hue-3.9-cdh-5.7.0安装
hue-3.9-cdh-5.7.0安装
本次安装踩了一个大坑，由于之前先rpm方式安装了mysql，把centos7自带的mariadb删除了，用yum方式安装mysql-devel失败了，所以使用了rpm的方式安装hue。
1.下载安装包
下载地址：http://archive.cloudera.com/cdh5/redhat/5/x86_64/cdh/5.7.0/RPMS/x86_64/
#一共是15个rpm包，beeswax要放在最后一个安装，别问，问就是踩过坑
hue-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-beeswax-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-common-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-doc-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-hbase-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-impala-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-pig-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-plugins-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-rdbms-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-search-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-security-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-server-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-spark-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-sqoop-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
hue-zookeeper-3.9.0+cdh5.15.1+8420-1.cdh5.15.1.p0.4.el7.x86_64.rpm
执行rpm -ivh xxxxxx.rpm, 报freely redistributed under the terms of the GNU GPL错误
解决方法：
rpm -ivh xxx.rpm --nodeps --force
2.编译
#安装结束之后进入hue目录进行编译
[hadoop@hadoop001 bin]$ cd /usr/lib/hue/
[hadoop@hadoop001 hue]$ make apps
如果编译会提示缺失这两个文件libexslt.so.0，libxslt.so.1，目录在/use/lib64/，直接从其他机器上拷贝过来,或者是直接从网上下载。
3.集成大数据环境配置
hdfs-site.xml配置
<!-- 启用WebHdfs. -->
<property>
<name>dfs.webhdfs.enabled</name>
<value>true</value>
</property>
<property>
<name>dfs.permissions</name>
<value>false</value>
</property>
core-site.xml配置
<!-- Hue配置 -->
<property>
<name>hadoop.proxyuser.hadoop.hosts</name>
<value>*</value>
</property>
<property>
<name>hadoop.proxyuser.hadoop.groups</name>
<value>*</value>
</property>
yarn-site.xml配置
<!--打开HDFS上日志记录功能-->
<property>
<name>yarn.log-aggregation-enable</name>
<value>true</value>
</property>
<!--在HDFS上聚合的日志最长保留多少秒。3天-->
<property>
<name>yarn.log-aggregation.retain-seconds</name>
<value>259200</value>
</property>
</configuration>
httpfs-site.xml配置
<property>
<name>httpfs.proxyuser.hue.hosts</name>
<value>*</value>
</property>
<property>
<name>httpfs.proxyuser.hue.groups</name>
<value>*</value>
</property>
集成hive
#/etc/hue/conf/hue.ini配置文件改
[hadoop]
[[hdfs_clusters]]
[[[default]]]
fs_defaultfs=hdfs://hadoop001:8020
webhdfs_url=http://hadoop001:50070/webhdfs/v1
[[yarn_clusters]]
[[[default]]]
resourcemanager_host=hadoop001
resourcemanager_port=8032
submit_to=True
resourcemanager_api_url=http://hadoop001:8088
proxy_api_url=http://hadoop001:8088
history_server_api_url=http://hadoop001:19888
[beeswax]
hive_server_host=hadoop001
hive_server_port=10000
hive-site.xml配置
<property>
<name>hive.metastore.uris</name>
<value>thrift://192.168.9.128:9083</value>
<description>Thrift URI for the remote metastore. Used by metastore client to connect to remote metastore.</description>
</property>
<property>
<name>hive.server2.thrift.bind.host</name>
<value>192.168.9.128</value>
<description>Bind host on which to run the HiveServer2 Thrift service.</description>
</property>
</configuration>
如果是集群操作，配置要同步到其他机器上
4.hue环境配置
#环境变量
export HUE_HOME=/usr/lib/hue
export PATH=$HUE_HOME/build/env/bin:$PATH
#修改desktop的权限
[hadoop@hadoop001 desktop]$ pwd
/usr/lib/hue/desktop
[hadoop@hadoop desktop]$ chmod 777 hue/desktop/
[hadoop@hadoop desktop]$ chmod 766 hue/desktop/desktop.db
[hadoop@hadoop001 desktop]$ ll
total 4
lrwxrwxrwx. 1 hadoop hadoop 13 Apr 17 09:10 conf -> /etc/hue/conf
drwxr-xr-x. 4 hadoop hadoop 172 Apr 17 09:17 core
lrwxrwxrwx. 1 hadoop hadoop 23 Apr 17 09:10 desktop.db -> /var/lib/hue/desktop.db
lrwxrwxrwx. 1 hadoop hadoop 23 Apr 17 09:10 desktop.db.rpmsave.20190417.091101OURCE -> /var/lib/hue/desktop.db
drwxr-xr-x. 15 hadoop hadoop 210 Apr 17 09:11 libs
lrwxrwxrwx. 1 hadoop hadoop 12 Apr 17 09:11 logs -> /var/log/hue
-rw-r--r--. 1 hadoop hadoop 3467 Mar 23 2016 Makefile
#修改hue.ini文件
[hadoop@hadoop001 conf]$ pwd
/etc/hue/conf
# Webserver listens on this address and port
http_host=hadoop001
http_port=8888
#修改hue的数据库为mysql
[[database]]
engine=mysql
host=localhost
port=3306
user=root
password=123456
name=hue
# mysql, oracle, or postgresql configuration.
[[[mysql]]]
nice_name="My SQL DB"
name=hue
engine=mysql
host=localhost
port=3306
5.hue数据库初始化
[hadoop@hadoop001 bin]$ hue syncdb
[hadoop@hadoop001 bin]$ hue migrate
6.启动hue
#1、启动Hive metastore
[hadoop@hadoop001 bin]$ nohup hive --service metastore &
#2、启动hiveserver2
[hadoop@hadoop001 bin]$ nohup hive --service hiveserver2 &
#3、启动Hue
[hadoop@hadoop001 bin]$ nohup supervisor &
7.Hue Web界面查看
8.踩坑
编译的时候出现这个错误Error in sasl_client_start (-4) SASL(-4): no mechanism available: No worthy mechs found，安装一下软件
yum install cyrus-sasl-plain cyrus-sasl-devel cyrus-sasl-gssapi
出现 hue unable to open database file 错误
desktop以及desktop.db权限问题，参考上面的环境配置
hue界面出现 hue database is locked ，这个是没有修改mysql的配置数据的问题
修改hue.ini中配置，修改为mysql
hue界面出现 Could not connect to hadoop001:10000 错误
没有启动 hive --service metastore 或者 hive --service hiveserver2
使用hue查看hdfs系统报无法访问：/user/hadoop。 Note: you are a Hue admin but not a HDFS superuser, “hdfs” or part of HDFS supergroup, “supergroup”.
出现这个问题,是因为默认的超级用户是hdfs ,我的是hadoop用户启动的, 所以要修改hue.ini配置
# This should be the hadoop cluster admin default_hdfs_superuser=hadoop