================
   echo "================================================="
   echo "Starting Phoenix Kerberos JDBC Demo..."
   echo "Using JDBC URL: ${JDBC_URL}"
   echo "================================================="
   
   # Build the Classpath. Order: current directory -> config directory -> your JAR -> dependency JARs
   # `hbase classpath` automatically loads core Hadoop/HBase dependencies
   CLASS_PATH=".:${HADOOP_CONF_DIR}:${HBASE_CONF_DIR}:${YOUR_JAR_FILE}:${PHOENIX_JAR}:$(hbase classpath)"
   
   # Execute the Java program
   java -cp "${CLASS_PATH}" \
        -Djava.security.krb5.conf="${KRB5_CONF_PATH}" \
        PhoenixKerberosDemo "${JDBC_URL}"
   
   # Check the exit code
   if [ $? -eq 0 ]; then
       echo -e "\n[SUCCESS] Program finished successfully."
   else
       echo -e "\n[FAILED] Program terminated with an error."
   fi
   # =================================================================================

   1. Upload the JAR package created in Step 2 and the kerberos-phoenix.sh script to a specified directory on the master node.

   2. Run the following command to grant execute permissions to the script.

      chmod +x kerberos-phoenix.sh

   3. Run the following command to run the script.

      ./kerberos-phoenix.sh

      The following example shows a part of the returned information.