How to compile the Linux64 bit operating system (CentOS6.6) spark1.3

1. After downloading 1.3.0 source code, execute the following command:

./make-distribution.sh--tgz--skip-java-test--with-tachyon-dhadoop.version=2.4.0-djava.version=1.7- Dprotobuf.version=2.5.0-pyarn-phive-phive-thriftserver

2. Parameter Description:

• --tgz Build the deployment package;

• --skip-java-test filter the test phase;

• --with-tachyon feel tachyon is a trend, so add tachyon support;

• -dhadoop.version Specifies the version of Hadoop, which defaults to 1.0.4;

• -djava.version Specifies the Java version, which is 1.6 by default;

• The

  -dprotobuf.version Specifies the version of PROTOBUF, which defaults to 2.4.1; This is specified because Hadoop uses 2.5.0, and does not modify the words when paired with the problem;

• -pyarn-phive-pthriftserver added support for yarn, hive, respectively.

3. after more than one hour, build SUCCESS is successful, and the compiled file can be found in the spark-1.3.0 folder.

This article is from "Levy" blog, please make sure to keep this source http://liyong0802.blog.51cto.com/2574432/1679137

How to compile the Linux64 bit operating system (CentOS6.6) spark1.3