Is it possible to install Beeline to run Hive queries without installing Hive?

Question

I'm on a Mac OSX machine and I'd like to run queries against a Hadoop db on a CentOS 6.6 machine. I can log in to the CentOS machine and run hive queries there. But I need to be able to run queries from my machine to troubleshoot connection issues.

Is there a way to install Beeline (the newer version of Hive CLI) or Hive on OSX without installing/configuring Hadoop? The information that I've seen says that you need to install Hadoop first, which seems like overkill just to test whether a remote database is listening for connections.

Answer 1

$ brew install hive worked well enough. I guess I'll leave this question up since I couldn't find the answer on the internet. 141 megs of disk space though, boo.

Answer 2

It is not necessary to install beeline/hive. All you have to do is collect the relevant jars from your system and copy them and place them in a single folder.

Suppose, we have a source system where you have beeline and a target system where you want to run beeline.

On the source system collect the relevant jars into one folder. The best way I have found to identify the exact jars involved is to use the jvm option: -verbose:class

I.e., you should be able to issue a java command that will replicate a typical beeline command invocation on the source system.

Then copy those files into one folder on the target system. cd to that folder to make the -classpath reference later simple.

I use an HDP 2.5 Hortonworks distro. For me, the following invocation on the target system works: java -Xmx1024m -classpath apache-log4j-extras-1.2.17.jar:avatica-1.8.0.2.5.0.0-1245.jar:calcite-core-1.2.0.2.5.0.0-1245.jar:calcite-linq4j-1.2.0.2.5.0.0-1245.jar:commons-cli-1.2.jar:commons-codec-1.4.jar:commons-collections-3.2.2.jar:commons-configuration-1.6.jar:commons-lang-2.6.jar:commons-logging-1.1.3.jar:curator-client-2.6.0.jar:curator-framework-2.6.0.jar:derby-10.10.2.0.jar:guava-14.0.1.jar:hadoop-annotations-2.7.3.2.5.0.0-1245.jar:hadoop-auth-2.7.3.2.5.0.0-1245.jar:hadoop-common-2.7.3.2.5.0.0-1245.jar:hadoop-mapreduce-client-core-2.7.3.2.5.0.0-1245.jar:hive-beeline-1.2.1000.2.5.0.0-1245.jar:hive-exec-1.2.1000.2.5.0.0-1245.jar:hive-jdbc-1.2.1000.2.5.0.0-1245.jar:hive-jdbc-1.2.1000.2.5.0.0-1245-standalone.jar:jce.jar:jline-2.12.jar:jsse.jar:log4j-1.2.16.jar:rt.jar:slf4j-log4j12-1.7.10.jar:sunec.jar:sunjce_provider.jar:super-csv-2.2.0.jar:xercesImpl-2.9.1.jar -Dhdp.version=2.5.0.0-1245 -Djava.net.preferIPv4Stack=true -Dhdp.version=2.5.0.0-1245 -Dhadoop.log.dir=/home/userid -Dhadoop.log.file=hadoop.log -Dhadoop.home.dir=/home/userid -Dhadoop.id.str=userid -Dhadoop.root.logger=INFO,console -Djava.library.path=:/home/userid -Dhadoop.policy.file=hadoop-policy.xml -Djava.net.preferIPv4Stack=true -Djava.util.logging.config.file=/home/userid/parquet-logging.properties -Dlog4j.configuration=beeline-log4j.properties -Dhadoop.security.logger=INFO,NullAppender org.apache.hadoop.util.RunJar /home/userid/hive-beeline-1.2.1000.2.5.0.0-1245.jar org.apache.hive.beeline.BeeLine -n userid -p pass -u "jdbc:hive2://knox.company.com:8000/;ssl=true;transportMode=http;httpPath=gateway/tdcprd/hive"

Some of the parameters are probably not necessary, but I kept them because that is how it is done on the source system. You should use source system's java invocation as a reference pattern.