Table of Contents |
---|
This post concentrates on installing, configuring and testing Apache Hadoop, Spark, Hive and HBase built using Apache BigTop v1.2 and was provided as part of ERP 17.08 release.
...
These packages are: flume, mahout, pig, sqoop, zookeeper. If all components were already built in earlier step, skip this step.
Code Block | ||
---|---|---|
| ||
docker run -v `pwd`:/ws bigtop/slaves:erp17.08-deb-8-aarch64 bash -l -c 'cd /ws ; ./gradlew flume-deb mahout-deb pig-deb sqoop-deb zookeeper-deb' |
...
- To run the Spark smoke tests you need to apply 2 patches https://issues.apache.org/jira/browse/BIGTOP-2860 and https://issues.apache.org/jira/browse/BIGTOP-2866
Code Block | ||
---|---|---|
| ||
$ vim provisioner/docker/config_debian-8-aarch64.yaml Do the below changes components: [hdfs, yarn, mapreduce, hive, spark] enable_local_repo: true smoke_test_components: [spark] jdk: "openjdk-8-jdk" save and quit. $ vim provisioner/utils/smoke-tests.sh Add the below lines export HADOOP_HOME=/usr/lib/hadoop export HADOOP_PREFIX=$HADOOP_HOME export HADOOP_OPTS="-Djava.library.path=$HADOOP_PREFIX/lib/native" export HADOOP_LIBEXEC_DIR=/usr/lib/hadoop/libexec export HADOOP_CONF_DIR=/etc/hadoop/conf export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_HOME/lib/native export HADOOP_COMMON_HOME=$HADOOP_HOME export HADOOP_MAPRED_HOME=/usr/lib/hadoop-mapreduce export HADOOP_HDFS_HOME=/usr/lib/hadoop-hdfs export YARN_HOME=/usr/lib/hadoop-yarn export HADOOP_YARN_HOME=/usr/lib/hadoop-yarn/ export HADOOP_USER_NAME=hdfs export CLASSPATH=$CLASSPATH:. export CLASSPATH=$CLASSPATH:$HADOOP_HOME/hadoop-common-2.7.2.jar:$HADOOP_HOME/client/hadoop-hdfs-2.7.2.jar:$HADOOP_HOME/hadoop-auth-2.7.2.jar:/usr/lib/hadoop-mapreduce/*:/usr/lib/hive/lib/*:/usr/lib/hadoop/lib/*: export JAVA_HOME=$(readlink -f /usr/bin/java | sed "s:bin/java::") export PATH=/usr/lib/hadoop/libexec:/etc/hadoop/conf:$HADOOP_HOME/bin/:$PATH export SPARK_HOME=/usr/lib/spark export PATH=$HADOOP_HOME\bin:$PATH export SPARK_DIST_CLASSPATH=$HADOOP_HOME\bin\hadoop:$CLASSPATH:/usr/lib/hadoop/lib/*:/usr/lib/hadoop/lib/*:/usr/lib/hadoop-mapreduce/*:. export CLASSPATH=$CLASSPATH:/usr/lib/hadoop/lib/*:. export SPARK_HOME=/usr/lib/spark export SPARK_MASTER_IP=<IP Address of your host> export SPARK_MASTER_PORT=7077 save and quit. Run the command smoke test commands $ ./docker-hadoop.sh -C config_debian-8-aarch64.yaml -c 3 -s -d If the smoke test is still failing. It might be because of you have't set the "SPARK_MASTER_IP" properly. Another way to do is that remove the $masterMode and add local[*] as below $ vim bigtop-tests/smoke-tests/spark/TestSpark.groovy make the below change final String SPARK_SHELL = SPARK_HOME + "/bin/spark-shell --master local[*]" save and quit. Now run the spark smoke-tests again. |
...
TestSparkExample |
TestSparkPythonExample |
ShellTest |
HDFSTest |
JobTest |
Hive
TestHiveSmoke
Component versions
- alluxio v1.0.1
- Apache ambari v2.5.0
- Apache apex v3.5.0
- groovy v2.4.10
- Apache commons - jsvc v1.0.15
- Apache tomcat v6.0.45
- bigtop_utils v1.2.0
- Apache crunch v0.14.0
- Pig UDF datafu v1.3.0
- Apache flink v1.1.3
- Apache flume v1.7.0
- Apache giraph v1.1.0
- Greenplum gpdb v5.0.0-alpha.0
- Apache hadoop v2.7.3
- Apache hama v0.7.0
- Apache hbase v1.1.3
- Apache hive v1.2.1
- Apache hue v3.11.0
- Apache ignite v1.9.0
- Apache kafka v0.10.1.1
- kite v1.1.0
- Apache mahout v0.12.2
- Apache oozie v4.3.0
- Apache phoenix v4.9.0-HBase-1.1
- Apache pig v0.15.0
- Quantcast qfs v1.1.4
- Apache solr v4.10.4
- Apache spark 1.1 v1.6.2
- Apache spark 2.0 v2.1.1
- Apache sqoop v1 v1.4.6
- Apache sqoop v2 v1.99.4
- Apache tajo v0.11.1
- Apache tez v0.6.2
- ycsb v0.4.0
- Apache zeppelin v0.7.0
- Apache zookeeper v3.4.6
...