Local deployment Zeppelin 0.10.1

twitter

• Apache Zeppelin It is a web-based tool that makes interactive data analysis feasible notebook

• Zeppelin Provides a framework for data visualization

• To achieve what you need

  • Data collection
  • Data discovery
  • Data analysis
  • Data visualization and collaboration

The latest version is 0.10.1 edition , There are two official installation packages , One is that it contains all parsers （1.6G）, The other is with spark And network download parser package （568M）

Supported parsers

• alluxio
• angular
• beam
• bigquery
• cassandra
• elasticsearch
• file
• flink
• flink-cmd
• geode
• groovy
• hazelcastjet
• hbase
• ignite
• influxdb
• java
• jdbc
• jupyter
• kotlin
• ksql
• kylin
• lens
• livy
• md
• mongodb
• neo4j
• pig
• python
• r
• sap
• scalding
• scio
• sh
• spark
• spark-submit
• sparql
• submarine

Deploy zeppelin

Download the full parser installation package

wget --no-check-certificate -c \
https://dlcdn.apache.org/zeppelin/zeppelin-0.10.1/zeppelin-0.10.1-bin-all.tgz
tar xf zeppelin-0.10.1-bin-all.tgz

install java Environmental Science

zeppelin Deployment dependency java Environmental Science

yum install -y java-1.8.0-openjdk

Modify the configuration file

cd zeppelin-0.10.1-bin-all
cp conf/zeppelin-env.sh.template conf/zeppelin-env.sh
vim conf/zeppelin-env.sh

The official profile template is fully annotated , Here are some parameters that need attention

#  Definition  java  Command path ,yum  You can install without writing , Download the  Oracle java  I need to configure it 
export JAVA_HOME=
# zeppelin  Access address , If you don't configure , The default value is  127.0.0.1, Can also write  0.0.0.0
export ZEPPELIN_ADDR=
# zeppelin web  Page access port , The default value is  8080, If it conflicts with the local port , You can modify the parameters here 
export ZEPPELIN_PORT=
# sh  Parser bound machine  ip, Only one... Can be written , If you don't write , One will be randomly obtained on this machine  ip  Address  [ Multiple network cards or  docker  It's embarrassing ]
export ZEPPELIN_LOCAL_IP=
# zeppelin  Log directory , The default is the current path 
export ZEPPELIN_LOG_DIR=
# zeppelin pid  Record Directory , Default record in  ${ZEPPELIN_HOME}/run  Under the table of contents 
export ZEPPELIN_PID_DIR=
# notebook  Storage path 
export ZEPPELIN_NOTEBOOK_DIR=

modify xml The configuration file

cp conf/zeppelin-site.xml.template conf/zeppelin-site.xml
vim conf/zeppelin-site.xml

No special needs , You can not configure

• zeppelin The default index path is /tmp Under the table of contents , If necessary, you can modify it to your own defined path , Do unified management

<property>
  <name>zeppelin.search.index.path</name>
  <value>/tmp/zeppelin-index</value>
  <description>path for storing search index on disk.</description>
</property>

• Turn off anonymous access mode [ The default is anonymous login , also zeppelin-site.xml There is no such paragraph in the document , You need to increase it yourself ] （ Be careful xml Format ）
  • false - Indicates that anonymous mode is turned off
  • true - Indicates that anonymous mode is enabled

<property>
  <name>zeppelin.anonymous.allowed</name>
  <value>false</value>
  <description>Anonymous user allowed by default</description>
</property>

• When anonymous mode is turned off , Need to configure user name and password

cp conf/shiro.ini.template conf/shiro.ini
vim conf/shiro.ini

To configure < user name > = < password >,< role >

More details can be found in official documents ：Apache Shiro Configuration

admin = admin,admin

start-up zeppelin

bash bin/zeppelin-daemon.sh start

Access the configured through the browser ip Add ports to access zeppelin