• Impala编译部署-4


    这一步的目的是将程序运行必须的文件拷贝出来,便于发布和部署使用,我只想测试kudu,这里只拷贝了kudu的库。

    /etc/default  :准备放impala配置文件

    /etc/hadoop  :hadoop配置文件

    be      :impala运行文件

    lib64    :impala运行依赖库

    hadoop    :hadoop运行文件

    dependency  :jar包

    shell      :impala-shell

    www      :impala程序中web页面模板

    #!/bin/bash
    IMPALA_HOME=/root/impala
    IMPALA_BUILD=/root/impala-build
    
    # if exist the dest directory then clear it.
    if [ -d "${IMPALA_BUILD}" ]; then
            rm -rf ${IMPALA_BUILD}/*
    else
            mkdir -p ${IMPALA_BUILD}
    fi
    
    mkdir -p ${IMPALA_BUILD}/etc/default/
    mkdir -p ${IMPALA_BUILD}/etc/hadoop/
    mkdir -p ${IMPALA_BUILD}/be
    mkdir -p ${IMPALA_BUILD}/lib64/
    mkdir ${IMPALA_BUILD}/hadoop/
    mkdir ${IMPALA_BUILD}/dependency/
    mkdir ${IMPALA_BUILD}/shell
    
    cp -rf ${IMPALA_HOME}/be/build/debug/* ${IMPALA_BUILD}/be/
    cp -rf ${IMPALA_HOME}/toolchain/gcc-4.9.2/lib64/*   ${IMPALA_BUILD}/lib64/
    cp -rf ${IMPALA_HOME}/fe/target/impala-frontend-0.1-SNAPSHOT.jar   ${IMPALA_BUILD}/dependency/
    cp -rf ${IMPALA_HOME}/fe/target/dependency/*      ${IMPALA_BUILD}/dependency/
    cp -rf ${IMPALA_HOME}/toolchain/cdh_components/hadoop-2.6.0-cdh5.10.0-SNAPSHOT/*  ${IMPALA_BUILD}/hadoop/
    cp -rf ${IMPALA_BUILD}/hadoop/etc/hadoop ${IMPALA_BUILD}/etc/
    cp -rf ${IMPALA_HOME}/toolchain/kudu-a70c905006/debug/lib64/*  ${IMPALA_BUILD}/lib64/
    cp -rf ${IMPALA_HOME}/shell/build/impala-shell-2.8.0-RELEASE/*   ${IMPALA_BUILD}/shell
    cp -r ${IMPALA_HOME}/www ${IMPALA_BUILD}/
    
    echo "Finished"
  • 相关阅读:
    nginx.conf nginx反向代理配置文件
    linux shell date的用法
    shell find 命令 find命令报错 find: paths must precede expression:
    nginx平滑升级
    centos 6/7 tar包安装mysql 5.7
    3. Longest Substring Without Repeating Characters
    模板之类模板2
    排序之归并排序
    排序之堆排序
    排序之选择排序
  • 原文地址:https://www.cnblogs.com/fangjx/p/6863279.html
Copyright © 2020-2023  润新知