安装手册新版.docx

HUE 安装手册

        本文档重要参考cloudera和hortonworks的安装文档，但这两个文档重要是针对自己产品的安装，有些配置是产品特有的配置，特别是cloudera的安装文档，hortonwork的文档不错，本手册重要是参考hortonwork的安装文档而来，并增长了thrift的插件配置和安装。

以上是hue和hadoop组件的整体结构图：

1. 安装环节：
安装前，最佳先阅读下以上提到的两篇安装手册，本手册是安装中软国际的安装包中安装的hue，也可以把yum源地址配置成hortonworks的安装包，这里的说明的yum安装源是中软国际的安装包。
    主节点：master.hadoop
    jobtracker节点：node4.hadoop
    本手册是把hue安装在主节点master.hadoop上
2. 配置hadoop（通过ambari界面更改如下配置文献）
1.  hdfs-site.xml
<property>
  <name>dfs.webhdfs.enabled</name>
  <value>true</value>
</property>
2. core-site.xml
<property>
  <name>hadoop.proxyuser.hue.hosts</name>
  <value>*</value>
</property>
<property>
  <name>hadoop.proxyuser.hue.groups</name>
  <value>*</value>
</property>
3. webhcat-site.xml 
<property>
  <name>webhcat.proxyuser.hue.hosts</name>  
  <value>*</value>
</property>
<property>  
  <name>webhcat.proxyuser.hue.groups</name> 
  <value>*</value>
</property>
4. oozie-site.xml
<property>
  <name>oozie.service.ProxyUserService.proxyuser.hue.hosts</name>
  <value>*</value>
</property>
<property>  
  <name>oozie.service.ProxyUserService.proxyuser.hue.groups</name>
  <value>*</value>
</property>
5. 配置thrift插件
     1.假如安装hue server和JobTracker在同一个节点上,不需要复制插件包，该包已经在lib下了；
  2.假如安装hue server和JobTracker不在同一个节点上复master.hadoop的 /usr/lib/hadoop/lib/hue-plugins-2.2.0-SNAPSHOT.jar 至jobtracker节点node4.hadoop的 /usr/lib/hadoop/lib下

scp  /usr/lib/hadoop/lib   p: /usr/lib/hadoop/lib

假如没有该包，yum install hue-plugins
  3 .管理界面配置插件，修改mapred-site.xml
 <!-- Enable Hue plugins -->
<property>
         <name>mapreduce.jobtracker.plugins</name><!--HUE文档，规定设立mapred.jobtracker.plugins,但设立后不生效，通过查询源码发现实际是mapreduce.jobtracker.plugins-->
        <value>org.apache.hadoop.thriftfs.ThriftJobTrackerPlugin</value>
</property>
<property>
       <name>jobtracker.thrift.address</name>
       <value>node4.hadoop:9290</value>
</property>
配置以上信息后，重启服务。        
3.  安装 Hue
           yum install hue
4.  配置 Hue
     需要vi  /etc/hue/conf.empty/hue.ini，根据当前集群的信息，配置如下信息：
  5.1 配置Web Server
1. Hue HTTP Address.
# Webserver listens on this address and port
http_host=0.0.0.0
http_port=8000
2. 指定Secret Key.
输入长度在30-60的字符串：值随便定义
secret_key=secretkeysecretkeysecretkeysecretkey
3. 配置Hue for SSL.(一方面要安装openssl ,默认已经安装)，配置后，访问hue需要使用https://
Install pyOpenSSL in order to configure Hue to serve over HTTPS. To install pyOpenSSL, from the root of your Hue installation path, complete the following instructions:
生成key
$cd /etc/hue/
### Create a key
        $openssl genrsa 1024 > host.key
### Create a self-signed certificate
        $openssl req -new -x509 -nodes -sha1 -key host.key > host.cert
把生成的的两个文献信息配置到如下属性：
ssl_certificate=/etc/hue/host.cert
ssl_private_key=/etc/hue/host.key

  5.2. Configure Hadoop
1. 配置HDFS Cluster.
 # Configuration for HDFS NameNode
  # ------------------------------------------------------------------------
  [[hdfs_clusters]]

    [[[default]]]
      # Enter the filesystem uri
      fs_defaultfs=hdfs://master.hadoop:8020

      # Use WebHdfs/HttpFs as the communication mechanism. To fallback to
      # using the Thrift plugin (used in Hue 1.x), this must be uncommented
      # and explicitly set to the empty value.
      webhdfs_url=:50070/webhdfs/v1/

      ## security_enabled=true

      # Settings about this HDFS cluster. If you install HDFS in a
      # different location, you need to set the following.

      # Defaults to $HADOOP_HDFS_HOME or /usr/lib/hadoop-hdfs
      hadoop_hdfs_home=/usr/lib/hadoop/lib

      # Defaults to $HADOOP_BIN or /usr/bin/hadoop
       hadoop_bin=/usr/bin/hadoop

      # Defaults to $HADOOP_CONF_DIR or /etc/hadoop/conf
      hadoop_conf_dir=/etc/hadoop/conf

2. Configure YARN (MR2) Cluster.
# Configuration for MapReduce JobTracker
  # ------------------------------------------------------------------------
  [[mapred_clusters]]

    [[[default]]]
      # Enter the host on which you are running the Hadoop JobTracker
      jobtracker_host=node4.hadoop
      # The port where the JobTracker IPC listens on
      jobtracker_port=9001
      # Thrift plug-in port for the JobTracker
      thrift_port=9290
      # Whether to submit jobs to this cluster
      submit_to=true

      # Job tracker kerberos principal
      ## jt_kerberos_principal=jt

      ## security_enabled=true

      # Settings about this MR1 cluster. If you install MR1 in a
      # different location, you need to set the following.

      # Defaults to $HADOOP_MR1_HOME or /usr/lib/hadoop-0.20-mapreduce
      hadoop_mapred_home=/usr/lib/hadoop/lib

      # Defaults to $HADOOP_BIN or /usr/bin/hadoop
       hadoop_bin=/usr/bin/hadoop

      # Defaults to $HADOOP_CONF_DIR or /etc/hadoop/conf
      hadoop_conf_dir=/etc/hadoop/conf
4.      3. 配置Beeswax
   beeswax_server_host=master.hadoop
   beeswax_server_port=8002
   beeswax_meta_server_host=localhost
   beeswax_meta_server_port=8003
5.      4. Configure JobDesigner and Oozie
 # The URL where the Oozie service runs on. This is required in order for
  # users to submit jobs.
  oozie_url=:11000/oozie    （master.hadoop为oozie安装节点）
  ## security_enabled=true
  # Location on HDFS where the workflows/coordinator are deployed when submitted.
  ## remote_deployement_dir=/user/hue/oozie/deployments
6.         6. Configure WebHCat
             templeton_url=":50111/templeton/v1/"   （master.hadoop为WebHcat的安装节点）

5.  启动Hue
execute the following command on the Hue Server:
     /etc/init.d/hue start
    hue服务启动后，访问hue界面：
    :8000/
    用户名：hue   密码:hue
6. 假如要把hue的内置数据库SQLLite换成mysql，参考文档：
    cloudrea文档：Hue Installtion --->Administering HUe 文档