Nutch相关框架安装使用最佳指南

Chinese installing and using instruction  -  The best guidance in installing and using  Nutch in China 

国内首套免费的《Nutch相关框架视频教程》        

土豆在线观看地址:http://www.tudou.com/home/item_u106249539s0p1.html 
超清原版下载地址:  http://pan.baidu.com/share/home?uk=3157595467 

下载 Nutch相关框架安装使用最佳指南.docx 

一、nutch1.2 
二、nutch1.5.1 
三、nutch2.0 
四、配置SSH 
五、安装Hadoop Cluster(伪分布式运行模式)并运行Nutch 
六、安装Hadoop Cluster(分布式运行模式)并运行Nutch 
七、配置Ganglia监控Hadoop集群和HBase集群 
八、Hadoop配置Snappy压缩 
九、Hadoop配置Lzo压缩 
十、配置zookeeper集群以运行hbase 
十一、配置Hbase集群以运行nutch-2.1(Region Servers会因为内存的问题宕机) 
十二、配置Accumulo集群以运行nutch-2.1(gora存在BUG) 
十三、配置Cassandra 集群以运行nutch-2.1(Cassandra 采用去中心化结构) 
十四、配置MySQL 单机服务器以运行nutch-2.1 
十五、nutch2.1 使用DataFileAvroStore作为数据源 
十六、nutch2.1 使用AvroStore作为数据源 
十七、配置SOLR 
十八、Nagios监控 
十九、配置Splunk 
二十、配置Pig 
二十一、配置Hive 
二十二、配置Hadoop2.x集群 



一、nutch1.2 
步骤和二大同小异,在步骤 5、配置构建路径 中需要多两个操作:在左部Package Explorer的 nutch1.2文件夹上单击右键 > Build Path > Configure Build Path...   >  选中Source选项 > Default output folder:修改nutch1.2/bin为nutch1.2/_bin,在左部Package Explorer的 nutch1.2文件夹下的bin文件夹上单击右键 > Team > 还原 
二中黄色背景部分是版本号的差异,红色部分是1.2版本没有的,绿色部分是不一样的地方,如下: 
1、Add JARs... >  nutch1.2 > lib ,选中所有的.jar文件 > OK 
2、crawl-urlfilter.txt 
3、将crawl -urlfilter.txt.template改名为crawl -urlfilter.txt 
4、修改crawl-urlfilter.txt,将 
# accept hosts in MY.DOMAIN.NAME 
+^http://([a-z0-9]*\.)*MY.DOMAIN.NAME/ 

# skip everything else 
-. 
5、cd /home/ysc/workspace/nutch1.2 
nutch1.2是一个完整的搜索引擎,nutch1.5.1只是一个爬虫。nutch1.2可以把索引提交给SOLR,也可以直接生成LUCENE索引,nutch1.5.1则只能把索引提交给SOLR: 
1、cd /home/ysc 
2、wget http://mirrors.tuna.tsinghua.edu.cn/apache/tomcat/tomcat-7/v7.0.29/bin/apache-tomcat-7.0.29.tar.gz 
3、tar -xvf apache-tomcat-7.0.29.tar.gz 
4、在左部Package Explorer的 nutch1.2文件夹下的build.xml文件上单击右键 > Run As > Ant Build... > 选中war target > Run 
5、cd /home/ysc/workspace/nutch1.2/build 
6、unzip nutch-1.2.war -d nutch-1.2 
7、cp -r nutch-1.2 /home/ysc/apache-tomcat-7.0.29/webapps 
8、vi /home/ysc/apache-tomcat-7.0.29/webapps/nutch-1.2/WEB-INF/classes/nutch-site.xml 
加入以下配置: 
<property> 
  <name>searcher.dir</name> 
  <value>/home/ysc/workspace/nutch1.2/data</value> 
  <description> 
  Path to root of crawl.  This directory is searched (in 
  order) for either the file search-servers.txt, containing a list of 
  distributed search servers, or the directory "index" containing 
  merged indexes, or the directory "segments" containing segment 
  indexes. 
  </description> 
</property> 
9、vi /home/ysc/apache-tomcat-7.0.29/conf/server.xml 
将 
<Connector port="8080" protocol="HTTP/1.1" 
               connectionTimeout="20000" 
               redirectPort="8443"/> 
改为 
<Connector port="8080" protocol="HTTP/1.1" 
               connectionTimeout="20000" 
               redirectPort="8443" URIEncoding="utf-8"/> 

10、cd /home/ysc/apache-tomcat-7.0.29/bin 
11、./startup.sh 
12、访问:http://localhost:8080/nutch-1.2/ 

关于nutch1.2更多的BUG修复及资料,请参看我在CSDN发布的资源:http://download.csdn.net/user/yangshangchuan 

二、nutch1.5.1 
1、下载并解压eclipse(集成开发环境) 
下载地址:http://www.eclipse.org/downloads/,下载Eclipse IDE for Java EE Developers 
2、安装Subclipse插件(SVN客户端) 
插件地址:http://subclipse.tigris.org/update_1.8.x, 
3、安装IvyDE插件(下载依赖Jar) 
插件地址:http://www.apache.org/dist/ant/ivyde/updatesite/ 
4、签出代码 
File > New > Project > SVN > 从SVN 检出项目 
创建新的资源库位置 > URL:https://svn.apache.org/repos/asf/nutch/tags/release-1.5.1/ > 选中URL > Finish 
弹出New Project向导,选择Java Project > Next,输入Project name:nutch1.5.1 > Finish 
5、配置构建路径 
在左部Package Explorer的 nutch1.5.1文件夹上单击右键 > Build Path > Configure Build Path...   
> 选中Source选项 > 选择src > Remove > Add Folder... > 选择src/bin, src/java, src/test 和 src/testresources(对于插件,需要选中src/plugin目录下的每一个插件目录下的src/java , src/test文件夹) > OK 
切换到Libraries选项 > 
Add Class Folder... > 选中nutch1.5.1/conf > OK 
Add JARs... >  需要选中src/plugin目录下的每一个插件目录下的lib目录下的jar文件 > OK 
Add Library... > IvyDE Managed Dependencies > Next > Main > Ivy File > Browse > ivy/ivy.xml > Finish 
切换到Order and Export选项> 
选中conf > Top 
6、执行ANT 
在左部Package Explorer的 nutch1.5.1文件夹下的build.xml文件上单击右键 > Run As > Ant Build 
在左部Package Explorer的 nutch1.5.1文件夹上单击右键 > Refresh 
在左部Package Explorer的 nutch1.5.1文件夹上单击右键 > Build Path > Configure Build Path...   >  选中Libraries选项 > Add Class Folder... >  选中build > OK 
7、修改配置文件nutch-site.xml 和regex-urlfilter.txt 
将nutch-site.xml.template改名为nutch-site.xml 
将regex-urlfilter.txt.template改名为regex-urlfilter.txt 
在左部Package Explorer的 nutch1.5.1文件夹上单击右键 > Refresh 
将如下配置项加入文件nutch-site.xml: 
<property> 
  <name>http.agent.name</name> 
  <value>nutch</value> 
</property> 
<property> 
  <name>http.content.limit</name> 
  <value>-1</value> 
</property> 
修改regex-urlfilter.txt,将 
# accept anything else 
+. 
替换为: 
+^http://([a-z0-9]*\.)*news.163.com/ 
-. 
8、开发调试 
在左部Package Explorer的 nutch1.5.1文件夹上单击右键 > New > Folder > Folder name: urls 
在刚新建的urls目录下新建一个文本文件url,文本内容为:http://news.163.com 
打开src/java下的org.apache.nutch.crawl.Crawl.java类,单击右键Run As > Run Configurations > Arguments > 在Program arguments输入框中输入: urls -dir data -depth 3 > Run 
在需要调试的地方打上断点Debug As > Java Applicaton 
9、查看结果 
查看segments目录: 
打开src/java下的org.apache.nutch.segment.SegmentReader.java类 
单击右键Run As > Java Applicaton,控制台会输出该命令的使用方法 
单击右键Run As > Run Configurations > Arguments > 在Program arguments输入框中输入: -dump data/segments/*  data/segments/dump 
用文本编辑器打开文件data/segments/dump/dump查看segments中存储的信息 

查看crawldb目录: 
打开src/java下的org.apache.nutch.crawl.CrawlDbReader.java类 
单击右键Run As > Java Applicaton,控制台会输出该命令的使用方法 
单击右键Run As > Run Configurations > Arguments > 在Program arguments输入框中输入: data/crawldb -stats 
控制台会输出 crawldb统计信息 

查看linkdb目录: 
打开src/java下的org.apache.nutch.crawl.LinkDbReader.java类 
单击右键Run As > Java Applicaton,控制台会输出该命令的使用方法 
单击右键Run As > Run Configurations > Arguments > 在Program arguments输入框中输入: data/linkdb -dump data/linkdb_dump 
用文本编辑器打开文件data/linkdb_dump/part-00000查看linkdb中存储的信息 
10、全网分步骤抓取 
在左部Package Explorer的 nutch1.5.1文件夹下的build.xml文件上单击右键 > Run As > Ant Build 
cd  /home/ysc/workspace/nutch1.5.1/runtime/local 
#准备URL列表 
wget http://rdf.dmoz.org/rdf/content.rdf.u8.gz 
gunzip content.rdf.u8.gz 
mkdir dmoz 
bin/nutch org.apache.nutch.tools.DmozParser content.rdf.u8 -subset 5000 > dmoz/url 
#注入URL 
bin/nutch inject crawl/crawldb dmoz 
#生成抓取列表 
bin/nutch generate crawl/crawldb crawl/segments 
#第一次抓取 
s1=`ls -d crawl/segments/2* | tail -1` 
echo $s1 
#抓取网页 
bin/nutch fetch $s1 
#解析网页 
bin/nutch parse $s1 
#更新URL状态 
bin/nutch updatedb crawl/crawldb $s1 
#第二次抓取 
bin/nutch generate crawl/crawldb crawl/segments -topN 1000 
s2=`ls -d crawl/segments/2* | tail -1` 
echo $s2 
bin/nutch fetch $s2 
bin/nutch parse $s2 
bin/nutch updatedb crawl/crawldb $s2 
#第三次抓取 
bin/nutch generate crawl/crawldb crawl/segments -topN 1000 
s3=`ls -d crawl/segments/2* | tail -1` 
echo $s3 
bin/nutch fetch $s3 
bin/nutch parse $s3 
bin/nutch updatedb crawl/crawldb $s3 
#生成反向链接库 
bin/nutch invertlinks crawl/linkdb -dir crawl/segments 

11、索引和搜索 
cd  /home/ysc/ 
wget http://mirror.bjtu.edu.cn/apache/lucene/solr/3.6.1/apache-solr-3.6.1.tgz 
tar -xvf apache-solr-3.6.1.tgz 
cd apache-solr-3.6.1 /example 

NUTCH_RUNTIME_HOME=/home/ysc/workspace/nutch1.5.1/runtime/local 
APACHE_SOLR_HOME=/home/ysc/apache-solr-3.6.1 

cp ${NUTCH_RUNTIME_HOME}/conf/schema.xml ${APACHE_SOLR_HOME}/example/solr/conf/ 
如果需要把网页内容存储到索引中,则修改 schema.xml文件中的 
<field name="content" type="text" stored="false" indexed="true"/> 
为 
<field name="content" type="text" stored="true" indexed="true"/> 

修改${APACHE_SOLR_HOME}/example/solr/conf/solrconfig.xml,将里面的<str name="df">text</str>都替换为<str name="df">content</str> 

把${APACHE_SOLR_HOME}/example/solr/conf/schema.xml中的 <schema name="nutch" version="1.5.1">修改为<schema name="nutch" version="1.5"> 
#启动SOLR服务器 
java -jar start.jar 

http://127.0.0.1:8983/solr/admin/ 
http://127.0.0.1:8983/solr/admin/stats.jsp 

cd  /home/ysc/workspace/nutch1.5.1/runtime/local 
#提交索引 
bin/nutch solrindex http://127.0.0.1:8983/solr/ crawl/crawldb -linkdb crawl/linkdb crawl/segments/* 

执行完整crawl: 
bin/nutch crawl urls -dir data -depth 2 -topN 100 -solr http://127.0.0.1:8983/solr/ 

使用以下命令分页查看所有索引的文档: 
http://127.0.0.1:8983/solr/select/?q=*%3A*&version=2.2&start=0&rows=10&indent=on 
标题包含“网易”的文档: 
http://127.0.0.1:8983/solr/select/?q=title%3A%E7%BD%91%E6%98%93&version=2.2&start=0&rows=10&indent=on 

12、查看索引信息 
cd  /home/ysc/ 
wget http://luke.googlecode.com/files/lukeall-3.5.0.jar 
java -jar lukeall-3.5.0.jar 
Path: /home/ysc/apache-solr-3.6.1/example/solr/data 

13、配置SOLR的中文分词 
cd  /home/ysc/ 
wget http://mmseg4j.googlecode.com/files/mmseg4j-1.8.5.zip 
unzip mmseg4j-1.8.5.zip -d  mmseg4j-1.8.5 

APACHE_SOLR_HOME=/home/ysc/apache-solr-3.6.1 
mkdir $APACHE_SOLR_HOME/example/solr/lib 
mkdir $APACHE_SOLR_HOME/example/solr/dic 
cp mmseg4j-1.8.5/mmseg4j-all-1.8.5.jar $APACHE_SOLR_HOME/example/solr/lib 
cp mmseg4j-1.8.5/data/*.dic $APACHE_SOLR_HOME/example/solr/dic 

将${APACHE_SOLR_HOME}/example/solr/conf/schema.xml文件中的 
<tokenizer class="solr.WhitespaceTokenizerFactory"/> 
和 
<tokenizer class="solr.StandardTokenizerFactory"/> 
替换为 
<tokenizer class="com.chenlb.mmseg4j.solr.MMSegTokenizerFactory" mode="complex" dicPath="/home/ysc/apache-solr-3.6.1/example/solr/dic"/> 

#重新启动SOLR服务器 
java -jar start.jar 

#重建索引,演示在开发环境中如何操作 
打开src/java下的org.apache.nutch.indexer.solr.SolrIndexer.java类 
单击右键Run As > Java Applicaton,控制台会输出该命令的使用方法 
单击右键Run As > Run Configurations > Arguments > 在Program arguments输入框中输入: http://127.0.0.1:8983/solr/  data/crawldb -linkdb  data/linkdb  data/segments/* 
使用luke重新打开索引就会发现分词起作用了 

三、nutch2.0 
nutch2.0和二中的nutch1.5.1的步骤相同,但在8、开发调试之前需要做以下配置: 
在左部Package Explorer的 nutch2.0文件夹上单击右键 > New > Folder > Folder name: data并指定数据存储方式,选如下之一: 
1、使用mysql作为数据存储 
  1)、在nutch2.0/conf/nutch-site.xml中加入如下配置: 
<property> 
  <name>storage.data.store.class</name> 
  <value>org.apache.gora.sql.store.SqlStore</value> 
</property> 
  2)、将nutch2.0/conf/gora.properties文件中的  
  gora.sqlstore.jdbc.driver=org.hsqldb.jdbc.JDBCDriver 
gora.sqlstore.jdbc.url=jdbc:hsqldb:hsql://localhost/nutchtest 
gora.sqlstore.jdbc.user=sa 
gora.sqlstore.jdbc.password= 
  修改为 
  gora.sqlstore.jdbc.driver=com.mysql.jdbc.Driver 
gora.sqlstore.jdbc.url=jdbc:mysql://127.0.0.1:3306/nutch2 
gora.sqlstore.jdbc.user=root 
gora.sqlstore.jdbc.password=ROOT 
  3)、打开nutch2.0/ivy/ivy.xml中的mysql-connector-java依赖 
  4)、sudo apt-get install mysql-server 
2、使用hbase作为数据存储 
  1)、在nutch2.0/conf/nutch-site.xml中加入如下配置: 
<property> 
  <name>storage.data.store.class</name> 
  <value>org.apache.gora.hbase.store.HBaseStore</value> 
</property> 
  2)、打开nutch2.0/ivy/ivy.xml中的gora-hbase依赖 
  3)、cd /home/ysc 
  4)、wget http://mirror.bit.edu.cn/apache/hbase/hbase-0.90.5/hbase-0.90.5.tar.gz 
  5)、tar -xvf hbase-0.90.5.tar.gz 
  6)、vi  hbase-0.90.5/conf/hbase-site.xml 
   加入以下配置: 
  <property> 
    <name>hbase.rootdir</name> 
    <value>file:///home/ysc/hbase-0.90.5-database</value> 
  </property> 
7)、hbase-0.90.5/bin/start-hbase.sh 
8)、将/home/ysc/hbase-0.90.5/hbase-0.90.5.jar加入开发环境eclipse的build path 

四、配置SSH 
三台机器 devcluster01, devcluster02, devcluster03,分别在每一台机器上面执行如下操作: 
1、sudo vi /etc/hosts 
加入以下配置: 
192.168.1.1 devcluster01 
192.168.1.2 devcluster02 
192.168.1.3 devcluster03 
2、安装SSH服务: 
  sudo apt-get install openssh-server 
3、(有提示的时候回车键确认) 
  ssh-keygen -t rsa 
  该命令会在用户主目录下创建 .ssh 目录,并在其中创建两个文件:id_rsa 私钥文件。是基于 RSA 算法创建。该私钥文件要妥善保管,不要泄漏。id_rsa.pub 公钥文件。和 id_rsa 文件是一对儿,该文件作为公钥文件,可以公开。 
4、cp .ssh/id_rsa.pub .ssh/authorized_keys 
把 三台机器 devcluster01, devcluster02, devcluster03 的文件/home/ysc/.ssh/authorized_keys的内容复制出来合并成一个文件并替换每一台机器上的/home/ysc/.ssh/authorized_keys文件 
在devcluster01上面执行时,以下两条命令的主机为02和03 
在devcluster02上面执行时,以下两条命令的主机为01和03 
在devcluster03上面执行时,以下两条命令的主机为01和02 
5、ssh-copy-id -i .ssh/id_rsa.pub ysc@ devcluster02 
6、ssh-copy-id -i .ssh/id_rsa.pub ysc@ devcluster03 
以上两条命令实际上是将 .ssh/id_rsa.pub 公钥文件追加到远程主机 server 的 user 主目录下的 .ssh/authorized_keys 文件中。 

五、安装Hadoop Cluster(伪分布式运行模式)并运行Nutch 
步骤和四大同小异,只需要1台机器 devcluster01,所以黄色背景部分全部设置为devcluster01,不需要第11步 

六、安装Hadoop Cluster(分布式运行模式)并运行Nutch 
三台机器 devcluster01, devcluster02, devcluster03(vi /etc/hostname) 
使用用户ysc登陆 devcluster01: 
1、cd /home/ysc 
2、wget http://mirrors.tuna.tsinghua.edu.cn/apache/hadoop/common/hadoop-1.1.1/hadoop-1.1.1-bin.tar.gz 
3、tar -xvf hadoop-1.1.1-bin.tar.gz 
4、cd  hadoop-1.1.1 
5、vi conf/masters 
  替换内容为 : 
  devcluster01 
6、vi conf/slaves 
  替换内容为 : 
  devcluster02 
  devcluster03 
7、vi conf/core-site.xml 
  加入配置: 
  <property> 
    <name>fs.default.name</name> 
    <value>hdfs://devcluster01:9000</value> 
    <description> 
       Where to find the Hadoop Filesystem through the network. 
       Note 9000 is not the default port. 
       (This is slightly changed from previous versions which didnt have "hdfs") 
    </description> 
  </property> 
    <property> 
     <name>hadoop.security.authorization</name> 
      <value>true</value> 
    </property> 
编辑conf/hadoop-policy.xml 
8、vi conf/hdfs-site.xml 
  加入配置: 
<property> 
  <name>dfs.name.dir</name> 
  <value>/home/ysc/dfs/filesystem/name</value> 
</property> 

<property> 
  <name>dfs.data.dir</name> 
  <value>/home/ysc/dfs/filesystem/data</value> 
</property> 

<property> 
  <name>dfs.replication</name> 
  <value>1</value> 
</property> 

<property> 
  <name>dfs.block.size</name> 
  <value>671088640</value> 
  <description>The default block size for new files.</description> 
</property> 
9、vi conf/mapred-site.xml 
  加入配置: 
<property> 
  <name>mapred.job.tracker</name> 
  <value>devcluster01:9001</value> 
  <description> 
    The host and port that the MapReduce job tracker runs at. If 
    "local", then jobs are run in-process as a single map and 
    reduce task. 
    Note 9001 is not the default port. 
  </description> 
</property> 

<property> 
  <name>mapred.reduce.tasks.speculative.execution</name> 
  <value>false</value> 
  <description>If true, then multiple instances of some reduce tasks 
               may be executed in parallel.</description> 
</property> 

<property> 
  <name>mapred.map.tasks.speculative.execution</name> 
  <value>false</value> 
  <description>If true, then multiple instances of some map tasks 
               may be executed in parallel.</description> 
</property> 

<property> 
  <name>mapred.child.java.opts</name> 
  <value>-Xmx2000m</value> 
</property> 

<property> 
  <name>mapred.tasktracker.map.tasks.maximum</name> 
  <value>4</value> 
  <description> 
    the core number of host 
  </description> 
</property> 

<property> 
  <name>mapred.map.tasks</name> 
  <value>4</value> 
</property> 

<property> 
  <name>mapred.tasktracker.reduce.tasks.maximum</name> 
  <value>4</value> 
    <description> 
    define mapred.map tasks to be number of slave hosts.the best number is the  number of slave hosts plus the core numbers of per host 
    </description> 
</property> 

<property> 
  <name>mapred.reduce.tasks</name> 
  <value>4</value> 
  <description> 
    define mapred.reduce tasks to be number of slave hosts.the best number is the  number of slave hosts plus the core numbers of per host 
  </description> 
</property> 

<property> 
  <name>mapred.output.compression.type</name> 
  <value>BLOCK</value> 
  <description>If the job outputs are to compressed as SequenceFiles, how should they be compressed? Should be one of NONE, RECORD or BLOCK. 
  </description> 
</property> 

<property> 
  <name>mapred.output.compress</name> 
  <value>true</value> 
  <description>Should the job outputs be compressed? 
  </description> 
</property> 

<property> 
  <name>mapred.compress.map.output</name> 
  <value>true</value> 
  <description>Should the outputs of the maps be compressed before being                sent across the network. Uses SequenceFile compression. 
  </description> 
</property> 

<property> 
  <name>mapred.system.dir</name> 
  <value>/home/ysc/mapreduce/system</value> 
</property> 

<property> 
  <name>mapred.local.dir</name> 
  <value>/home/ysc/mapreduce/local</value> 
</property> 
10、vi conf/hadoop-env.sh 
  追加: 
export JAVA_HOME=/home/ysc/jdk1.7.0_05 
  export HADOOP_HEAPSIZE=2000 
  #替换掉默认的垃圾回收器,因为默认的垃圾回收器在多线程环境下会有更多的wait等待 
  export HADOOP_OPTS="-server -Xmn256m -XX:+UseParNewGC -XX:+UseConcMarkSweepGC -XX:CMSInitiatingOccupancyFraction=70" 
11、复制HADOOP文件 
  scp -r /home/ysc/hadoop-1.1.1 ysc@devcluster02:/home/ysc/hadoop-1.1.1 
  scp -r /home/ysc/hadoop-1.1.1 ysc@devcluster03:/home/ysc/hadoop-1.1.1 
12、sudo vi /etc/profile 
  追加并重启系统: 
  export PATH=/home/ysc/hadoop-1.1.1/bin:$PATH 
13、格式化名称节点并启动集群 
  hadoop namenode -format 
  start-all.sh 
14、cd /home/ysc/workspace/nutch1.5.1/runtime/deploy 
  mkdir urls 
  echo http://news.163.com > urls/url 
  hadoop dfs -put urls urls 
  bin/nutch crawl urls -dir data -depth 2 -topN 100 
15、访问 http://localhost:50030 可以查看 JobTracker 的运行状态。访问 http://localhost:50060 可以查看 TaskTracker 的运行状态。访问 http://localhost:50070 可以查看 NameNode 以及整个分布式文件系统的状态,浏览分布式文件系统中的文件以及 log 等 
16、通过stop-all.sh停止集群 
17、如果NameNode和SecondaryNameNode不在同一台机器上,则在SecondaryNameNode的conf/hdfs-site.xml文件中加入配置: 
   <property> 
     <name>dfs.http.address</name> 
     <value>namenode:50070</value> 
   </property> 

七、配置Ganglia监控Hadoop集群和HBase集群 
1、服务器端(安装到master devcluster01上) 
  1)、ssh devcluster01 
  2)、addgroup ganglia 
           adduser --ingroup ganglia ganglia 
  3)、sudo apt-get install  ganglia-monitor ganglia-webfront gmetad 
   //补充:在Ubuntu10.04上,ganglia-webfront这个package名字叫ganglia-webfrontend 
   //如果install出错,则运行sudo apt-get update,如果update出错,则删除出错路径 
  4)、vi /etc/ganglia/gmond.conf 
   先找到setuid = yes,改成setuid =no; 
   在找到cluster块中的name,改成name =”hadoop-cluster”; 
  5)、sudo apt-get install rrdtool 
  6)、vi /etc/ganglia/gmetad.conf 
   在这个配置文件中增加一些datasource,即其他2个被监控的节点,增加以下内容: 
   data_source “hadoop-cluster” devcluster01:8649 devcluster02:8649 devcluster03:8649 
   gridname "Hadoop" 
2、数据源端(安装到所有slaves上) 
  1)、ssh devcluster02 
   addgroup ganglia 
   adduser --ingroup ganglia ganglia 
   sudo apt-get install  ganglia-monitor 

  2)、ssh devcluster03 
   addgroup ganglia 
   adduser --ingroup ganglia ganglia 
   sudo apt-get install  ganglia-monitor 

  3)、ssh devcluster01 
   scp /etc/ganglia/gmond.conf devcluster02:/etc/ganglia/gmond.conf 
   scp /etc/ganglia/gmond.conf devcluster03:/etc/ganglia/gmond.conf 
3、配置WEB 
  1)、ssh devcluster01 
  2)、sudo ln -s /usr/share/ganglia-webfrontend /var/www/ganglia 
  3)、vi /etc/apache2/apache2.conf 
   添加: 
   ServerName devcluster01 
4、重启服务 
  1)、ssh devcluster02 
   sudo /etc/init.d/ganglia-monitor restart 
   ssh devcluster03 
   sudo /etc/init.d/ganglia-monitor restart 
  2)、ssh devcluster01 
   sudo /etc/init.d/ganglia-monitor restart 
   sudo /etc/init.d/gmetad restart 
   sudo /etc/init.d/apache2 restart 
5、访问页面 
  http:// devcluster01/ganglia 
6、集成hadoop 
  1)、ssh devcluster01 
  2)、cd /home/ysc/hadoop-1.1.1 
  3)、vi conf/hadoop-metrics2.properties 
  # 大于0.20以后的版本用ganglia31  *.sink.ganglia.class=org.apache.hadoop.metrics2.sink.ganglia.GangliaSink31 
  *.sink.ganglia.period=10 
  # default for supportsparse is false 
  *.sink.ganglia.supportsparse=true 
*.sink.ganglia.slope=jvm.metrics.gcCount=zero,jvm.metrics.memHeapUsedM=both 
*.sink.ganglia.dmax=jvm.metrics.threadsBlocked=70,jvm.metrics.memHeapUsedM=40 
  #广播IP地址,这是缺省的,统一设该值(只能用组播地址239.2.11.71) 
  namenode.sink.ganglia.servers=239.2.11.71:8649 
  datanode.sink.ganglia.servers=239.2.11.71:8649 
  jobtracker.sink.ganglia.servers=239.2.11.71:8649 
  tasktracker.sink.ganglia.servers=239.2.11.71:8649 
  maptask.sink.ganglia.servers=239.2.11.71:8649 
  reducetask.sink.ganglia.servers=239.2.11.71:8649 
  dfs.class=org.apache.hadoop.metrics.ganglia.GangliaContext31 
  dfs.period=10 
  dfs.servers=239.2.11.71:8649 
  mapred.class=org.apache.hadoop.metrics.ganglia.GangliaContext31 
  mapred.period=10 
  mapred.servers=239.2.11.71:8649 
  jvm.class=org.apache.hadoop.metrics.ganglia.GangliaContext31 
  jvm.period=10 
  jvm.servers=239.2.11.71:8649 
  4)、scp conf/hadoop-metrics2.properties root@devcluster02:/home/ysc/hadoop-1.1.1/conf/hadoop-metrics2.properties 
  5)、scp conf/hadoop-metrics2.properties root@devcluster03:/home/ysc/hadoop-1.1.1/conf/hadoop-metrics2.properties 
  6)、stop-all.sh 
  7)、start-all.sh 
7、集成hbase 
  1)、ssh devcluster01 
  2)、cd /home/ysc/hbase-0.92.2 
  3)、vi conf/hadoop-metrics.properties(只能用组播地址239.2.11.71) 
   hbase.extendedperiod = 3600 
   hbase.class=org.apache.hadoop.metrics.ganglia.GangliaContext31 
   hbase.period=10 
   hbase.servers=239.2.11.71:8649 
   jvm.class=org.apache.hadoop.metrics.ganglia.GangliaContext31 
   jvm.period=10 
   jvm.servers=239.2.11.71:8649 
   rpc.class=org.apache.hadoop.metrics.ganglia.GangliaContext31 
   rpc.period=10 
   rpc.servers=239.2.11.71:8649 
  4)、scp conf/hadoop-metrics.properties root@devcluster02:/home/ysc/ hbase-0.92.2/conf/hadoop-metrics.properties 
  5)、scp conf/hadoop-metrics.properties root@devcluster03:/home/ysc/ hbase-0.92.2/conf/hadoop-metrics.properties 
  6)、stop-hbase.sh 
  7)、start-hbase.sh 

八、Hadoop配置Snappy压缩 
1、wget http://snappy.googlecode.com/files/snappy-1.0.5.tar.gz 
2、tar -xzvf snappy-1.0.5.tar.gz 
3、cd snappy-1.0.5 
4、./configure 
5、make 
6、make install 
7、scp /usr/local/lib/libsnappy* devcluster01:/home/ysc/hadoop-1.1.1/lib/native/Linux-amd64-64/ 
scp /usr/local/lib/libsnappy* devcluster02:/home/ysc/hadoop-1.1.1/lib/native/Linux-amd64-64/ 
scp /usr/local/lib/libsnappy* devcluster03:/home/ysc/hadoop-1.1.1/lib/native/Linux-amd64-64/ 
8、vi /etc/profile 
  追加: 
  export LD_LIBRARY_PATH=/home/ysc/hadoop-1.1.1/lib/native/Linux-amd64-64 
9、修改mapred-site.xml 
  <property> 
    <name>mapred.output.compression.type</name> 
    <value>BLOCK</value> 
    <description>If the job outputs are to compressed as SequenceFiles, how should 
        they be compressed? Should be one of NONE, RECORD or BLOCK. 
    </description> 
  </property> 

  <property> 
    <name>mapred.output.compress</name> 
    <value>true</value> 
    <description>Should the job outputs be compressed? 
    </description> 
  </property> 

  <property> 
    <name>mapred.compress.map.output</name> 
    <value>true</value> 
    <description>Should the outputs of the maps be compressed before being 
        sent across the network. Uses SequenceFile compression. 
    </description> 
  </property> 

  <property> 
    <name>mapred.map.output.compression.codec</name> 
    <value>org.apache.hadoop.io.compress.SnappyCodec</value> 
    <description>If the map outputs are compressed, how should they be 
        compressed? 
    </description> 
  </property> 

  <property> 
    <name>mapred.output.compression.codec</name> 
    <value>org.apache.hadoop.io.compress.SnappyCodec</value> 
    <description>If the job outputs are compressed, how should they be compressed? 
    </description> 
  </property> 

九、Hadoop配置Lzo压缩 
1、wget http://www.oberhumer.com/opensource/lzo/download/lzo-2.06.tar.gz 
2、tar -zxvf lzo-2.06.tar.gz 
3、cd lzo-2.06 
4、./configure --enable-shared 
5、make 
6、make install 
7、scp /usr/local/lib/liblzo2.* devcluster01:/lib/x86_64-linux-gnu 
scp /usr/local/lib/liblzo2.* devcluster02:/lib/x86_64-linux-gnu 
scp /usr/local/lib/liblzo2.* devcluster03:/lib/x86_64-linux-gnu 
8、wget http://hadoop-gpl-compression.apache-extras.org.codespot.com/files/hadoop-gpl-compression-0.1.0-rc0.tar.gz 
9、tar -xzvf hadoop-gpl-compression-0.1.0-rc0.tar.gz 
10、cd hadoop-gpl-compression-0.1.0 
11、cp lib/native/Linux-amd64-64/* /home/ysc/hadoop-1.1.1/lib/native/Linux-amd64-64/ 
12、cp hadoop-gpl-compression-0.1.0.jar /home/ysc/hadoop-1.1.1/lib/(这里hadoop集群的版本要和compression使用的版本一致) 
13、scp -r /home/ysc/hadoop-1.1.1/lib devcluster02:/home/ysc/hadoop-1.1.1/ 
scp -r /home/ysc/hadoop-1.1.1/lib devcluster03:/home/ysc/hadoop-1.1.1/ 
14、vi /etc/profile 
  追加: 
  export LD_LIBRARY_PATH=/home/ysc/hadoop-1.1.1/lib/native/Linux-amd64-64 
15、修改core-site.xml 
  <property> 
    <name>io.compression.codecs</name> 
    <value>com.hadoop.compression.lzo.LzoCodec,org.apache.hadoop.io.compress.DefaultCodec,org.apache.hadoop.io.compress.GzipCodec,org.apache.hadoop.io.compress.BZip2Codec,org.apache.hadoop.io.compress.SnappyCodec</value>
    <description>A list of the compression codec classes that can be used 
        for compression/decompression.</description> 
  </property> 

  <property> 
    <name>io.compression.codec.lzo.class</name> 
    <value>com.hadoop.compression.lzo.LzoCodec</value> 
  </property> 

  <property> 
    <name>fs.trash.interval</name> 
    <value>1440</value> 
    <description>Number of minutes between trash checkpoints. 
    If zero, the trash feature is disabled. 
    </description> 
  </property> 
16、修改mapred-site.xml 
  <property> 
    <name>mapred.output.compression.type</name> 
    <value>BLOCK</value> 
    <description>If the job outputs are to compressed as SequenceFiles, how should 
        they be compressed? Should be one of NONE, RECORD or BLOCK. 
    </description> 
  </property> 

  <property> 
    <name>mapred.output.compress</name> 
    <value>true</value> 
    <description>Should the job outputs be compressed? 
    </description> 
  </property> 

  <property> 
    <name>mapred.compress.map.output</name> 
    <value>true</value> 
    <description>Should the outputs of the maps be compressed before being 
        sent across the network. Uses SequenceFile compression. 
    </description> 
  </property> 

  <property> 
    <name>mapred.map.output.compression.codec</name> 
    <value>com.hadoop.compression.lzo.LzoCodec</value> 
    <description>If the map outputs are compressed, how should they be 
        compressed? 
    </description> 
  </property> 

  <property> 
    <name>mapred.output.compression.codec</name> 
    <value>com.hadoop.compression.lzo.LzoCodec</value> 
    <description>If the job outputs are compressed, how should they be compressed? 
    </description> 
  </property> 

十、配置zookeeper集群以运行hbase 
1、ssh devcluster01 
2、cd /home/ysc 
3、wget http://mirror.bjtu.edu.cn/apache/zookeeper/stable/zookeeper-3.4.5.tar.gz 
4、tar -zxvf  zookeeper-3.4.5.tar.gz 
5、cd zookeeper-3.4.5 
6、cp conf/zoo_sample.cfg  conf/zoo.cfg 
7、vi conf/zoo.cfg 
  修改:dataDir=/home/ysc/zookeeper 
  添加: 
   server.1=devcluster01:2888:3888 
   server.2=devcluster02:2888:3888 
   server.3=devcluster03:2888:3888 
   maxClientCnxns=100 
8、scp -r  zookeeper-3.4.5  devcluster01:/home/ysc 
scp -r  zookeeper-3.4.5  devcluster02:/home/ysc 
scp -r  zookeeper-3.4.5  devcluster03:/home/ysc 
9、分别在三台机器上面执行: 
  ssh devcluster01 
  mkdir /home/ysc/zookeeper(注:dataDir是zookeeper的数据目录,需要手动创建) 
  echo 1 > /home/ysc/zookeeper/myid 
  ssh devcluster02 
  mkdir /home/ysc/zookeeper 
  echo 2 > /home/ysc/zookeeper/myid 
  ssh devcluster03 
  mkdir /home/ysc/zookeeper 
  echo 3 > /home/ysc/zookeeper/myid 
10、分别在三台机器上面执行: 
  cd /home/ysc/zookeeper-3.4.5 
  bin/zkServer.sh start 
  bin/zkCli.sh -server devcluster01:2181 
  bin/zkServer.sh status 

十一、配置Hbase集群以运行nutch-2.1(Region Servers会因为内存的问题宕机) 
1、nutch-2.1使用gora-0.2.1, gora-0.2.1使用hbase-0.90.4,hbase-0.90.4和hadoop-1.1.1不兼容,hbase-0.94.4和gora-0.2.1不兼容,hbase-0.92.2没问题。hbase存在系统时间同步的问题,并且误差要再30s以内。 
sudo apt-get install ntp 
sudo ntpdate -u 210.72.145.44 
2、HBase是数据库,会在同一时间使用很多的文件句柄。大多数linux系统使用的默认值1024是不能满足的。还需要修改 hbase 用户的 nproc,在压力下,如果过低会造成 OutOfMemoryError异常。 
vi /etc/security/limits.conf 
添加: 
   ysc soft nproc 32000 
   ysc hard nproc 32000 
   ysc soft nofile 32768 
   ysc hard nofile 32768 
vi /etc/pam.d/common-session 
添加: 
   session required  pam_limits.so 
3、登陆master,下载并解压hbase 
  ssh devcluster01 
  cd /home/ysc 
  wget http://apache.etoak.com/hbase/hbase-0.92.2/hbase-0.92.2.tar.gz 
  tar -zxvf hbase-0.92.2.tar.gz 
  cd hbase-0.92.2 
4、修改配置文件hbase-env.sh 
  vi conf/hbase-env.sh 
  追加: 
  export JAVA_HOME=/home/ysc/jdk1.7.0_05 
  export HBASE_MANAGES_ZK=false 
  export HBASE_HEAPSIZE=10000 
  #替换掉默认的垃圾回收器,因为默认的垃圾回收器在多线程环境下会有更多的wait等待 
  export HBASE_OPTS="-server -Xmn256m -XX:+UseParNewGC -XX:+UseConcMarkSweepGC -XX:CMSInitiatingOccupancyFraction=70" 
5、修改配置文件hbase-site.xml 
  vi conf/hbase-site.xml 
  <property>  
   <name>hbase.rootdir</name>  
   <value>hdfs://devcluster01:9000/hbase</value>     
  </property> 
  <property>  
   <name>hbase.cluster.distributed</name>  
   <value>true</value>  
  </property>  
  <property>   
   <name>hbase.zookeeper.quorum</name>        
   <value>devcluster01,devcluster02,devcluster03</value>   
  </property> 
  <property> 
   <name>hfile.block.cache.size</name> 
   <value>0.25</value> 
   <description> 
    Percentage of maximum heap (-Xmx setting) to allocate to block cache 
    used by HFile/StoreFile. Default of 0.25 means allocate 25%. 
    Set to 0 to disable but it's not recommended. 
   </description> 
  </property> 
  <property> 
   <name>hbase.regionserver.global.memstore.upperLimit</name> 
   <value>0.4</value> 
   <description>Maximum size of all memstores in a region server before new 
     updates are blocked and flushes are forced. Defaults to 40% of heap 
   </description> 
  </property> 
    <property> 
   <name>hbase.regionserver.global.memstore.lowerLimit</name> 
   <value>0.35</value> 
   <description>When memstores are being forced to flush to make room in 
    memory, keep flushing until we hit this mark. Defaults to 35% of heap. 
    This value equal to hbase.regionserver.global.memstore.upperLimit causes 
    the minimum possible flushing to occur when updates are blocked due to 
    memstore limiting. 
   </description> 
    </property> 
  <property> 
   <name>hbase.hregion.majorcompaction</name> 
   <value>0</value> 
   <description>The time (in miliseconds) between 'major' compactions of all 
    HStoreFiles in a region.  Default: 1 day. 
    Set to 0 to disable automated major compactions. 
   </description> 
  </property> 
6、修改配置文件regionservers 
  vi conf/regionservers 
  devcluster01 
  devcluster02 
  devcluster03 
7、因为HBase建立在Hadoop之上,Hadoop使用的hadoop*.jar和HBase使用的 必须 一致。所以要将 HBase lib 目录下的hadoop*.jar替换成Hadoop里面的那个,防止版本冲突。 
  cp  /home/ysc/hadoop-1.1.1/hadoop-core-1.1.1.jar  /home/ysc/hbase-0.92.2/lib 
  rm  /home/ysc/hbase-0.92.2/lib/hadoop-core-1.0.3.jar 
8、复制文件到regionservers 
  scp -r /home/ysc/hbase-0.92.2 devcluster01:/home/ysc 
  scp -r /home/ysc/hbase-0.92.2 devcluster02:/home/ysc 
  scp -r /home/ysc/hbase-0.92.2 devcluster03:/home/ysc 
9、启动hadoop并创建目录 
  hadoop fs -mkdir /hbase 
10、管理HBase集群: 
  启动初始 HBase 集群: 
   bin/start-hbase.sh 
  停止HBase 集群: 
   bin/stop-hbase.sh 
  启动额外备份主服务器,可以启动到 9 个备份服务器 (总数10 个): 
   bin/local-master-backup.sh start 1 
   bin/local-master-backup.sh start 2 3 
  启动更多 regionservers, 支持到 99 个额外regionservers (总100个): 
   bin/local-regionservers.sh start 1 
   bin/local-regionservers.sh start 2 3 4 5 
  停止备份主服务器: 
   cat /tmp/hbase-ysc-1-master.pid |xargs kill -9 
  停止单独 regionserver: 
   bin/local-regionservers.sh stop 1 
  使用HBase命令行模式: 
   bin/hbase shell 
11、web界面 
  http://devcluster01:60010 
  http://devcluster01:60030 
12、如运行nutch2.1则方法一: 
  cp conf/hbase-site.xml /home/ysc/nutch-2.1/conf 
  cd /home/ysc/nutch-2.1 
  ant 
  cd runtime/deploy 
  unzip -d apache-nutch-2.1 apache-nutch-2.1.job 
  rm  apache-nutch-2.1.job 
  cd apache-nutch-2.1 
  rm lib/hbase-0.90.4.jar 
  cp /home/ysc/hbase-0.92.2/hbase-0.92.2.jar  lib 
  zip -r ../apache-nutch-2.1.job ./* 
  cd .. 
  rm -r apache-nutch-2.1 
13、如运行nutch2.1则方法二: 
  cp conf/hbase-site.xml /home/ysc/nutch-2.1/conf 
  cd /home/ysc/nutch-2.1 
  cp /home/ysc/hbase-0.92.2/hbase-0.92.2.jar  lib 
  ant 
  cd runtime/deploy 
  zip -d apache-nutch-2.1.job lib/hbase-0.90.4.jar 

启用snappy压缩: 
1、vi conf/gora-hbase-mapping.xml 
  在family上面添加属性:compression="SNAPPY" 
2、mkdir /home/ysc/hbase-0.92.2/lib/native/Linux-amd64-64 
3、cp /home/ysc/hadoop-1.1.1/lib/native/Linux-amd64-64/* /home/ysc/hbase-0.92.2/lib/native/Linux-amd64-64 
4、vi /home/ysc/hbase-0.92.2/conf/hbase-site.xml 
  增加: 
                <property> 
                        <name>hbase.regionserver.codecs</name> 
                        <value>snappy</value> 
                </property> 



十二、配置Accumulo集群以运行nutch-2.1(gora存在BUG) 
1、wget http://apache.etoak.com/accumulo/1.4.2/accumulo-1.4.2-dist.tar.gz 
2、tar -xzvf accumulo-1.4.2-dist.tar.gz 
3、cd accumulo-1.4.2 
4、cp conf/examples/3GB/standalone/* conf 
5、vi conf/accumulo-env.sh 
  export HADOOP_HOME=/home/ysc/cluster3 
  export ZOOKEEPER_HOME=/home/ysc/zookeeper-3.4.5 
  export JAVA_HOME=/home/jdk1.7.0_01 
  export ACCUMULO_HOME=/home/ysc/accumulo-1.4.2 
6、vi conf/slaves 
  devcluster01 
  devcluster02 
  devcluster03 
7、vi conf/masters 
  devcluster01 
8、vi conf/accumulo-site.xml 
  <property> 
    <name>instance.zookeeper.host</name> 
    <value>host6:2181,host8:2181</value> 
    <description>comma separated list of zookeeper servers</description> 
  </property> 

  <property> 
    <name>logger.dir.walog</name> 
    <value>walogs</value> 
    <description>The directory used to store write-ahead logs on the local filesystem. It is possible to specify a comma-separated list of directories.</description> 
  </property> 

  <property> 
    <name>instance.secret</name> 
    <value>ysc</value> 
    <description>A secret unique to a given instance that all servers must know in order to communicate with one another. 
        Change it before initialization. To change it later use ./bin/accumulo org.apache.accumulo.server.util.ChangeSecret [oldpasswd] [newpasswd], 
        and then update this file. 
    </description> 
  </property> 

  <property> 
    <name>tserver.memory.maps.max</name> 
    <value>3G</value> 
  </property> 

  <property> 
    <name>tserver.cache.data.size</name> 
    <value>50M</value> 
  </property> 

  <property> 
    <name>tserver.cache.index.size</name> 
    <value>512M</value> 
  </property> 

  <property> 
    <name>trace.password</name> 
    <!-- 
   change this to the root user's password, and/or change the user below 
     --> 
    <value>ysc</value> 
  </property> 

  <property> 
    <name>trace.user</name> 
    <value>root</value> 
  </property> 
9、bin/accumulo init 
10、bin/start-all.sh 
11、bin/stop-all.sh 
12、web访问:http://devcluster01:50095/ 

修改nutch2.1: 
1、cd  /home/ysc/nutch-2.1 
2、vi  conf/gora.properties 
  增加: 
  gora.datastore.default=org.apache.gora.accumulo.store.AccumuloStore 
  gora.datastore.accumulo.mock=false 
  gora.datastore.accumulo.instance=accumulo 
  gora.datastore.accumulo.zookeepers=host6,host8 
  gora.datastore.accumulo.user=root 
  gora.datastore.accumulo.password=ysc 
3、vi  conf/nutch-site.xml 
  增加: 
  <property> 
    <name>storage.data.store.class</name> 
    <value>org.apache.gora.accumulo.store.AccumuloStore</value> 
  </property> 
4、vi ivy/ivy.xml 
  增加: 
  <dependency org="org.apache.gora" name="gora-accumulo" rev="0.2.1" conf="*->default" /> 
5、升级accumulo 
  cp /home/ysc/accumulo-1.4.2/lib/accumulo-core-1.4.2.jar  /home/ysc/nutch-2.1/lib 
  cp /home/ysc/accumulo-1.4.2/lib/accumulo-start-1.4.2.jar  /home/ysc/nutch-2.1/lib 
  cp /home/ysc/accumulo-1.4.2/lib/cloudtrace-1.4.2.jar  /home/ysc/nutch-2.1/lib 
6、ant 
7、cd runtime/deploy 
8、删除旧jar 
  zip -d apache-nutch-2.1.job lib/accumulo-core-1.4.0.jar 
  zip -d apache-nutch-2.1.job lib/accumulo-start-1.4.0.jar 
  zip -d apache-nutch-2.1.job lib/cloudtrace-1.4.2.jar 

十三、配置Cassandra 集群以运行nutch-2.1(Cassandra 采用去中心化结构) 
1、vi /etc/hosts(注意:需要登录到每一台机器上面,将localhost解析到实际地址) 
  192.168.1.1       localhost 
2、wget http://labs.mop.com/apache-mirror/cassandra/1.2.0/apache-cassandra-1.2.0-bin.tar.gz 
3、tar -xzvf  apache-cassandra-1.2.0-bin.tar.gz 
4、cd apache-cassandra-1.2.0 
5、vi conf/cassandra-env.sh 
  增加: 
  MAX_HEAP_SIZE="4G" 
  HEAP_NEWSIZE="800M" 
6、vi conf/log4j-server.properties 
  修改: 
  log4j.appender.R.File=/home/ysc/cassandra/system.log 
7、vi conf/cassandra.yaml 
  修改: 
  cluster_name: 'Cassandra  Cluster' 
  data_file_directories: 
      - /home/ysc/cassandra/data 
  commitlog_directory: /home/ysc/cassandra/commitlog 
  saved_caches_directory: /home/ysc/cassandra/saved_caches 

  - seeds: "192.168.1.1" 
  listen_address: 192.168.1.1 
  rpc_address: 192.168.1.1 

  thrift_framed_transport_size_in_mb: 1023 
  thrift_max_message_length_in_mb: 1024 
8、vi bin/stop-server 
  增加: 
  user=`whoami` 
  pgrep -u $user -f cassandra | xargs kill -9 
9、复制cassandra到其他节点: 
  cd .. 
  scp -r apache-cassandra-1.2.0 devcluster02:/home/ysc 
  scp -r apache-cassandra-1.2.0 devcluster03:/home/ysc 
  分别在devcluster02和devcluster03上面修改: 
  vi conf/cassandra.yaml 
   listen_address: 192.168.1.2 
   rpc_address: 192.168.1.2 
  vi conf/cassandra.yaml 
   listen_address: 192.168.1.3 
   rpc_address: 192.168.1.3 
10、分别在3个节点上面运行 
  bin/cassandra 
  bin/cassandra -f   参数 -f 的作用是让 Cassandra 以前端程序方式运行,这样有利于调试和观察日志信息,而在实际生产环境中这个参数是不需要的(即 Cassandra 会以 daemon 方式运行) 
11、bin/nodetool -host devcluster01 ring 
        bin/nodetool -host devcluster01 info 
12、bin/stop-server 
13、bin/cassandra-cli 

修改nutch2.1: 
1、cd  /home/ysc/nutch-2.1 
2、vi  conf/gora.properties 
  增加: 
  gora.cassandrastore.servers=host2:9160,host6:9160,host8:9160 
3、vi  conf/nutch-site.xml 
  增加: 
  <property> 
    <name>storage.data.store.class</name> 
    <value>org.apache.gora.cassandra.store.CassandraStore</value> 
  </property> 
4、vi ivy/ivy.xml 
  增加: 
  <dependency org="org.apache.gora" name="gora-cassandra" rev="0.2.1" conf="*->default" /> 
5、升级cassandra 
  cp /home/ysc/apache-cassandra-1.2.0/lib/apache-cassandra-1.2.0.jar  /home/ysc/nutch-2.1/lib 
  cp /home/ysc/apache-cassandra-1.2.0/lib/apache-cassandra-thrift-1.2.0.jar  /home/ysc/nutch-2.1/lib 
  cp /home/ysc/apache-cassandra-1.2.0/lib/jline-1.0.jar  /home/ysc/nutch-2.1/lib 
6、ant 
7、cd runtime/deploy 
8、删除旧jar 
  zip -d apache-nutch-2.1.job lib/cassandra-thrift-1.1.2.jar 
  zip -d apache-nutch-2.1.job lib/jline-0.9.1.jar 

十四、配置MySQL 单机服务器以运行nutch-2.1 
1、apt-get install mysql-server mysql-client 
2、vi /etc/mysql/my.cnf 
  修改: 
  bind-address            = 221.194.43.2 
  在[client]下增加: 
  default-character-set=utf8 
  在[mysqld]下增加: 
  default-character-set=utf8 
3、mysql –uroot –pysc 
  SHOW VARIABLES LIKE '%character%'; 
4、service mysql restart 
5、mysql –uroot –pysc 
  GRANT ALL PRIVILEGES ON *.* TO root@"%" IDENTIFIED BY "ysc"; 
6、vi conf/gora-sql-mapping.xml 
  修改字段的长度 
  <primarykey column="id" length="333"/> 
  <field name="content" column="content" /> 
  <field name="text" column="text" length="19892"/> 
7、启动nutch之后登陆mysql 
   ALTER TABLE webpage MODIFY COLUMN content MEDIUMBLOB; 
   ALTER TABLE webpage MODIFY COLUMN text MEDIUMTEXT; 
   ALTER TABLE webpage MODIFY COLUMN title MEDIUMTEXT; 
   ALTER TABLE webpage MODIFY COLUMN reprUrl MEDIUMTEXT; 
   ALTER TABLE webpage MODIFY COLUMN baseUrl MEDIUMTEXT; 
   ALTER TABLE webpage MODIFY COLUMN typ MEDIUMTEXT; 
   ALTER TABLE webpage MODIFY COLUMN inlinks MEDIUMBLOB; 
   ALTER TABLE webpage MODIFY COLUMN outlinks MEDIUMBLOB; 

修改nutch2.1: 
1、cd  /home/ysc/nutch-2.1 
2、vi  conf/gora.properties 
  增加: 
   gora.sqlstore.jdbc.driver=com.mysql.jdbc.Driver 
gora.sqlstore.jdbc.url=jdbc:mysql://host2:3306/nutch?createDatabaseIfNotExist=true&useUnicode=true&characterEncoding=utf8 
  gora.sqlstore.jdbc.user=root 
  gora.sqlstore.jdbc.password=ysc 
3、vi  conf/nutch-site.xml 
  增加: 
  <property> 
    <name>storage.data.store.class</name> 
    <value>org.apache.gora.sql.store.SqlStore </value> 
  </property> 

  <property> 
    <name>encodingdetector.charset.min.confidence</name> 
    <value>1</value> 
    <description>A integer between 0-100 indicating minimum confidence value 
    for charset auto-detection. Any negative value disables auto-detection. 
    </description> 
  </property> 
4、vi ivy/ivy.xml 
  增加: 
  <dependency org="mysql" name="mysql-connector-java" rev="5.1.18" conf="*->default"/> 

十五、nutch2.1 使用DataFileAvroStore作为数据源 
1、cd  /home/ysc/nutch-2.1 
2、vi  conf/gora.properties 
  增加: 
  gora.datafileavrostore.output.path=datafileavrostore 
  gora.datafileavrostore.input.path=datafileavrostore 
3、vi  conf/nutch-site.xml 
  增加: 
  <property> 
    <name>storage.data.store.class</name> 
    <value>org.apache.gora.avro.store.DataFileAvroStore</value> 
  </property> 

  <property> 
    <name>encodingdetector.charset.min.confidence</name> 
    <value>1</value> 
    <description>A integer between 0-100 indicating minimum confidence value 
    for charset auto-detection. Any negative value disables auto-detection. 
    </description> 
  </property> 



十六、nutch2.1 使用AvroStore作为数据源 
1、cd  /home/ysc/nutch-2.1 
2、vi  conf/gora.properties 
  增加: 
  gora.avrostore.codec.type=BINARY 
  gora.avrostore.input.path=avrostore 
  gora.avrostore.output.path=avrostore 
3、vi  conf/nutch-site.xml 
  增加: 
  <property> 
    <name>storage.data.store.class</name> 
    <value>org.apache.gora.avro.store.AvroStore</value> 
  </property> 

  <property> 
    <name>encodingdetector.charset.min.confidence</name> 
    <value>1</value> 
    <description>A integer between 0-100 indicating minimum confidence value 
    for charset auto-detection. Any negative value disables auto-detection. 
    </description> 
  </property> 



十七、配置SOLR 
配置tomcat: 
1、wget http://www.fayea.com/apache-mirror/tomcat/tomcat-7/v7.0.35/bin/apache-tomcat-7.0.35.tar.gz 
2、tar -xzvf apache-tomcat-7.0.35.tar.gz 
3、cd apache-tomcat-7.0.35 
4、vi conf/server.xml 
增加URIEncoding="UTF-8": 
  <Connector port="8080" protocol="HTTP/1.1" 
       connectionTimeout="20000" 
       redirectPort="8443" URIEncoding="UTF-8"/> 
5、mkdir conf/Catalina 
6、mkdir conf/Catalina/localhost 
7、vi conf/Catalina/localhost/solr.xml 
增加: 
  <Context path="/solr"> 
   <Environment name="solr/home" type="java.lang.String" value="/home/ysc/solr/configuration/" override="false"/> 
  </Context> 
8、cd .. 

下载SOLR: 
1、wget http://mirrors.tuna.tsinghua.edu.cn/apache/lucene/solr/4.1.0/solr-4.1.0.tgz 
2、tar -xzvf solr-4.1.0.tgz 

复制资源: 
1、mkdir /home/ysc/solr 
2、cp -r solr-4.1.0/example/solr  /home/ysc/solr/configuration 
3、unzip solr-4.1.0/example/webapps/solr.war -d /home/ysc/apache-tomcat-7.0.35/webapps/solr 

配置nutch: 
1、复制schema: 
  cp /home/ysc/nutch-1.6/conf/schema-solr4.xml /home/ysc/solr/configuration/collection1/conf/schema.xml 
2、vi /home/ysc/solr/configuration/collection1/conf/schema.xml 
  在<fields>下增加: 
  <field name="_version_" type="long" indexed="true" stored="true"/> 

配置中文分词: 
1、wget http://mmseg4j.googlecode.com/files/mmseg4j-1.9.1.v20130120-SNAPSHOT.zip 
2、unzip mmseg4j-1.9.1.v20130120-SNAPSHOT.zip 
3、cp mmseg4j-1.9.1-SNAPSHOT/dist/* /home/ysc/apache-tomcat-7.0.35/webapps/solr/WEB-INF/lib 
4、unzip mmseg4j-1.9.1-SNAPSHOT/dist/mmseg4j-core-1.9.1-SNAPSHOT.jar -d  mmseg4j-1.9.1-SNAPSHOT/dist/mmseg4j-core-1.9.1-SNAPSHOT 
5、mkdir /home/ysc/dic 
6、cp   mmseg4j-1.9.1-SNAPSHOT/dist/mmseg4j-core-1.9.1-SNAPSHOT/data/* /home/ysc/dic 
7、vi /home/ysc/solr/configuration/collection1/conf/schema.xml 
  将文件中的 
  <tokenizer class="solr.WhitespaceTokenizerFactory"/> 
  和 
  <tokenizer class="solr.StandardTokenizerFactory"/> 
  替换为 
  <tokenizer class="com.chenlb.mmseg4j.solr.MMSegTokenizerFactory" mode="complex" dicPath="/home/ysc/dic"/> 

配置tomcat本地库: 
1、wget http://apache.spd.co.il/apr/apr-1.4.6.tar.gz 
2、tar -xzvf apr-1.4.6.tar.gz 
3、cd apr-1.4.6 
4、./configure 
5、make 
6、make  install 

1、wget http://mirror.bjtu.edu.cn/apache/apr/apr-util-1.5.1.tar.gz 
2、tar -xzvf apr-util-1.5.1.tar.gz 
3、cd apr-util-1.5.1 
4、./configure --with-apr=/usr/local/apr 
5、make 
6、make  install 

1、wget http://mirror.bjtu.edu.cn/apache//tomcat/tomcat-connectors/native/1.1.24/source/tomcat-native-1.1.24-src.tar.gz 
2、tar -zxvf tomcat-native-1.1.24-src.tar.gz 
3、cd tomcat-native-1.1.24-src/jni/native 
4、./configure --with-apr=/usr/local/apr \ 
                --with-java-home=/home/ysc/jdk1.7.0_01 \ 
                --with-ssl=no \ 
                --prefix=/home/ysc/apache-tomcat-7.0.35 
5、make 
6、make  install 
7、vi /etc/profile 
增加: 
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/ysc/apache-tomcat-7.0.35/lib:/usr/local/apr/lib 
8、source /etc/profile 

启动tomcat: 
cd apache-tomcat-7.0.35 
bin/catalina.sh start 
http://devcluster01:8080/solr/ 

十八、Nagios监控 
服务端: 
1、apt-get install apache2 nagios3 nagios-nrpe-plugin 
  输入密码:nagiosadmin 
2、apt-get install nagios3-doc 
3、vi /etc/nagios3/conf.d/hostgroups_nagios2.cfg 
   define hostgroup { 
     hostgroup_name  nagios-servers 
     alias           nagios servers 
     members         devcluster01,devcluster02,devcluster03 
   } 
4、cp  /etc/nagios3/conf.d/localhost_nagios2.cfg /etc/nagios3/conf.d/devcluster01_nagios2.cfg 
  vi /etc/nagios3/conf.d/devcluster01_nagios2.cfg 
  替换: 
   g/localhost/s//devcluster01/g 
   g/127.0.0.1/s//192.168.1.1/g 
5、cp  /etc/nagios3/conf.d/localhost_nagios2.cfg /etc/nagios3/conf.d/devcluster02_nagios2.cfg 
  vi /etc/nagios3/conf.d/devcluster02_nagios2.cfg 
  替换: 
   g/localhost/s//devcluster02/g 
   g/127.0.0.1/s//192.168.1.2/g 
6、cp  /etc/nagios3/conf.d/localhost_nagios2.cfg /etc/nagios3/conf.d/devcluster03_nagios2.cfg 
  vi /etc/nagios3/conf.d/devcluster03_nagios2.cfg 
  替换: 
   g/localhost/s//devcluster03/g 
   g/127.0.0.1/s//192.168.1.3/g 

7、vi /etc/nagios3/conf.d/services_nagios2.cfg 
  将hostgroup_name改为nagios-servers 
  增加: 
   # check that web services are running 
   define service { 
     hostgroup_name                  nagios-servers 
     service_description             HTTP 
     check_command                   check_http 
     use                             generic-service 
     notification_interval           0 ; set > 0 if you want to be renotified 
   } 

   # check that ssh services are running 
   define service { 
     hostgroup_name                  nagios-servers 
     service_description             SSH 
     check_command                   check_ssh 
     use                             generic-service 
     notification_interval           0 ; set > 0 if you want to be renotified 
   } 
8、vi /etc/nagios3/conf.d/extinfo_nagios2.cfg 
  将hostgroup_name改为nagios-servers 
  增加: 
   define hostextinfo{ 
     hostgroup_name   nagios-servers 
     notes            nagios-servers 
   #       notes_url        http://webserver.localhost.localdomain/hostinfo.pl?host=netware1 
     icon_image       base/debian.png 
     icon_image_alt   Debian GNU/Linux 
     vrml_image       debian.png 
     statusmap_image  base/debian.gd2 
     } 
9、sudo /etc/init.d/nagios3 restart 
10、访问http://devcluster01/nagios3/ 
  用户名:nagiosadmin密码:nagiosadmin 

监控端: 
1、apt-get install nagios-nrpe-server 
2、vi /etc/nagios/nrpe.cfg 
  替换: 
  g/127.0.0.1/s//192.168.1.1/g 
3、sudo /etc/init.d/nagios-nrpe-server restart 

十九、配置Splunk 
1、wget http://download.splunk.com/releases/5.0.2/splunk/linux/splunk-5.0.2-149561-Linux-x86_64.tgz 
2、tar -zxvf splunk-5.0.2-149561-Linux-x86_64.tgz 
3、cd splunk 
4、bin/splunk start --answer-yes --no-prompt --accept-license 
5、访问http://devcluster01:8000 
  用户名:admin 密码:changeme 
6、添加数据 -> 从 UDP 端口 -> UDP 端口 *: 1688 -> 来源类型 从列表 log4j -> 保存 
7、配置hadoop 
  vi /home/ysc/hadoop-1.1.1/conf/log4j.properties 
  修改: 
   log4j.rootLogger=${hadoop.root.logger}, EventCounter, SYSLOG 
  增加: 
   log4j.appender.SYSLOG=org.apache.log4j.net.SyslogAppender  
   log4j.appender.SYSLOG.facility=local1  
   log4j.appender.SYSLOG.layout=org.apache.log4j.PatternLayout  
   log4j.appender.SYSLOG.layout.ConversionPattern=%p %c{2}: %m%n  
   log4j.appender.SYSLOG.SyslogHost=host6:1688 
   log4j.appender.SYSLOG.threshold=INFO  
   log4j.appender.SYSLOG.Header=true 
   log4j.appender.SYSLOG.FacilityPrinting=true  
8、配置hbase 
  vi /home/ysc/hbase-0.92.2/conf/log4j.properties 
  修改: 
   log4j.rootLogger=${hbase.root.logger},SYSLOG 
  增加: 
   log4j.appender.SYSLOG=org.apache.log4j.net.SyslogAppender  
   log4j.appender.SYSLOG.facility=local1  
   log4j.appender.SYSLOG.layout=org.apache.log4j.PatternLayout  
   log4j.appender.SYSLOG.layout.ConversionPattern=%p %c{2}: %m%n  
   log4j.appender.SYSLOG.SyslogHost=host6:1688 
   log4j.appender.SYSLOG.threshold=INFO  
   log4j.appender.SYSLOG.Header=true 
   log4j.appender.SYSLOG.FacilityPrinting=true 
9、配置nutch 
  vi /home/lanke/ysc/nutch-2.1-hbase/conf/log4j.properties 
  修改: 
   log4j.rootLogger=INFO,DRFA,SYSLOG 
  增加: 
   log4j.appender.SYSLOG=org.apache.log4j.net.SyslogAppender  
   log4j.appender.SYSLOG.facility=local1  
   log4j.appender.SYSLOG.layout=org.apache.log4j.PatternLayout  
   log4j.appender.SYSLOG.layout.ConversionPattern=%p %c{2}: %m%n  
   log4j.appender.SYSLOG.SyslogHost=host6:1688 
   log4j.appender.SYSLOG.threshold=INFO  
   log4j.appender.SYSLOG.Header=true 
   log4j.appender.SYSLOG.FacilityPrinting=true 
10、启动hadoop和hbase 
  start-all.sh 
  start-hbase.sh 

二十、配置Pig 
1、wget http://labs.mop.com/apache-mirror/pig/pig-0.11.0/pig-0.11.0.tar.gz 
2、tar -xzvf pig-0.11.0.tar.gz 
3、cd pig-0.11.0 
4、vi /etc/profile 
  增加: 
  export PIG_HOME=/home/ysc/pig-0.11.0 
  export PATH=$PIG_HOME/bin:$PATH 
5、source /etc/profile 
6、cp conf/log4j.properties.template conf/log4j.properties 
7、vi conf/log4j.properties 
8、pig 

二十一、配置Hive 
1、wget http://mirrors.cnnic.cn/apache/hive/hive-0.10.0/hive-0.10.0.tar.gz 
2、tar -xzvf hive-0.10.0.tar.gz 
3、cd hive-0.10.0 
4、vi /etc/profile 
  增加: 
  export HIVE_HOME=/home/ysc/hive-0.10.0 
  export PATH=$HIVE_HOME/bin:$PATH 
5、source /etc/profile 
6、cp conf/hive-log4j.properties.template conf/hive-log4j.properties 
7、vi conf/hive-log4j.properties 
  替换: 
  log4j.appender.EventCounter=org.apache.hadoop.metrics.jvm.EventCounter 
  为: 
  log4j.appender.EventCounter=org.apache.hadoop.log.metrics.EventCounter 


二十二、配置Hadoop2.x集群 
1、wget http://labs.mop.com/apache-mirror/hadoop/common/hadoop-2.0.2-alpha/hadoop-2.0.2-alpha.tar.gz 
2、tar -xzvf hadoop-2.0.2-alpha.tar.gz 
3、cd hadoop-2.0.2-alpha 
4、vi etc/hadoop/hadoop-env.sh 
  追加: 
export JAVA_HOME=/home/ysc/jdk1.7.0_05 
  export HADOOP_HEAPSIZE=2000 
5、vi etc/hadoop/core-site.xml 
  <property> 
   <name>fs.defaultFS</name> 
   <value>hdfs://devcluster01:9000</value> 
   <description> 
      Where to find the Hadoop Filesystem through the network. 
      Note 9000 is not the default port. 
      (This is slightly changed from previous versions which didnt have "hdfs") 
   </description> 
   </property> 
   <property> 
    <name>io.file.buffer.size</name> 
    <value>131072</value> 
    <description>The size of buffer for use in sequence files. 
    The size of this buffer should probably be a multiple of hardware 
    page size (4096 on Intel x86), and it determines how much data is 
    buffered during read and write operations.</description> 
  </property> 
6、vi etc/hadoop/mapred-site.xml 
  <property> 
    <name>mapreduce.framework.name</name> 
    <value>yarn</value> 
  </property> 

  <property> 
    <name>mapred.job.reduce.input.buffer.percent</name> 
    <value>1</value> 
    <description>The percentage of memory- relative to the maximum heap size- to 
    retain map outputs during the reduce. When the shuffle is concluded, any 
    remaining map outputs in memory must consume less than this threshold before 
    the reduce can begin. 
    </description> 
  </property> 

  <property> 
    <name>mapred.job.shuffle.input.buffer.percent</name> 
    <value>1</value> 
    <description>The percentage of memory to be allocated from the maximum heap 
    size to storing map outputs during the shuffle. 
    </description> 
  </property> 

  <property> 
    <name>mapred.inmem.merge.threshold</name> 
    <value>0</value> 
    <description>The threshold, in terms of the number of files 
    for the in-memory merge process. When we accumulate threshold number of files 
    we initiate the in-memory merge and spill to disk. A value of 0 or less than 
    0 indicates we want to DON'T have any threshold and instead depend only on 
    the ramfs's memory consumption to trigger the merge. 
    </description> 
  </property> 

  <property> 
    <name>io.sort.factor</name> 
    <value>100</value> 
    <description>The number of streams to merge at once while sorting 
    files.  This determines the number of open file handles.</description> 
  </property> 

  <property> 
    <name>io.sort.mb</name> 
    <value>240</value> 
    <description>The total amount of buffer memory to use while sorting 
    files, in megabytes.  By default, gives each merge stream 1MB, which 
    should minimize seeks.</description> 
  </property> 
    <property> 
      <name>mapred.map.output.compression.codec</name> 
      <value>org.apache.hadoop.io.compress.SnappyCodec</value> 
      <description>If the map outputs are compressed, how should they be 
          compressed? 
      </description> 
    </property> 

    <property> 
      <name>mapred.output.compression.codec</name> 
      <value>org.apache.hadoop.io.compress.SnappyCodec</value> 
      <description>If the job outputs are compressed, how should they be compressed? 
      </description> 
    </property> 
  <property> 
    <name>mapred.output.compression.type</name> 
    <value>BLOCK</value> 
    <description>If the job outputs are to compressed as SequenceFiles, how should 
        they be compressed? Should be one of NONE, RECORD or BLOCK. 
    </description> 
  </property> 
  <property> 
    <name>mapred.child.java.opts</name> 
    <value>-Xmx2000m</value> 
  </property> 

  <property> 
    <name>mapred.output.compress</name> 
    <value>true</value> 
    <description>Should the job outputs be compressed? 
    </description> 
  </property> 

  <property> 
    <name>mapred.compress.map.output</name> 
    <value>true</value> 
    <description>Should the outputs of the maps be compressed before being 
        sent across the network. Uses SequenceFile compression. 
    </description> 
  </property> 

  <property> 
    <name>mapred.tasktracker.map.tasks.maximum</name> 
    <value>5</value> 
  </property> 

  <property> 
    <name>mapred.map.tasks</name> 
    <value>15</value> 
  </property> 

  <property> 
    <name>mapred.tasktracker.reduce.tasks.maximum</name> 
    <value>5</value> 
   <description> 
   define mapred.map tasks to be number of slave hosts.the best number is the  number of slave hosts plus the core numbers of per host 
   </description> 
  </property> 

  <property> 
    <name>mapred.reduce.tasks</name> 
    <value>15</value> 
    <description> 
   define mapred.reduce tasks to be number of slave hosts.the best number is the  number of slave hosts plus the core numbers of per host 
    </description> 
  </property> 
  <property> 
    <name>mapred.system.dir</name> 
    <value>/home/ysc/mapreduce/system</value> 
  </property> 

  <property> 
    <name>mapred.local.dir</name> 
    <value>/home/ysc/mapreduce/local</value> 
  </property> 

  <property> 
    <name>mapreduce.job.counters.max</name> 
    <value>12000</value> 
    <description>Limit on the number of counters allowed per job. 
    </description> 
  </property> 
7、vi etc/hadoop/yarn-site.xml 
  <property>    
    <name>yarn.resourcemanager.resource-tracker.address</name>   
    <value>devcluster01:8031</value> 
   </property>   
   <property>  
    <name>yarn.resourcemanager.address</name>     
    <value>devcluster01:8032</value>  
   </property> 
   <property>    
    <name>yarn.resourcemanager.scheduler.address</name>  
    <value>devcluster01:8030</value> 
   </property> 
   <property>  
    <name>yarn.resourcemanager.admin.address</name>  
    <value>devcluster01:8033</value>   
   </property>   
   <property>    
    <name>yarn.resourcemanager.webapp.address</name>    
    <value>devcluster01:8088</value>  
   </property>  
   <property>   
    <description>Classpath for typical applications.</description> 
    <name>yarn.application.classpath</name>  
    <value>       
    $HADOOP_CONF_DIR,      
    $HADOOP_COMMON_HOME/*,$HADOOP_COMMON_HOME/lib/*,    
    $HADOOP_HDFS_HOME/*,$HADOOP_HDFS_HOME/lib/*,       
    $HADOOP_MAPRED_HOME/*,$HADOOP_MAPRED_HOME/lib/*,   
    $YARN_HOME/*,$YARN_HOME/lib/*   
    </value>  
   </property> 
   <property>  
    <name>yarn.nodemanager.aux-services</name>  
    <value>mapreduce.shuffle</value>  
   </property>   
   <property>    
    <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>  
    <value>org.apache.hadoop.mapred.ShuffleHandler</value>  
   </property>  
   <property>   
    <name>yarn.nodemanager.local-dirs</name>     <value>/home/ysc/h2/data/1/yarn/local,/home/ysc/h2/data/2/yarn/local,/home/ysc/h2/data/3/yarn/local</value>  
   </property> 
   <property> 
    <name>yarn.nodemanager.log-dirs</name>      <value>/home/ysc/h2/data/1/yarn/logs,/home/ysc/h2/data/2/yarn/logs,/home/ysc/h2/data/3/yarn/logs</value>  
   </property>  
   <property>   
    <description>Where to aggregate logs</description> 
    <name>yarn.nodemanager.remote-app-log-dir</name>    
    <value>/home/ysc/h2/var/log/hadoop-yarn/apps</value> 
   </property>    
   <property>    
    <name>mapreduce.jobhistory.address</name>   
    <value>devcluster01:10020</value> 
   </property>   
   <property>    
    <name>mapreduce.jobhistory.webapp.address</name>   
    <value>devcluster01:19888</value> 
   </property>   
8、vi etc/hadoop/hdfs-site.xml 
  <property>  
   <name>dfs.permissions.superusergroup</name>  
   <value>root</value> 
  </property> 
  <property> 
    <name>dfs.name.dir</name> 
    <value>/home/ysc/dfs/filesystem/name</value> 
  </property> 
  <property> 
    <name>dfs.data.dir</name> 
    <value>/home/ysc/dfs/filesystem/data</value> 
  </property> 
  <property> 
    <name>dfs.replication</name> 
    <value>3</value> 
  </property> 
  <property> 
    <name>dfs.block.size</name> 
    <value>6710886400</value> 
    <description>The default block size for new files.</description> 
  </property> 
9、启动hadoop 
  bin/hdfs namenode -format 
  sbin/start-dfs.sh 
  sbin/start-yarn.sh 
10、访问管理页面 
  http://devcluster01:8088 
  http://devcluster01:50070

你可能感兴趣的:(hadoop,爬虫,hbase,Nutch,Sorl)