Go to

Monday, September 10, 2018

Big Data: Installing HUE WebUI ontop of Hadoop and Hive

APACHE HUE: Apache hue is a web UI for Hadoop ecosystem, certified by apache. Hue provide connection to all of Apache Hadoop ecosystem tools like HDFS, hive, pig, impala, spark and comes with interactive web UI interface.






Hue depends on these following packages:

gcc,    g++ ,    libxml2-dev ,    libxlst-dev,     libsasl2-dev ,    libsasl2-modules-gssapi-mit ,   
libmysqlclient-dev,     python-dev ,    python-setuptools,     libsqlite3-dev,     ant ,   
libkrb5-dev,     libtidy-0.99-0,     libldap2-dev,     libssl-dev,     libgmp3-dev .   


Installing all the packages: To install all of these packages follow below use commands which are giving below.


sudo apt-get update
sudo apt-get install gcc g++ libxml2-dev libxslt-dev libsasl2-dev libsasl2-modules-gssapi-mit libmysqlclient-dev python-dev python-setuptools libsqlite3-dev ant libkrb5-dev libtidy-0.99-0 libldap2-dev libssl-dev libgmp3-dev


Installation and Configuration
Performing installation as hadoop user( if you have specific hadoop use as hadoop admin).
su - hduser   
Download Hue from gethue.com (this link is an example     obtained from Hue website)   
wget     https://dl.dropboxusercontent.com/u/730827/hue/releases/4.1.0/hue-4.1.0.tgz

Extract the downloaded tarball

tar -xvf hue-4.1.0.tgz

Execute install command

    cd hue-4.1.0
    make install   
Note: if you receive an error like c/_cffi_backend.c:15:17: fatal error: ffi.h: No such file or directory hue error


Run below command to resolve it:
    sudo apt-get update
    sudo apt-get install libffi-dev


Once the above process is completed,

Update ~/.bashrc file,

export HUE_HOME=/home/hadoop/hue
    export PATH=$PATH:$HUE_HOME/build/env/bin


source after adding the entries,
    source ~/.bashrc



Configure Hue ( 3 files to edit)

  • make this changes in hue.ini file
    open this file.
    cd $HUE_HOME/desktop/conf

[desktop]
    server_user=hduser
    server_group=hduser
    default_user=hduser
    default_hdfs_superuser=hduser


 app_blacklist=impala,security
 fs_defaultfs=hdfs://localhost:50070
 hadoop_conf_dir=$HADOOP_CONF_DIR
 resourcemanager_host=localhost
 resourcemanager_api_url=http://localhost:8088

 hive_server_host=http://localhost:10000
 hive_conf_dir=/usr/local/hive2.2/conf


 proxy_api_url=http://localhost:8088
 history_server_api_url=http://localhost:19888
 hbase_clusters=(Cluster|localhost:9090)
 oozie_url=http://localhost:11000/oozie


Make change in Hadoop Conf files:
open conf directory and edit below given files.
cd $HADOOP_CONF_DIR

  • core-site.xml
    Note: hduser = your hadoop  useper huser


<property>
   <name>hadoop.proxyuser.hduser.hosts</name>
   <value>*</value>
</property>
<property>
   <name>hadoop.proxyuser.hduser.groups</name>
   <value>*</value>
</property>



  • hdfs-site.xml
    Note: enabling webUI for Hadoop so that Hue can get access.
<property>
   <name>dfs.webhdfs.enabled</name>
   <value>true</value>
</property>

  • httpfs-site.xml
    Note: Here we are giving HDFS access to our Hue User, which we has created earlier.
<property>
 <name>httpfs.proxyuser.hduser.hosts</name>
 <value>*</value>
 <description> this properties are for hue to access hdfs </description>
</property>


<property>
 <name>httpfs.proxyuser.hduser.groups</name>
 <value>*</value>
  <description> this properties are for hue to access hdfs </description>
</property>

Starting Hadoop and hiveserver2 for hue:
Start Hadoop:
start-all.sh
  Start Hive server:
$HIVE_HOME/bin/hiveserver2 


Start Hue 

nohup     supervisor    
or
$HUE_HOME/build/env/bin/hue     runserver

Login to Hue Web Interface: http://localhost:8888


Reference link for help:

No comments:

Post a Comment

Power BI Report and Dataset Performance Optimization

  Power BI Report and Dataset Performance Optimization     For any organization developing Power BI reports, there is a strong desire to des...