Source code

git clone https://github.com/apache/kylin.git -b kylin-on-parquet-v2
# Compile 
mvn clean install -DskipTests

Environment on the dev machine

Install Maven

The latest maven can be found at http://maven.apache.org/download.cgi, we create a symbolic so that mvn can be run anywhere.

cd ~
wget http://xenia.sote.hu/ftp/mirrors/www.apache.org/maven/maven-3/3.2.5/binaries/apache-maven-3.2.5-bin.tar.gz
tar -xzvf apache-maven-3.2.5-bin.tar.gz
ln -s /root/apache-maven-3.2.5/bin/mvn /usr/bin/mvn

Install Spark

Manually install the Spark binary in in a local folder like /usr/local/spark. Kylin is not support community version currently. Download spark with the following:

# spark version is spark-2.4.1-os-kylin-r3
wget https://download-resource.s3.cn-north-1.amazonaws.com.cn/osspark/spark-2.4.1-os-kylin-r3.tgz

How to Debug

There are two modes to debug source code : Debug with local metadata(recommend) and debug with hadoop sandbox.

Configuration

Debug with local metadata 

1. Edit the properties of $KYLIN_SOURCE_DIR/examples/test_case_data/sandbox/kylin.properties

# Need to use absolute path
kylin.metadata.url=${KYLIN_SOURCE_DIR}/examples/test_case_data/sample_local
kylin.storage.url=${KYLIN_SOURCE_DIR}/examples/test_case_data/sample_local
kylin.env.zookeeper-is-local=true
kylin.env.hdfs-working-dir=file://$KYLIN_SOURCE_DIR/examples/test_case_data/sample_local
kylin.engine.spark-conf.spark.master=local
kylin.engine.spark-conf.spark.eventLog.dir=/path/to/local/dir
kylin.engine.spark-conf.spark.sql.shuffle.partitions=1
kylin.env=LOCAL

     2. Open DebugTomcat.java, start to debug. 

     3. Edit Configuration

   VM options "-Dspark.local=true"   is for query engine.

Debug with Hadoop sandbox

Local configuration must be modified to point to your hadoop sandbox (or CLI) machine.

• In examples/test_case_data/sandbox/kylin.properties
  • Find sandbox and replace with your hadoop hosts (if you’re using HDP sandbox, this can be skipped)
  • Find kylin.job.use-remote-cli and change it to “true” (in code repository the default is false, which assume running it on hadoop CLI)
  • Find kylin.job.remote.cli.username and kylin.job.remote.cli.password, fill in the user name and password used to login hadoop cluster for hadoop command execution; If you’re using HDP sandbox, the default username is root and password is hadoop.
• In examples/test_case_data/sandbox
  • For each configuration xml file, find all occurrences of sandbox and sandbox.hortonworks.com, replace with your hadoop hosts; (if you’re using HDP sandbox, this can be skipped)

An alternative to the host replacement is updating your hosts file to resolve sandbox and sandbox.hortonworks.com to the IP of your sandbox machine.

Launch Kylin Web Server

Copy server/src/main/webapp/WEB-INF to webapp/app/WEB-INF

cp -r server/src/main/webapp/WEB-INF webapp/app/WEB-INF

Download JS for Kylin web GUI. npm is part of Node.js, please search about how to install it on your OS.

cd webapp
npm install -g bower
bower --allow-root install

If you encounter network problem when run “bower install”, you may try:

git config --global url."git://".insteadOf https://

Note, if on Windows, after install bower, need to add the path of “bower.cmd” to system environment variable ‘PATH’, and then run:

bower.cmd --allow-root install

In IDE, launch org.apache.kylin.rest.DebugTomcat. Please set the path of “server” module as the “Working directory”, set “kylin-server” for “Use classpath of module”, and check “Include dependencies with ‘Provided’ scope” option in IntelliJ IDEA 2018. If you’re using IntelliJ IDEA 2017 and older, you need modify “server/kylin-server.iml” file, replace all “PROVIDED” to “COMPILE”, otherwise an “java.lang.NoClassDefFoundError: org/apache/catalina/LifecycleListener” error may be thrown..

You may also need to tune the VM options:

-Dhdp.version=2.4.0.0-169 -DSPARK_HOME=/usr/local/spark -Dkylin.hadoop.conf.dir=/workspace/kylin/examples/test_case_data/sandbox -Xms800m -Xmx800m -XX:PermSize=64M -XX:MaxNewSize=256m -XX:MaxPermSize=128m 

Also remeber that if you debug with local mode, you should add VM option for query engine:

-Dspark.local=true

If you worked with Kerberized Hadoop Cluster, the additional VM options should be set:

-Djava.security.krb5.conf=/etc/krb5.conf -Djava.security.krb5.principal=kylin -Djava.security.krb5.keytab=/path/to/kylin/keytab

And Hadoop environment variable:

HADOOP_USER_NAME=root

By default Kylin server will listen on 7070 port; If you want to use another port, please specify it as a parameter when run DebugTomcat.

Check Kylin Web at http://localhost:7070/kylin (user:ADMIN, password:KYLIN)

Setup IDE code formatter

In case you’re writting code for Kylin, you should make sure that your code in expected formats.

For Eclipse users, just format the code before committing the code.

For intellij IDEA users, you have to do a few more steps:

1. Install “Eclipse Code Formatter” and use “org.eclipse.jdt.core.prefs” and “org.eclipse.jdt.ui.prefs” in core-common/.settings to configure “Eclipse Java Formatter config file” and “Import order”

2. Go to Preference => Code Style => Java, set “Scheme” to Default, and set both “Class count to use import with ‘*’” and “Names count to use static import with ‘*’” to 99.

3. Disable intellij IDEA’s “Optimize imports on the fly”

4. Format the code before committing the code.

Setup IDE license header template

Each source file should include the following Apache License header

Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements.  See the NOTICE file
distributed with this work for additional information
regarding copyright ownership.  The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License.  You may obtain a copy of the License at

     http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

The checkstyle plugin will check the header rule when packaging also. The license file locates under dev-support/checkstyle-apache-header.txt. To make it easy for developers, please add the header as Copyright Profile and set it as default for Kylin project.

How to Package and Deploy

cd ${KYLIN_SOURCE_CODE}
# For HDP2.x
./build/script/package.sh

# For CDH5.7
./build/script/package.sh -P cdh5.7
# After finished, the package will be avaliable in the directory ${KYLIN_SOURCE_CODE}/dist/

# If running on HDP, you need to uncomment the following properties in kylin.properties
kylin.engine.spark-conf.spark.driver.extraJavaOptions=-Dhdp.version=current
kylin.engine.spark-conf.spark.yarn.am.extraJavaOptions=-Dhdp.version=current
kylin.engine.spark-conf.spark.executor.extraJavaOptions=-Dhdp.version=current