Installing, configuring and running Hive

You can install a stable release of Hive by downloading and unpacking a tarball, or you can download the source code and build Hive using Maven (release 3.6.3 and later).

Prerequisites

  • Java 8.
  • Maven 3.6.3
  • Protobuf 2.5
  • Hadoop 3.3.6 (As a preparation, configure it in single-node cluster, pseudo-distributed mode)
  • Tez. The default is MapReduce but we will change the execution engine to Tez.
  • Hive is commonly used in production Linux environment. Mac is a commonly used development environment. The instructions in this document are applicable to Linux and Mac.  

Install the prerequisites

Java 8

Building Hive requires JDK 8 installed. Some notes in case you have ARM chipset (Apple M1 or later).

You will have to build protobuf 2.5 later. And it doesn't compile with ARM JDK. So we will install intel architecture's Java with brew and configure maven with this. It will enable us to compile protobuf. 

JDK install on apple ARM: 

brew install homebrew/cask-versions/adoptopenjdk8 --cask
brew untap adoptopenjdk/openjdk

Maven: 

Just install maven and configure the JAVA_HOME properly. 

Notes for arm: after a proper configuration, you should see something like this:

mvn -version
Apache Maven 3.6.3 (cecedd343002696d0abb50b32b541b8a6ba2883f)
Maven home: /Users/yourusername/programs/apache-maven-3.6.3
Java version: 1.8.0_292, vendor: AdoptOpenJDK, runtime: /Library/Java/JavaVirtualMachines/adoptopenjdk-8.jdk/Contents/Home/jre
Default locale: en_HU, platform encoding: UTF-8
OS name: "mac os x", version: "10.16", arch: "x86_64", family: "mac"

As you can see, even if it is an arm processor, maven thinks the architecture is Intel based.

Protobuf

You have to download and compile protobuf. And also, install it into the local maven repository. Protobuf 2.5.0 is not ready for ARM. On this chipset, you will need to do some extra steps.

wget https://github.com/google/protobuf/releases/download/v2.5.0/protobuf-2.5.0.tar.bz2
tar -xvf protobuf-2.5.0.tar.bz2
cd protobuf-2.5.0
./configure

On ARM, edit the src/google/protobuf/stubs/platform_macros.h and add arm to the part, processor architecture detection, after the last elif branch: 
 

#elif defined(__arm64__)
#define GOOGLE_PROTOBUF_ARCH_ARM 1
#define GOOGLE_PROTOBUF_ARCH_64_BIT 1

Now, you can compile and install protobuf:

make
make check
sudo make install

You can validate your install:

protoc --version

Hadoop

Firstly, move through the instructions on the official documentation, single-node, pseudo-distributed configuration: https://hadoop.apache.org/docs/stable/hadoop-project-dist/hadoop-common/SingleCluster.html#Pseudo-Distributed_Operation.

After that, set up HADOOP_HOME:

export HADOOP_HOME=/yourpathtohadoop/hadoop-3.3.6

Tez

Tez will require some additional steps. Hadoop uses a tez tarball but it expects it in other compressed directory structure than it is released. So we will extract the tarball and compress again. And also, we will put the extracted jars into hdfs. After that we set the necessary environment variables.

Download tez, extract and re-compress the tar: 

wget https://dlcdn.apache.org/tez/0.10.2/apache-tez-0.10.2-bin.tar.gz
tar -xzvf apache-tez-0.10.2-bin.tar.gz
cd apache-tez-0.10.2-bin
tar zcvf ../apache-tez-0.10.2-bin.tar.gz * && cd ..

Replace hdfs-client jar to 3.3.6 (tez is shipped with 3.3.1)

rm apache-tez-0.10.2-bin/lib/hadoop-hdfs-client-3.3.1.jar
cp $HADOOP_HOME/share/hadoop/hdfs/hadoop-hdfs-client-3.3.6.jar $TEZ_HOME/lib

Add the necessary tez files to hdfs

$HADOOP_HOME/sbin/start-dfs.sh	# start hdfs
$HADOOP_HOME/bin/hadoop fs -mkdir -p /apps/tez
$HADOOP_HOME/bin/hadoop fs -put apache-tez-0.10.2-bin.tar.gz /apps/tez # copy the tarball
$HADOOP_HOME/bin/hadoop fs -put apache-tez-0.10.2-bin /apps/tez # copy the whole folder
$HADOOP_HOME/bin/hadoop fs -ls /apps/tez # verify
$HADOOP_HOME/sbin/stop-all.sh # stop hdfs

Set up TEZ_HOME and HADOOP_CLASSPATH environment variables

export TEZ_HOME=/yourpathtotez/apache-tez-0.10.2-bin
export HADOOP_CLASSPATH=$TEZ_HOME/*:$TEZ_HOME/conf

Create a new config file for Tez: $TEZ_HOME/conf/tez-site.xml

<configuration>
    <property>
        <name>tez.lib.uris</name>
        <value>hdfs://localhost:9000/apps/tez/apache-tez-0.10.2-bin.tar.gz,hdfs://localhost:9000/apps/tez/apache-tez-0.10.2-bin/lib,hdfs://localhost:9000/apps/tez/apache-tez-0.10.2-bin</value>
    </property>
</configuration>

Extra hadoop configurations to make everything working

Modify $HADOOP_HOME/etc/hadoop/core-site.xml

<configuration>
    <property>
        <name>fs.defaultFS</name>
        <value>hdfs://localhost:9000</value>
    </property>

    <property>
      <name>hadoop.proxyuser.yourusername.groups</name>
      <value>*</value>
    </property>

    <property>
      <name>hadoop.proxyuser.yourusername.hosts</name>
      <value>*</value>
    </property>
</configuration>

Modify $HADOOP_HOME/etc/hadoop/hadoop-env.sh

# JAVA_HOME
export JAVA_HOME=/yourpathtojavahome/javahome
# tez
export TEZ_CONF_DIR=/yourpathtotezconf/conf
export TEZ_JARS=/yourpathtotez/apache-tez-0.10.2-bin
export HADOOP_CLASSPATH=${TEZ_CONF_DIR}:${TEZ_JARS}/*:${TEZ_JARS}/lib/*:${HADOOP_CLASSPATH}:
${JAVA_JDBC_LIBS}:${MAPREDUCE_LIBS}

Modify $HADOOP_HOME/etc/hadoop/mapred-site.xml

<configuration>
    <property>
        <name>mapreduce.framework.name</name>
        <value>yarn</value>
    </property>
    <property>
        <name>mapreduce.application.classpath</name>
        <value>$HADOOP_CLASSPATH:$HADOOP_HOME/share/hadoop/mapreduce/*:$HADOOP_MAPRED_HOME/share/hadoop/mapreduce/lib/*</value>
    </property>
</configuration>

Modify $HADOOP_HOME/etc/hadoop/yarn-site.xml

<configuration>
<!-- Site specific YARN configuration properties -->
  <property>
    <name>yarn.nodemanager.aux-services</name>
    <value>mapreduce_shuffle</value>
  </property>
  <property>
    <name>yarn.nodemanager.env-whitelist</name>
    <value>JAVA_HOME,HADOOP_COMMON_HOME,HADOOP_HDFS_HOME,HADOOP_CONF_DIR,CLASSPATH_PREPEND_DISTCACHE,HADOOP_YARN_HOME,HADOOP_HOME,PATH,LANG,TZ,HADOOP_MAPRED_HOME</value>
  </property>
    <property>
    <name>yarn.nodemanager.resource.memory-mb</name>
  <value>4096</value>
  </property>
  <property>
    <name>yarn.scheduler.minimum-allocation-mb</name>
    <value>2048</value>
  </property>
  <property>
    <name>yarn.nodemanager.vmem-pmem-ratio</name>
    <value>2.1</value>
  </property>
</configuration>

Installing Hive from a Tarball

Start by downloading the most recent stable release of Hive from one of the Apache download mirrors (see Hive Releases).

Next you need to unpack the tarball. This will result in the creation of a subdirectory named hive-x.y.z (where x.y.z is the release number):

wget https://dlcdn.apache.org/hive/hive-4.0.0-beta-1/apache-hive-4.0.0-beta-1-bin.tar.gz
tar -xzvf apache-hive-4.0.0-beta-1-bin.tar.gz

Set the environment variable HIVE_HOME to point to the installation directory:

cd apache-hive-4.0.0-beta-1-bin
export HIVE_HOME=/yourpathtohive/apache-hive-4.0.0-beta-1-bin

Add $HIVE_HOME/bin to your PATH:

export PATH=$HIVE_HOME/bin:$PATH

Create a directory for external tables: 

mkdir /yourpathtoexternaltables/warehouse

Create a new config file for Hive: $HIVE_HOME/conf/hive-site.xml

<configuration>
    <property>
        <name>hive.tez.container.size</name>
        <value>1024</value>
    </property>

    <property>
        <name>hive.metastore.warehouse.external.dir</name>
        <value>/yourpathtowarehousedirectory/warehouse</value>
    </property>

    <property>
        <name>hive.execution.engine</name>
        <value>tez</value>
    </property>

    <property>
        <name>tez.lib.uris</name>
        <value>hdfs://localhost:9000/apps/tez/apache-tez-0.10.2-bin.tar.gz,hdfs://localhost:9000/apps/tez/apache-tez-0.10.2-bin/lib,hdfs://localhost:9000/apps/tez/apache-tez-0.10.2-bin</value>
    </property>

    <property>
        <name>tez.configuration</name>
        <value>/yourpathtotez/apache-tez-0.10.2-bin/conf/tez-site.xml</value>
    </property>

    <property>
        <name>tez.use.cluster.hadoop-libs</name>
        <value>true</value>
    </property>
</configuration>

Initialize metastore schema. It will create a directore called metastore_db. It contains an embedded Derby database for metastore

$HIVE_HOME/bin/schematool -dbType derby -initSchema --verbose

Run HiveServer2

$HIVE_HOME/bin/hiveserver2

Run beeline:

bin/beeline -u 'jdbc:hive2://localhost:10000/' -n yourusername

As a test, create a table insert some value

create table test (message string);
insert into test values ('Hello, from Hive!');

Installing from Source Code 

Configuring is the same as when we do it from tarball. The only difference is that we have to build Hive for ourself and we will find the compiled binaries in a different directory.


Hive is available via Git at https://github.com/apache/hive. You can download it by running the following command.

git clone git@github.com:apache/hive.git

In case you want to get a specific release branch, like 4.0.0, you can run that command: 

git clone -b branch-4.0 --single-branch git@github.com:apache/hive.git


To build Hive, execute the following command on the base directory:

  $ mvn clean install -Pdist,itests,iceberg -DskipTests

It will create the subdirectory packaging/target/apache-hive-<release_string>-bin/apache-hive-<release_string>-bin/. That will be your HIVE_HOME directory.

It has a content like:

  • bin/: directory containing all the shell scripts
  • lib/: directory containing all required jar files
  • conf/: directory with configuration files
  • examples/: directory with sample input and query files

That directory should contain all the files necessary to run Hive. You can run it from there or copy it to a different location, if you prefer.

From now, you can follow the steps described in the section Installing Hive from a Tarball

Next Steps

You can begin using Hive as soon as it is installed, it should be work on you computer. There are some extra information in the following sections.

Beeline CLI

HiveServer2 has a CLI called Beeline (see Beeline – New Command Line Shell). To use Beeline, execute the following command in the Hive home directory:

$ bin/beeline

Hive Metastore

Metadata is stored in a relational database. In our example (and as a default) it is a Derby database. By default, it's location is ./metastore_db. (See conf/hive-default.xml). You can change it by modifying the configuration variable javax.jdo.option.ConnectionURL.

Using Derby in embedded mode allows at most one user at a time. To configure Derby to run in server mode, see Hive Using Derby in Server Mode.

To configure a database other than Derby for the Hive metastore, see Hive Metastore Administration.

Next Step: Configuring Hive.

HCatalog and WebHCat

HCatalog

If you install Hive from the binary tarball, the hcat command is available in the hcatalog/bin directory. However, most hcat commands can be issued as hive commands except for "hcat -g" and "hcat -p". Note that the hcat command uses the -p flag for permissions but hive uses it to specify a port number. The HCatalog CLI is documented here and the Hive CLI is documented here.

HCatalog installation is documented here.

WebHCat (Templeton)

If you install Hive from the binary tarball, the WebHCat server command webhcat_server.sh is in the hcatalog/webhcat/svr/src/main/bin/webhcat_server.sh directory.

WebHCat installation is documented here.

  • No labels