hive容器镜像,本分支为元数据使用mysql的版本

本项目涉及到1个镜像的制作，使用hive2.3.9版本

hive-server和hive-metastore服务共同使用的hive:2.3.9-mysql-metastore镜像

本项目可直接快速启动一个包含hadoop及hive的容器环境,具体参考docker-compose.yml

#除hive-server和hive-metastore服务使用本项目中的Dockerfile构建外，其他服务具体请看其他镜像部分的说明

#启动脚本
git clone http://gitlab.software.dc/mp-data/dss/docker-hive.git
cd docker-hive
git checkout 2.3.9-mysql-metastore
docker-compose up -d
docker-compose logs -f hive-server

#测试HDFS服务是否正常
docker-compose exec hive-server /bin/bash -c 'hdfs dfs -ls /'

#测试Hive服务是否正常
docker-compose exec hive-server /bin/bash -c 'hive -S -e "show databases;"'
docker-compose exec hive-server /bin/bash -c 'hive -S -e "select * from default.pokes;"'

端口开放列表

节点	用途	容器端口	主机端口	说明
namenode	fs.defaultFS	8020	8020
namenode	dfs.namenode.http.address	50070	50070
datanode	dfs.datanode.address	50010	50010
datanode	dfs.datanode.ipc.address	50020	50020
datanode	dfs.datanode.http.address	50075	50075
nodemanager	yarn.nodemanager.localizer.address	8040	8040
nodemanager	yarn.nodemanager.address	8041	8041
nodemanager	yarn.nodemanager.webapp.address	8042	8042
resourcemanager	yarn.resourcemanager.scheduler.address	8030	8030
resourcemanager	yarn.resourcemanager.resource-tracker.address	8031	8031
resourcemanager	yarn.resourcemanager.address	8032	8032
resourcemanager	yarn.resourcemanager.webapp.address	8088	8088
historyserver	mapreduce.jobhistory.webapp.address	19888	19888
historyserver	mapreduce.jobhistory.address	10020	10020
historyserver	yarn.timeline-service.webapp.address	8188	8188	timelineseerver端口和historyserver共用容器
historyserver	yarn.timeline-service.address	10200	10200	timelineseerver端口和historyserver共用容器
hive-server	hive.server2.thrift.port	10000	10000
hive-server	hive.server2.webui.port	10002	10002	hive2.0以上版本支持web UI
hive-metastore-mysql	mysql	3306	13306	数据库端口
hive-metastore	metastore	9083	9083	元数据服务端口

hive:2.3.9-mysql-metastore镜像

此镜像Dockerfile中包括hive2.3.9版本的安装、hive启动脚本及atlas hive hook插件的安装，其中atlas插件根据环境变量参数选择是否启用，具体参考entrypoint.sh

镜像制作脚本

git clone http://gitlab.software.dc/mp-data/dss/docker-hive.git
cd docker-hive
git checkout 2.3.9-mysql-metastore
docker build -t  harbor.software.dc/mpdata/hive:2.3.9-mysql-metastore -f Dockerfile ./
docker push harbor.software.dc/mpdata/hive:2.3.9-mysql-metastore

添加hive集成Atlas插件/对Dockerfile进行优化,根据环境变量选择是否启动Atlas插件

if [ -n "$ATLAS_HOOK" ]&&[ "$ATLAS_HOOK" = "true" ]&&[ `grep -c "HiveHook" /opt/hive/conf/hive-site.xml` -eq '0' ] ; then
echo " - add atlas hook -"
addProperty /opt/hive/conf/hive-site.xml hive.exec.post.hooks org.apache.atlas.hive.hook.HiveHook
echo "export HIVE_AUX_JARS_PATH=/opt/atlas/apache-atlas-hive-hook-2.1.0/hook/hive">>/opt/hive/conf/hive-env.sh
fi

atlas配置文件/opt/hive/conf/atlas-application.properties

atlas.rest.address=http://atlas-server:21000

测试hive元数据同步到Atlas

#创建新表同步到Atlas
docker-compose exec hive-server /bin/bash -c 'hive -S -e "create table ljgk_dw.d_area3(id int, name string) row format delimited fields terminated by \"\t\";"'
#创建新表生成数据血缘数据
docker-compose exec hive-server /bin/bash -c 'hive -S -e "create table ljgk_dw.d_area1 as select * from ljgk_dw.d_area;"'

其他镜像

本项目docker-compose.yml还包含其他镜像

hive-metastore-mysql服务使用的mysql:5.7镜像
Hadoop组件相关的镜像

mysql:5.7

此镜像为hive使用的元数据库，使用mysql数据库5.7版本

docker pull mysql:5.7
docker tag mysql:5.7 harbor.software.dc/mysql/mysql:5.7

Hadoop组件相关的镜像

#其中hadoop-base基础镜像需要替换部分文件，其他镜像在基础镜像基础上重新编译
# hadoop配置下的mapred-site.xml和yarn-env.sh
# hive配置下的hive-site.xml
vim ./base/DockerFile

docker build -t harbor.software.dc/mpdata/hadoop-base:2.0.0-hadoop2.7.4-java8 ./base

docker build -t harbor.software.dc/mpdata/hadoop-namenode:2.0.0-hadoop2.7.4-java8 ./namenode
docker build -t harbor.software.dc/mpdata/hadoop-datanode:2.0.0-hadoop2.7.4-java8 ./datanode
docker build -t harbor.software.dc/mpdata/hadoop-resourcemanager:2.0.0-hadoop2.7.4-java8 ./resourcemanager
docker build -t harbor.software.dc/mpdata/hadoop-nodemanager:2.0.0-hadoop2.7.4-java8 ./nodemanager
docker build -t harbor.software.dc/mpdata/hadoop-historyserver:2.0.0-hadoop2.7.4-java8 ./historyserver
docker build -t harbor.software.dc/mpdata/hadoop-submit:2.0.0-hadoop2.7.4-java8 ./submit

#镜像上传到harbor

docker push harbor.software.dc/mpdata/hadoop-namenode:2.0.0-hadoop2.7.4-java8
docker push harbor.software.dc/mpdata/hadoop-datanode:2.0.0-hadoop2.7.4-java8
docker push harbor.software.dc/mpdata/hadoop-resourcemanager:2.0.0-hadoop2.7.4-java8
docker push harbor.software.dc/mpdata/hadoop-nodemanager:2.0.0-hadoop2.7.4-java8
docker push harbor.software.dc/mpdata/hadoop-historyserver:2.0.0-hadoop2.7.4-java8
docker push harbor.software.dc/mpdata/hadoop-submit:2.0.0-hadoop2.7.4-java8

hadoop容器镜像制作,请使用分支2.0.0-hadoop2.7.4-java8

具体镜像制作方法详见http://gitlab.software.dc/mp-data/dss/docker-hadoop/-/tree/2.0.0-hadoop2.7.4-java8，！！！此工程已不在维护！！！

docker-hive

Testing

Load data into Hive:

  $ docker-compose exec hive-server bash
  # /opt/hive/bin/beeline -u jdbc:hive2://localhost:10000
  > CREATE TABLE pokes (foo INT, bar STRING);
  > LOAD DATA LOCAL INPATH '/opt/hive/examples/files/kv1.txt' OVERWRITE INTO TABLE pokes;

Then query it from PrestoDB. You can get presto.jar from PrestoDB website:

  $ wget https://repo1.maven.org/maven2/io/prestosql/presto-cli/308/presto-cli-308-executable.jar
  $ mv presto-cli-308-executable.jar presto.jar
  $ chmod +x presto.jar
  $ ./presto.jar --server localhost:8080 --catalog hive --schema default
  presto> select * from pokes;

Contributors

Ivan Ermilov @earthquakesan (maintainer)
Yiannis Mouchakis @gmouchakis
Ke Zhu @shawnzhu

poenr/docker-hive