大数据环境搭建@Hive编译

Hive3.1.3编译

1.编译原因
- 1.1Guava依赖冲突
- 1.2开启MetaStore后运行有StatsTask报错
- 1.3Spark版本过低
2.环境部署
- 2.1jdk安装
- 2.2maven部署
- 2.3安装图形化桌面
- 2.4安装Git
- 2.5安装IDEA
3.拉取Hive源码
4.Hive源码编译
- 4.1环境测试
- - 1.测试方法——编译
  - 2.问题及解决方案
  - - 💥问题1：下载不到 pentaho-aggdesigner-algorithm-5.1.5-jhyde.jar
    - 💥问题2：阿里云镜像没有被使用
    - 💥问题3：jdk版本冲突，<2.环境部署>做的不细致
- 4.2解决Guava版本冲突问题
- - 1.修改内容
  - 2.问题及解决方案
  - - 💥问题1：`Futures.addCallback()`方法27.0-jre中3个参数，19.0中2个参数
    - 💥问题2：Iterators的 `emptyIterator` 方法过时了
- 4.3开启MetaStore之后StatsTask报错
- - 1.修改内容
  - 2.问题及解决方案
  - - 💥问题1：cherry-pick失败
- 4.4Spark兼容问题
- - 1.修改内容
  - 2.问题及解决方案
  - - 💥问题1：SparkCounter中方法过时，需要替换
    - 💥问题2：ShuffleWriteMetrics中方法过时，需要替换
    - 💥问题3：TestStatsUtils中方法过时，需要替换
- 4.5编译成功

各组件版本选择：
hadoop-3.3.2
hive-3.1.3
spark-3.3.4 Scala version 2.12.15 （spark-3.3.4依赖hadoop-3.3.2）

1.编译原因

1.1Guava依赖冲突

tail -200 /tmp/root/hive.log > /home/log/hive-200.log

hive的github地址

https://github.com/apache/hive

查询guava依赖

https://github.com/apache/hive/blob/rel/release-3.1.3/pom.xml
<guava.version>19.0</guava.version>

hadoop的github地址

https://github.com/apache/hadoop

查询guava依赖

https://github.com/apache/hadoop/blob/rel/release-3.3.2/hadoop-project/pom.xml
<guava.version>27.0-jre</guava.version>

1.2开启MetaStore后运行有StatsTask报错

FAILED: Execution Error, return code 1 from org.apache.hadoop.hive.ql.exec.StatsTask
MapReduce Jobs Launched:

1.3Spark版本过低

Hive3.1.3默认支持Spark2.3.0，版本过低很多新的高效方法都没用到，所以替换为spark-3.3.4（Hadoop3.3.2支持的最高spark版本）

<spark.version>2.3.0</spark.version>

2.环境部署

2.1jdk安装

已安装1.8版本

(base) [root@bigdata01 opt]# java -version
java version "1.8.0_301"
Java(TM) SE Runtime Environment (build 1.8.0_301-b09)
Java HotSpot(TM) 64-Bit Server VM (build 25.301-b09, mixed mode)

2.2maven部署

下载3.6.3安装包 https://archive.apache.org/dist/maven/maven-3/3.6.3/binaries/

(base) [root@bigdata01 ~]# cd /opt
(base) [root@bigdata01 opt]# tar -zxvf apache-maven-3.6.3-bin.tar.gz
(base) [root@bigdata01 opt]# mv apache-maven-3.6.3 maven
(base) [root@bigdata01 opt]# vim /etc/profile
# 增加MAVEN_HOME
export MAVEN_HOME=/opt/maven
export PATH=$MAVEN_HOME/bin:$PATH
(base) [root@bigdata01 opt]# source /etc/profile

监测 maven 是否安装成功

(base) [root@bigdata01 opt]# mvn -version
Apache Maven 3.6.3 ()
Maven home: /opt/maven
Java version: 1.8.0_301, vendor: Oracle Corporation, runtime: /opt/jdk/jre
Default locale: zh_CN, platform encoding: UTF-8
OS name: "linux", version: "4.18.0-365.el8.x86_64", arch: "amd64", family: "unix"

配置仓库镜像，阿里云公共仓库

vim /opt/maven/conf/settings.xml

<?xml version="1.0" encoding="UTF-8"?>
<settings xmlns="http://maven.apache.org/SETTINGS/1.0.0"xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"xsi:schemaLocation="http://maven.apache.org/SETTINGS/1.0.0 http://maven.apache.org/xsd/settings-1.0.0.xsd"><localRepository>/repo</localRepository><mirrors><mirror><id>aliyunmaven</id><mirrorOf>central</mirrorOf><name>阿里云公共仓库</name><url>https://maven.aliyun.com/repository/public</url></mirror></mirrors>
</settings>

2.3安装图形化桌面

采用带图形界面的Centos，卸载多余的jdk，避免版本冲突。这里的操作非常重要，不处理会报稀奇古怪的错误。例如：Fatal error compiling: 无效的目标发行版: 1.11 ，报错java1.11没有这个版本，即使升了java11也没有用。

找到多余安装的jdk

(base) [root@bigdata01 ~]# yum list installed |grep jdk
copy-jdk-configs.noarch                            4.0-2.el8                                                  @appstream        
java-1.8.0-openjdk.x86_64                          1:1.8.0.362.b08-3.el8                                      @appstream        
java-1.8.0-openjdk-devel.x86_64                    1:1.8.0.362.b08-3.el8                                      @appstream        
java-1.8.0-openjdk-headless.x86_64                 1:1.8.0.362.b08-3.el8                                      @appstream

卸载多余安装的jdk

(base) [root@bigdata01 ~]# yum remove -y copy-jdk-configs.noarch java-1.8.0-openjdk.x86_64 java-1.8.0-openjdk-devel.x86_64 java-1.8.0-openjdk-headless.x86_64

验证当前的jdk

(base) [root@bigdata01 ~]# java -version
java version "1.8.0_301"
Java(TM) SE Runtime Environment (build 1.8.0_301-b09)
Java HotSpot(TM) 64-Bit Server VM (build 25.301-b09, mixed mode)

2.4安装Git

安装第三方仓库

(base) [root@bigdata01 opt]# yum install https://repo.ius.io/ius-release-el7.rpm https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm

在这里插入图片描述

安装Git

(base) [root@bigdata01 opt]# yum install -y git

git版本检查

(base) [root@bigdata01 ~]# git -v
git version 2.43.0

2.5安装IDEA

https://download.jetbrains.com.cn/idea/ideaIU-2021.1.3.tar.gz 下载 linux版

(base) [root@bigdata01 opt]# tar -zxvf ideaIU-2021.1.3.tar.gz

启动IDEA，启动图形化界面要在VMware中

cd /opt/idea-IU-211.7628.21
./bin/idea.sh

这里试用30天

在这里插入图片描述

配置 maven，settings.xml中已配置阿里云公共仓库地址

设置 idea 快捷图标（这里的 bluetooth-sendto.desktop 是随便复制了一个，可以任意换）

(base) [root@bigdata01 bin]# cd /usr/share/applications
(base) [root@bigdata01 applications]# cp bluetooth-sendto.desktop idea.desktop
(base) [root@bigdata01 applications]# vim idea.desktop
# 删掉原有的，补充这个内容
[Desktop Entry]
Name=idea
Exec=sh /opt/idea-IU-211.7628.21/bin/idea.sh
Terminal=false
Type=Application
Icon=/opt/idea-IU-211.7628.21/bin/idea.png
Comment=idea
Categories=Application;

在这里插入图片描述

3.拉取Hive源码

Get from VCS拉取hive源码，拉取的全过程大约需要1小时
在这里插入图片描述

配置URL https://gitee.com/apache/hive.git，并设置文件地址

在这里插入图片描述

信任项目后注意配置，这里按图填，否则容易jdk版本异常造成错误(-Xmx2048m)
在这里插入图片描述

加载hive3.1.3

在这里插入图片描述

并建立分支 slash-hive-3.1.3

在这里插入图片描述

4.Hive源码编译

4.1环境测试

1.测试方法——编译

https://hive.apache.org/development/gettingstarted/ 点击Getting Started Guide

https://cwiki.apache.org/confluence/display/Hive/GettingStarted#GettingStarted-BuildingHivefromSource 点击Building Hive from Source

获得编码方式，执行在 Terminal 终端执行，运行成功的7min左右

mvn clean package -Pdist -DskipTests -Dmaven.javadoc.skip=true

2.问题及解决方案

💥问题1：下载不到 pentaho-aggdesigner-algorithm-5.1.5-jhyde.jar

[ERROR] Failed to execute goal on project hive-upgrade-acid: Could not resolve dependencies for project org.apache.hive:hive-upgrade-acid:jar:3.1.3

Downloading from conjars: http://conjars.org/repo/org/pentaho/pentaho-aggdesigner-algorithm/5.1.5-jhyde/pentaho-aggdesigner-algorithm-5.1.5-jhyde.pom 下载不到 pentaho-aggdesigner-algorithm-5.1.5-jhyde.jar

这个问题是一个已知问题，它是由于Pentaho公司的Maven存储库服务器已被永久关闭，所以无法从该仓库获取它的依赖项导致的。

解决方案，先修改 /opt/maven/conf/setting.xml 文件如下

<?xml version="1.0" encoding="UTF-8"?>
<settings xmlns="http://maven.apache.org/SETTINGS/1.0.0"xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"xsi:schemaLocation="http://maven.apache.org/SETTINGS/1.0.0 http://maven.apache.org/xsd/settings-1.0.0.xsd"><localRepository>/repo</localRepository><mirrors><!-- 添加这个镜像仓库在阿里云公共仓库前面 --><mirror><id>aliyunmaven</id><mirrorOf>*</mirrorOf><name>spring-plugin</name><url>https://maven.aliyun.com/repository/spring-plugin</url></mirror><mirror><id>aliyunmaven</id><mirrorOf>central</mirrorOf><name>阿里云公共仓库</name><url>https://maven.aliyun.com/repository/public</url></mirror></mirrors>
</settings>

成功下载 /org/pentaho/ 相关内容后再改回去！！！！

💥问题2：阿里云镜像没有被使用

[ERROR] Failed to execute goal on project hive-upgrade-acid: Could not resolve dependencies for project org.apache.hive:hive-upgrade-acid:jar:3.1.3

Downloading from conjars: https://maven.glassfish.org/content/groups/glassfish/asm/asm/3.1/asm-3.1.jar 下载不到 asm-3.1.jar

修改/opt/maven/conf/settings.xml文件，之前的阿里云镜像没有被使用。复制如下内容，覆盖整个settings.xml文件

<?xml version="1.0" encoding="UTF-8"?>
<settings xmlns="http://maven.apache.org/SETTINGS/1.0.0"xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"xsi:schemaLocation="http://maven.apache.org/SETTINGS/1.0.0 http://maven.apache.org/xsd/settings-1.0.0.xsd">
<localRepository>/repo</localRepository>
<mirrors><mirror><id>aliyunmaven</id><mirrorOf>central</mirrorOf><name>阿里云公共仓库</name><url>https://maven.aliyun.com/repository/public</url></mirror>
</mirrors>
</settings>

配置后重启服务，阿里云镜像被成功使用

💥问题3：jdk版本冲突，<2.环境部署>做的不细致

[ERROR] Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.5.1:compile (default-compile) on project hive-upgrade-acid: Fatal error compiling: 无效的目标发行版: 1.11 -> [Help 1]

报这个错误是jdk版本冲突了，Linux版尽管 java -version 都显示了 1.8版本，但图形化、IDEA没做处理就会有很多jdk存在，需要做的就是重新做<2.环境部署><3.拉取Hive源码>

4.2解决Guava版本冲突问题

1.修改内容

修改pom.xml中的guava.version的版本为 27.0-jre

# 原来版本
<guava.version>19.0</guava.version>
# 修改后版本
<guava.version>27.0-jre</guava.version>

修改版本后执行编译 mvn clean package -Pdist -DskipTests -Dmaven.javadoc.skip=true

结果保存：/home/slash/hive/packaging/target/apache-hive-3.1.3-bin.tar.gz

2.问题及解决方案

💥问题1：`Futures.addCallback()`方法27.0-jre中3个参数，19.0中2个参数

[ERROR] Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.6.1:compile (default-compile) on project hive-llap-common: Compilation failure: Compilation failure: 
[ERROR] /home/slash/hive/llap-common/src/java/org/apache/hadoop/hive/llap/AsyncPbRpcProxy.java:[173,16] 无法将类 com.google.common.util.concurrent.Futures中的方法 addCallback应用到给定类型;
[ERROR]   需要: com.google.common.util.concurrent.ListenableFuture<V>,com.google.common.util.concurrent.FutureCallback<? super V>,java.util.concurrent.Executor
[ERROR]   找到: com.google.common.util.concurrent.ListenableFuture<U>,org.apache.hadoop.hive.llap.AsyncPbRpcProxy.ResponseCallback<U>
[ERROR]   原因: 无法推断类型变量 V
[ERROR]     (实际参数列表和形式参数列表长度不同)

修改 Futures.addCallback()，为其增加第3个参数,MoreExecutors.directExecutor()，这个修改大概15处，方法相同

// 原来的
@VisibleForTesting<T extends Message , U extends Message> void submitToExecutor(CallableRequest<T, U> request, LlapNodeId nodeId) {ListenableFuture<U> future = executor.submit(request);Futures.addCallback(future, new ResponseCallback<U>(request.getCallback(), nodeId, this));}// 修改后的
@VisibleForTesting<T extends Message , U extends Message> void submitToExecutor(CallableRequest<T, U> request, LlapNodeId nodeId) {ListenableFuture<U> future = executor.submit(request);Futures.addCallback(future, new ResponseCallback<U>(request.getCallback(), nodeId, this),MoreExecutors.directExecutor());}

过程中如果出现 ”找不到MoreExecutors方法“的问题可以手动 import 这个方法，具体方法可以拷贝其他文件中的 import
在这里插入图片描述

💥问题2：Iterators的 `emptyIterator` 方法过时了

[ERROR] Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.6.1:compile (default-compile) on project hive-druid-handler: Compilation failure
[ERROR] /home/slash/hive/druid-handler/src/java/org/apache/hadoop/hive/druid/serde/DruidScanQueryRecordReader.java:[46,61] <T>emptyIterator()在com.google.common.collect.Iterators中不是公共的; 无法从外部程序包中对其进行访问

修改Iterators中的emptyIterator方法

# org.apache.hadoop.hive.druid.serde.DruidScanQueryRecordReader
# 原始代码private Iterator<List<Object>> compactedValues = Iterators.emptyIterator();
# 修改后代码private Iterator<List<Object>> compactedValues = ImmutableSet.<List<Object>>of().iterator();

在这里插入图片描述

4.3开启MetaStore之后StatsTask报错

1.修改内容

# 错误信息
FAILED: Execution Error, return code 1 from org.apache.hadoop.hive.ql.exec.StatsTask
MapReduce Jobs Launched: # 错误日志 /tmp/root/hive.log
exec.StatsTask: Failed to run stats task
org.apache.hadoop.hive.ql.metadata.HiveException: org.apache.thrift.transport.TTransp
ortException

错误分析见 https://issues.apache.org/jira/browse/HIVE-19316

IDEA点击 Cherry-pick，将StatsTask fails due to ClassCastException的补丁合并到当前分支

在这里插入图片描述

修改版本后执行编译 mvn clean package -Pdist -DskipTests -Dmaven.javadoc.skip=true

结果保存：/home/slash/hive/packaging/target/apache-hive-3.1.3-bin.tar.gz

2.问题及解决方案

💥问题1：cherry-pick失败

Cherry-pick failed
3d21bc38 HIVE-19316: StatsTask fails due to ClassCastException (Jaume Marhuenda, reviewed by Jesus Camacho Rodriguez)
Committer identity unknown
*** Please tell me who you are.
Run
git config --global user.email “you@example.com”
git config --global user.name “Your Name”
to set your account’s default identity.
Omit --global to set the identity only in this repository.
unable to auto-detect email address (got ‘root@bigdata01.(none)’)

需要提交修复的版本信息

Cherry-pick failed

3d21bc38 HIVE-19316: StatsTask fails due toClassCastException (Jaume Marhuenda, reviewedby Jesus Camacho Rodriguez)your local changes would be overwritten bycherry-pick.hint: commit your changes or stash them toproceed.cherry-pick failed

工作目录中已经存在一些未提交的更改。git 不允许在未提交更改的情况下进行 cherry-pick

# 提交修复的版本信息
git config --global user.email "360322495@qq.com"
git config --global user.name "slash"# 添加并commit提交
git add .
git commit -m "resolve conflict  guava"

4.4Spark兼容问题

1.修改内容

修改pom.xml中的 spark.version、scala.version、hadoop.version

<!-- 原始代码 -->
<spark.version>2.3.0</spark.version>
<scala.binary.version>2.11</scala.binary.version>
<scala.version>2.11.8</scala.version><!-- 修改后代码 -->
<spark.version>3.3.4</spark.version>
<scala.binary.version>2.12</scala.binary.version>
<scala.version>2.12.15</scala.version>

spark中消除部分hadoop依赖，hive3.1.3依赖的是hadoop3.1.0，不同于spark-3.3.4依赖hadoop3.3.2，不用改hive pom的hadoop依赖

<!-- 修改后代码 -->
<groupId>org.apache.spark</groupId>
<artifactId>spark-core_${scala.binary.version}</artifactId>
<version>${spark.version}</version>
<exclusions><exclusion><groupId>org.apache.hadoop</groupId><artifactId>hadoop-core</artifactId></exclusion><exclusion><groupId>org.apache.hadoop</groupId><artifactId>hadoop-client</artifactId></exclusion><exclusion><groupId>org.apache.hadoop</groupId><artifactId>hadoop-client-api</artifactId></exclusion><exclusion><groupId>org.apache.hadoop</groupId><artifactId>hadoop-client-runtime</artifactId></exclusion>
</exclusions>

2.问题及解决方案

💥问题1：SparkCounter中方法过时，需要替换

[ERROR] Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.6.1:compile (default-compile) on project hive-spark-client: Compilation failure: Compilation failure: 
[ERROR] /home/slash/hive/spark-client/src/main/java/org/apache/hive/spark/counter/SparkCounter.java:[22,24] 找不到符号
[ERROR]   符号:   类 Accumulator
[ERROR]   位置: 程序包 org.apache.spark
[ERROR] /home/slash/hive/spark-client/src/main/java/org/apache/hive/spark/counter/SparkCounter.java:[23,24] 找不到符号
[ERROR]   符号:   类 AccumulatorParam
[ERROR]   位置: 程序包 org.apache.spark
[ERROR] /home/slash/hive/spark-client/src/main/java/org/apache/hive/spark/counter/SparkCounter.java:[30,11] 找不到符号
[ERROR]   符号:   类 Accumulator
[ERROR]   位置: 类 org.apache.hive.spark.counter.SparkCounter
[ERROR] /home/slash/hive/spark-client/src/main/java/org/apache/hive/spark/counter/SparkCounter.java:[91,41] 找不到符号
[ERROR]   符号:   类 AccumulatorParam
[ERROR]   位置: 类 org.apache.hive.spark.counter.SparkCounter

移除无用的方法，并修改相关内容，最终结果如下

/** Licensed to the Apache Software Foundation (ASF) under one* or more contributor license agreements.  See the NOTICE file* distributed with this work for additional information* regarding copyright ownership.  The ASF licenses this file* to you under the Apache License, Version 2.0 (the* "License"); you may not use this file except in compliance* with the License.  You may obtain a copy of the License at* <p/>* http://www.apache.org/licenses/LICENSE-2.0* <p/>* Unless required by applicable law or agreed to in writing, software* distributed under the License is distributed on an "AS IS" BASIS,* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.* See the License for the specific language governing permissions and* limitations under the License.*/
package org.apache.hive.spark.counter;import java.io.Serializable;import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.util.LongAccumulator;public class SparkCounter implements Serializable {private String name;private String displayName;private LongAccumulator accumulator;// Values of accumulators can only be read on the SparkContext side. This field is used when// creating a snapshot to be sent to the RSC client.private long accumValue;public SparkCounter() {// For serialization.}private SparkCounter(String name,String displayName,long value) {this.name = name;this.displayName = displayName;this.accumValue = value;}public SparkCounter(String name,String displayName,String groupName,long initValue,JavaSparkContext sparkContext) {this.name = name;this.displayName = displayName;String accumulatorName = groupName + "_" + name;this.accumulator = sparkContext.sc().longAccumulator(accumulatorName);this.accumulator.setValue(initValue);}public long getValue() {if (accumulator != null) {return accumulator.value();} else {return accumValue;}}public void increment(long incr) {accumulator.add(incr);}public String getName() {return name;}public String getDisplayName() {return displayName;}public void setDisplayName(String displayName) {this.displayName = displayName;}SparkCounter snapshot() {return new SparkCounter(name, displayName, accumulator.value());}}

💥问题2：ShuffleWriteMetrics中方法过时，需要替换

[ERROR] Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.6.1:compile (default-compile) on project hive-spark-client: Compilation failure: Compilation failure: 
[ERROR] /home/slash/hive/spark-client/src/main/java/org/apache/hive/spark/client/metrics/ShuffleWriteMetrics.java:[50,39] 找不到符号
[ERROR]   符号:   方法 shuffleBytesWritten()
[ERROR]   位置: 类 org.apache.spark.executor.ShuffleWriteMetrics
[ERROR] /home/slash/hive/spark-client/src/main/java/org/apache/hive/spark/client/metrics/ShuffleWriteMetrics.java:[51,36] 找不到符号
[ERROR]   符号:   方法 shuffleWriteTime()
[ERROR]   位置: 类 org.apache.spark.executor.ShuffleWriteMetrics

修改相关方法

// 原始代码
public ShuffleWriteMetrics(TaskMetrics metrics) {this(metrics.shuffleWriteMetrics().shuffleBytesWritten(),metrics.shuffleWriteMetrics().shuffleWriteTime());}// 修改后
public ShuffleWriteMetrics(TaskMetrics metrics) {this(metrics.shuffleWriteMetrics().bytesWritten(),metrics.shuffleWriteMetrics().writeTime());}

💥问题3：TestStatsUtils中方法过时，需要替换

[ERROR] Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.6.1:testCompile (default-testCompile) on project hive-exec: Compilation failure
[ERROR] /home/slash/hive/ql/src/test/org/apache/hadoop/hive/ql/stats/TestStatsUtils.java:[34,39] 程序包org.spark_project.guava.collect不存在

修改相关方法

// 原始代码
import org.spark_project.guava.collect.Sets;// 修改后
import org.sparkproject.guava.collect.Sets;

4.5编译成功

mvn clean package -Pdist -DskipTests -Dmaven.javadoc.skip=true

Hive3.1.3-spark-3.3.4-hadoop-3.3.2编译成功，结果保存：/home/slash/hive/packaging/target/apache-hive-3.1.3-bin.tar.gz

在这里插入图片描述

声明：本文所载信息不保证准确性和完整性。文中所述内容和意见仅供参考，不构成实际商业建议，可收藏可转发但请勿转载，如有雷同纯属巧合