HBase + Solr Cloud实现HBase二级索引

1. 执行流程

 

2. Solr Cloud实现

http://blog.csdn.net/u011462328/article/details/53008344

3. HBase实现

1) 自定义Observer

①　代码

package cn.bfire.coprocessor;

import com.typesafe.config.Config;
import com.typesafe.config.ConfigFactory;
import org.apache.hadoop.hbase.Cell;
import org.apache.hadoop.hbase.CellUtil;
import org.apache.hadoop.hbase.client.Delete;
import org.apache.hadoop.hbase.client.Durability;
import org.apache.hadoop.hbase.client.Put;
import org.apache.hadoop.hbase.coprocessor.BaseRegionObserver;
import org.apache.hadoop.hbase.coprocessor.ObserverContext;
import org.apache.hadoop.hbase.coprocessor.RegionCoprocessorEnvironment;
import org.apache.hadoop.hbase.regionserver.wal.WALEdit;
import org.apache.hadoop.hbase.util.Bytes;
import org.apache.solr.common.SolrInputDocument;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import java.io.IOException;
import java.util.List;

/**
 * 为hbase提供二级索引的协处理器 Coprocesser
 */
public class UserDevPiSolrObserver extends BaseRegionObserver {

    //加载配置文件属性
    static Config config = ConfigFactory.load("userdev_pi_solr.properties");

    //log记录
    private static final Logger logger = LoggerFactory.getLogger(UserDevPiSolrObserver.class);

    @Override
    public void postPut(ObserverContext<RegionCoprocessorEnvironment> e, Put put, WALEdit edit, Durability durability) throws IOException {
        // 获取行键值
        String rowkey = Bytes.toString(put.getRow());
        //实例化 SolrDoc
        SolrInputDocument doc = new SolrInputDocument();
        //添加Solr uniqueKey值
        doc.addField("rowkey", rowkey);
        // 获取需要索引的列
        String[] hbase_columns = config.getString("hbase_column").split(",");

        // 获取需要索引的列的值并将其添加到SolrDoc
        for (int i = 0; i < hbase_columns.length; i++) {
            String colName = hbase_columns[i];
            String colValue = "";
            // 获取指定列
            List<Cell> cells = put.get("cf".getBytes(), colName.getBytes());
            if (cells != null) {
                try {
                    colValue = Bytes.toString(CellUtil.cloneValue(cells.get(0)));
                } catch (Exception ex) {
                    logger.error("添加solrdoc错误", ex);
                }
            }

            doc.addField(colName, colValue);
        }

        //发送数据到本地缓存
        SolrIndexTools.addDoc(doc);
    }

    @Override
    public void postDelete(ObserverContext<RegionCoprocessorEnvironment> e, Delete delete, WALEdit edit, Durability durability) throws IOException {
        //得到rowkey
        String rowkey = Bytes.toString(delete.getRow());
        //发送数据本地缓存
        String solr_collection = config.getString("solr_collection");
        SolrIndexTools.delDoc(rowkey);
    }
}

package cn.bfire.coprocessor;

import com.typesafe.config.Config;
import com.typesafe.config.ConfigFactory;
import org.apache.solr.client.solrj.impl.CloudSolrClient;
import org.apache.solr.common.SolrInputDocument;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import java.util.ArrayList;
import java.util.List;
import java.util.Timer;
import java.util.TimerTask;
import java.util.concurrent.Semaphore;

/**
 * solr索引处理客户端
 * 注意问题，并发提交时，需要线程协作资源
 */
public class SolrIndexTools {
    //加载配置文件属性
    static Config config = ConfigFactory.load("userdev_pi_solr.properties");
    //log记录
    private static final Logger logger = LoggerFactory.getLogger(SolrIndexTools.class);
    //实例化solr的client
    static CloudSolrClient client = null;
    //添加批处理阈值
    static int add_batchCount = config.getInt("add_batchCount");
    //删除的批处理阈值
    static int del_batchCount = config.getInt("del_batchCount");
    //添加的集合缓冲
    static List<SolrInputDocument> add_docs = new ArrayList<SolrInputDocument>();
    //删除的集合缓冲
    static List<String> del_docs = new ArrayList<String>();


    static final List<String> zkHosts = new ArrayList<String>();

    static {
        logger.info("初始化索引调度........");
        String zk_host = config.getString("zk_host");
        String[] data = zk_host.split(",");
        for (String zkHost : data) {
            zkHosts.add(zkHost);
        }
        client = new CloudSolrClient.Builder().withZkHost(zkHosts).build();
        // 获取Solr collection
        String solr_collection = config.getString("solr_collection");
        client.setDefaultCollection(solr_collection);
        client.setZkClientTimeout(10000);
        client.setZkConnectTimeout(10000);

        //启动定时任务，第一次延迟1s执行,之后每隔指定时间30S执行一次
        Timer timer = new Timer();
        timer.schedule(new SolrCommit(), config.getInt("first_delay") * 1000, config.getInt("interval_commit_index") * 1000);
    }

    public static class SolrCommit extends TimerTask {
        @Override
        public void run() {

            logger.info("索引线程运行中........");
            //只有等于true时才执行下面的提交代码
            try {
                semp.acquire();//获取信号量
                if (add_docs.size() > 0) {
                    client.add(add_docs);//添加
                }
                if (del_docs.size() > 0) {
                    client.deleteById(del_docs);//删除
                }
                //确保都有数据才提交
                if (add_docs.size() > 0 || del_docs.size() > 0) {
                    client.commit();//共用一个提交策略
                    //清空缓冲区的添加和删除数据
                    add_docs.clear();
                    del_docs.clear();
                } else {
                    logger.info("暂无索引数据，跳过commit，继续监听......");
                }
            } catch (Exception e) {
                logger.error("间隔提交索引数据出错！", e);
            } finally {
                semp.release();//释放信号量
            }


        }
    }


    /**
     * 添加数据到临时存储中，如果
     * 大于等于batchCount时，就提交一次，
     * 再清空集合,其他情况下走对应的时间间隔提交
     *
     * @param doc 单个document对象
     */
    public static void addDoc(SolrInputDocument doc) {
        commitIndex(add_docs, add_batchCount, doc, true);
    }


    /***
     * 删除的数据添加到临时存储中，如果大于
     * 对应的批处理就直接提交，再清空集合，
     * 其他情况下走对应的时间间隔提交
     *
     * @param rowkey 删除的rowkey
     */
    public static void delDoc(String rowkey) {
        commitIndex(del_docs, del_batchCount, rowkey, false);
    }

    // 任何时候，保证只能有一个线程在提交索引，并清空集合
    final static Semaphore semp = new Semaphore(1);


    /***
     * 此方法需要加锁，并且提交索引时，与时间间隔提交是互斥的
     * 百分百确保不会丢失数据
     *
     * @param datas 用来提交的数据集合
     * @param count 对应的集合提交数量
     * @param doc   添加的单个doc
     * @param isAdd 是否为添加动作
     */
    public synchronized static void commitIndex(List datas, int count, Object doc, boolean isAdd) {
        try {
            semp.acquire();//获取信号量
            if (datas.size() >= count) {

                if (isAdd) {
                    client.add(datas);//添加数据到服务端中
                } else {
                    client.deleteById(datas);//删除数据
                }
                client.commit();//提交数据

                datas.clear();//清空临时集合


            }
        } catch (Exception e) {
            e.printStackTrace();
            logger.error("按阈值" + (isAdd == true ? "添加" : "删除") + "操作索引数据出错！", e);
        } finally {
            datas.add(doc);//添加单条数据
            semp.release();//释放信号量
        }

    }

}

<pre code_snippet_id="1962705" snippet_file_name="blog_20161102_1_8333418" style="font-family: Consolas; font-size: 11.3pt; background-color: rgb(255, 255, 255);">pom文件配置</pre>
<pre style="font-family:Consolas; font-size:11.3pt; background-color:rgb(255,255,255)"><pre code_snippet_id="1962705" snippet_file_name="blog_20161227_4_7977704" name="code" class="html"><?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/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>

    <groupId>cn.gcks</groupId>
    <artifactId>hbase</artifactId>
    <version>1.0-SNAPSHOT</version>

    <dependencies>
        <!-- https://mvnrepository.com/artifact/org.apache.solr/solr-solrj -->
        <dependency>
            <groupId>org.apache.solr</groupId>
            <artifactId>solr-solrj</artifactId>
            <version>6.2.1</version>
            <exclusions>
                <exclusion>
                    <groupId>org.slf4j</groupId>
                    <artifactId>slf4j-api</artifactId>
                </exclusion>
            </exclusions>
        </dependency>

        <!-- https://mvnrepository.com/artifact/org.apache.hbase/hbase-client -->
        <dependency>
            <groupId>org.apache.hbase</groupId>
            <artifactId>hbase-client</artifactId>
            <version>1.1.2</version>
            <exclusions>
                <exclusion>
                    <groupId>org.apache.hadoop</groupId>
                    <artifactId>*</artifactId>
                </exclusion>
            </exclusions>
        </dependency>
        <dependency>
            <groupId>org.apache.hbase</groupId>
            <artifactId>hbase-server</artifactId>
            <version>1.1.2</version>
            <exclusions>
                <exclusion>
                    <groupId>org.apache.hadoop</groupId>
                    <artifactId>*</artifactId>
                </exclusion>
            </exclusions>
        </dependency>

        <!-- https://mvnrepository.com/artifact/com.typesafe/config -->
        <dependency>
            <groupId>com.typesafe</groupId>
            <artifactId>config</artifactId>
            <version>1.3.1</version>
        </dependency>

    </dependencies>
</project></pre></pre>
<pre style="font-family:Consolas; font-size:11.3pt; background-color:rgb(255,255,255)"><p>
</p><p><span style="font-weight:bold; color:rgb(0,128,0); font-size:11.3pt; background-color:rgb(228,228,255)">userdev_pi_solr.properties</span></p><p></p><pre code_snippet_id="1962705" snippet_file_name="blog_20161227_5_7563783" name="code" class="plain">#需要建索引的列
hbase_column=oid,pi_id,statdate
# solr的collection名称
solr_collection=userdev_pi_day
#定义solr的url地址，如果是cloud模式，可以配置多个以逗号分隔
zk_host=1.1.1.1:2181,1.1.1.2:2181,1.1.1.3:2181
#调度第一次开始时，延迟多少秒执行
first_delay=10
#后台线程多久提交一次索引，单位秒
interval_commit_index=30
#添加索引的批处理数量
add_batchCount=10000
#删除索引的批处理数量
del_batchCount=2000</pre><br><br><p></p><p></p><p>②　打包代码并上传到<span style="font-family:Calibri">hdfs</span><span style="font-family:宋体">目录</span></p><p>③　修改<span style="font-family:Calibri">HBase</span><span style="font-family:宋体">表（设置自定义</span><span style="font-family:Calibri">observer</span><span style="font-family:宋体">所在</span><span style="font-family:Calibri">hdfs</span><span style="font-family:宋体">位置，以及指定自定义</span><span style="font-family:Calibri">Observer</span><span style="font-family:宋体">全类名）</span></p><p>alter 'radius:raduserlog', 'coprocessor' => 'hdfs://<span style="color:rgb(0,112,192)">/apps/hbase/jars/hbase_solr.jar</span>|cn.bfire.coprocessor.UserDevPiSolrObserver|'</p><p>2) 数据查询代码</p><p></p><pre code_snippet_id="1962705" snippet_file_name="blog_20161102_4_5934630" name="code" class="java">package cn.bfire.solr;

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.hbase.Cell;
import org.apache.hadoop.hbase.CellUtil;
import org.apache.hadoop.hbase.HBaseConfiguration;
import org.apache.hadoop.hbase.TableName;
import org.apache.hadoop.hbase.client.*;
import org.apache.hadoop.hbase.util.Bytes;
import org.apache.solr.client.solrj.SolrQuery;
import org.apache.solr.client.solrj.impl.CloudSolrClient;
import org.apache.solr.client.solrj.response.QueryResponse;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.common.SolrDocumentList;
import org.apache.solr.common.SolrInputDocument;

import java.util.ArrayList;
import java.util.Collection;
import java.util.List;

public class SolrCloudTest {
    public static final Log LOG = LogFactory.getLog(SolrCloudTest.class);
    private static CloudSolrClient cloudSolrClient;

    private static Connection connection;
    private static Table table;
    private static Get get;
    private static String defaultCollection = "userdev_pi_day";
    private static String hbaseTable = "<span style="font-family: Arial, Helvetica, sans-serif;">userdev_pi_day</span><span style="font-family: Arial, Helvetica, sans-serif;">";</span>
    List<Get> list = new ArrayList<Get>();

    static {
        final List<String> zkHosts = new ArrayList<String>();
        zkHosts.add("1.1.1.1:2181");
        zkHosts.add("1.1.1.2:2181");
        zkHosts.add("1.1.1.3:2181");
        cloudSolrClient = new CloudSolrClient.Builder().withZkHost(zkHosts).build();
        final int zkClientTimeout = 10000;
        final int zkConnectTimeout = 10000;
        cloudSolrClient.setDefaultCollection(defaultCollection);
        cloudSolrClient.setZkClientTimeout(zkClientTimeout);
        cloudSolrClient.setZkConnectTimeout(zkConnectTimeout);
        try {
            connection = ConnectionFactory.createConnection(HBaseConfiguration.create());
            table = connection.getTable(TableName.valueOf(hbaseTable));
        } catch (Exception e) {
            e.printStackTrace();
        }
    }

    private void addIndex(CloudSolrClient cloudSolrClient) throws Exception {
        Collection<SolrInputDocument> docs = new ArrayList<SolrInputDocument>();
        for (int i = 0; i <= 100; i++) {
            SolrInputDocument doc = new SolrInputDocument();
            String key = "";
            key = String.valueOf(i);
            doc.addField("rowkey", key);
            doc.addField("usermac", key + "usermac");
            doc.addField("userid", key + "userid");
            doc.addField("usertype", key + "usertype");
            doc.addField("city_id", key + "city_id");
            docs.add(doc);
        }
        LOG.info("docs info:" + docs + "
");
        cloudSolrClient.add(docs);
        cloudSolrClient.commit();
    }

    public void search(CloudSolrClient cloudSolrClient, String Str) throws Exception {
        SolrQuery query = new SolrQuery();
        query.setRows(100);
        query.setQuery(Str);
        LOG.info("query string: " + Str);
        QueryResponse response = cloudSolrClient.query(query);
        SolrDocumentList docs = response.getResults();
        System.out.println("文档个数：" + docs.getNumFound()); //数据总条数也可轻易获取
        System.out.println("查询时间：" + response.getQTime());
        System.out.println("查询总时间：" + response.getElapsedTime());
        for (SolrDocument doc : docs) {
            String rowkey = (String) doc.getFieldValue("rowkey");
            get = new Get(Bytes.toBytes(rowkey));
            list.add(get);
        }

        Result[] res = table.get(list);

        for (Result rs : res) {
            Cell[] cells = rs.rawCells();

            for (Cell cell : cells) {
                System.out.println("============");
                System.out.println(new String(CellUtil.cloneRow(cell)));
                System.out.println(new String(CellUtil.cloneFamily(cell)));
                System.out.println(new String(CellUtil.cloneQualifier(cell)));
                System.out.println(new String(CellUtil.cloneValue(cell)));
                System.out.println("============");
                break;
            }
        }


        table.close();
    }

    public static void main(String[] args) throws Exception {
        cloudSolrClient.connect();
        SolrCloudTest solrt = new SolrCloudTest();
//            solrt.addIndex(cloudSolrClient);
        solrt.search(cloudSolrClient, "userid:11111");
        cloudSolrClient.close();
    }
}
</pre><br><br><p></p><p></p><pre></pre><pre></pre></pre>
<pre></pre>
<link rel="stylesheet" href="http://static.blog.csdn.net/public/res-min/markdown_views.css?v=2.0">