hadoop/hadoop-tools/hadoop-aws/pom.xml
Steve Loughran e199da3fae
HADOOP-17833. Improve Magic Committer performance (#3289)
Speed up the magic committer with key changes being

* Writes under __magic always retain directory markers

* File creation under __magic skips all overwrite checks,
  including the LIST call intended to stop files being
	created over dirs.
* mkdirs under __magic probes the path for existence
  but does not look any further.  	

Extra parallelism in task and job commit directory scanning
Use of createFile and openFile with parameters which all for
HEAD checks to be skipped.

The committer can write the summary _SUCCESS file to the path
`fs.s3a.committer.summary.report.directory`, which can be in a
different file system/bucket if desired, using the job id as
the filename. 

Also: HADOOP-15460. S3A FS to add `fs.s3a.create.performance`

Application code can set the createFile() option
fs.s3a.create.performance to true to disable the same
safety checks when writing under magic directories.
Use with care.

The createFile option prefix `fs.s3a.create.header.`
can be used to add custom headers to S3 objects when
created.


Contributed by Steve Loughran.
2022-06-17 19:11:35 +01:00

582 lines
24 KiB
XML

<?xml version="1.0" encoding="UTF-8"?>
<!--
Licensed 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
http://www.apache.org/licenses/LICENSE-2.0
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. See accompanying LICENSE file.
-->
<project xmlns="http://maven.apache.org/POM/4.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 https://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-project</artifactId>
<version>3.4.0-SNAPSHOT</version>
<relativePath>../../hadoop-project</relativePath>
</parent>
<artifactId>hadoop-aws</artifactId>
<version>3.4.0-SNAPSHOT</version>
<name>Apache Hadoop Amazon Web Services support</name>
<description>
This module contains code to support integration with Amazon Web Services.
It also declares the dependencies needed to work with AWS services.
</description>
<packaging>jar</packaging>
<properties>
<file.encoding>UTF-8</file.encoding>
<downloadSources>true</downloadSources>
<hadoop.tmp.dir>${project.build.directory}/test</hadoop.tmp.dir>
<javadoc.skip.jdk11>true</javadoc.skip.jdk11>
<!-- are scale tests enabled ? -->
<fs.s3a.scale.test.enabled>unset</fs.s3a.scale.test.enabled>
<!-- Size in MB of huge files. -->
<fs.s3a.scale.test.huge.filesize>unset</fs.s3a.scale.test.huge.filesize>
<!-- Size in MB of the partion size in huge file uploads. -->
<fs.s3a.scale.test.huge.partitionsize>unset</fs.s3a.scale.test.huge.partitionsize>
<!-- Timeout in seconds for scale tests.-->
<fs.s3a.scale.test.timeout>3600</fs.s3a.scale.test.timeout>
<!-- Set a longer timeout for integration test (in milliseconds) -->
<test.integration.timeout>200000</test.integration.timeout>
<!-- should directory marker retention be audited? -->
<fs.s3a.directory.marker.audit>false</fs.s3a.directory.marker.audit>
<!-- marker retention policy -->
<fs.s3a.directory.marker.retention></fs.s3a.directory.marker.retention>
</properties>
<profiles>
<profile>
<id>tests-off</id>
<activation>
<file>
<missing>src/test/resources/auth-keys.xml</missing>
</file>
</activation>
<properties>
<skipITs>true</skipITs>
</properties>
</profile>
<profile>
<id>tests-on</id>
<activation>
<file>
<exists>src/test/resources/auth-keys.xml</exists>
</file>
</activation>
<properties>
<skipITs>false</skipITs>
</properties>
</profile>
<profile>
<id>parallel-tests</id>
<activation>
<property>
<name>parallel-tests</name>
</property>
</activation>
<build>
<plugins>
<plugin>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-maven-plugins</artifactId>
<executions>
<execution>
<id>parallel-tests-createdir</id>
<goals>
<goal>parallel-tests-createdir</goal>
</goals>
</execution>
</executions>
</plugin>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-surefire-plugin</artifactId>
<configuration>
<forkCount>${testsThreadCount}</forkCount>
<reuseForks>false</reuseForks>
<argLine>${maven-surefire-plugin.argLine} -DminiClusterDedicatedDirs=true</argLine>
<systemPropertyVariables>
<testsThreadCount>${testsThreadCount}</testsThreadCount>
<test.build.data>${test.build.data}/${surefire.forkNumber}</test.build.data>
<test.build.dir>${test.build.dir}/${surefire.forkNumber}</test.build.dir>
<hadoop.tmp.dir>${hadoop.tmp.dir}/${surefire.forkNumber}</hadoop.tmp.dir>
<!-- Due to a Maven quirk, setting this to just -->
<!-- surefire.forkNumber won't do the parameter -->
<!-- substitution. Putting a prefix in front of it like -->
<!-- "fork-" makes it work. -->
<!-- Important: Those leading 0s are needed to guarantee that -->
<!-- trailing three chars are always numeric and unique -->
<test.unique.fork.id>fork-000${surefire.forkNumber}</test.unique.fork.id>
<!-- Propagate scale parameters -->
<fs.s3a.scale.test.enabled>${fs.s3a.scale.test.enabled}</fs.s3a.scale.test.enabled>
<fs.s3a.scale.test.huge.filesize>${fs.s3a.scale.test.huge.filesize}</fs.s3a.scale.test.huge.filesize>
<fs.s3a.scale.test.huge.huge.partitionsize>${fs.s3a.scale.test.huge.partitionsize}</fs.s3a.scale.test.huge.huge.partitionsize>
<fs.s3a.scale.test.timeout>${fs.s3a.scale.test.timeout}</fs.s3a.scale.test.timeout>
<!-- Markers-->
<fs.s3a.directory.marker.retention>${fs.s3a.directory.marker.retention}</fs.s3a.directory.marker.retention>
<fs.s3a.directory.marker.audit>${fs.s3a.directory.marker.audit}</fs.s3a.directory.marker.audit>
</systemPropertyVariables>
</configuration>
</plugin>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-failsafe-plugin</artifactId>
<executions>
<execution>
<id>default-integration-test</id>
<goals>
<goal>integration-test</goal>
<goal>verify</goal>
</goals>
<configuration>
<forkCount>${testsThreadCount}</forkCount>
<reuseForks>false</reuseForks>
<argLine>${maven-surefire-plugin.argLine} -DminiClusterDedicatedDirs=true</argLine>
<forkedProcessTimeoutInSeconds>${fs.s3a.scale.test.timeout}</forkedProcessTimeoutInSeconds>
<trimStackTrace>false</trimStackTrace>
<systemPropertyVariables>
<!-- Tell tests that they are being executed in parallel -->
<test.parallel.execution>true</test.parallel.execution>
<test.build.data>${test.build.data}/${surefire.forkNumber}</test.build.data>
<test.build.dir>${test.build.dir}/${surefire.forkNumber}</test.build.dir>
<hadoop.tmp.dir>${hadoop.tmp.dir}/${surefire.forkNumber}</hadoop.tmp.dir>
<!-- Due to a Maven quirk, setting this to just -->
<!-- surefire.forkNumber won't do the parameter -->
<!-- substitution. Putting a prefix in front of it like -->
<!-- "fork-" makes it work. -->
<test.unique.fork.id>fork-000${surefire.forkNumber}</test.unique.fork.id>
<!-- Propagate scale parameters -->
<fs.s3a.scale.test.enabled>${fs.s3a.scale.test.enabled}</fs.s3a.scale.test.enabled>
<fs.s3a.scale.test.huge.filesize>${fs.s3a.scale.test.huge.filesize}</fs.s3a.scale.test.huge.filesize>
<fs.s3a.scale.test.huge.huge.partitionsize>${fs.s3a.scale.test.huge.partitionsize}</fs.s3a.scale.test.huge.huge.partitionsize>
<fs.s3a.scale.test.timeout>${fs.s3a.scale.test.timeout}</fs.s3a.scale.test.timeout>
<fs.s3a.directory.marker.retention>${fs.s3a.directory.marker.retention}</fs.s3a.directory.marker.retention>
<test.default.timeout>${test.integration.timeout}</test.default.timeout>
</systemPropertyVariables>
<!-- Some tests cannot run in parallel. Tests that cover -->
<!-- access to the root directory must run in isolation -->
<!-- from anything else that could modify the bucket. -->
<!-- S3A tests that cover multi-part upload must run in -->
<!-- isolation, because the file system is configured to -->
<!-- purge existing multi-part upload data on -->
<!-- initialization. MiniYARNCluster has not yet been -->
<!-- changed to handle parallel test execution gracefully. -->
<!-- Exclude all of these tests from parallel execution, -->
<!-- and instead run them sequentially in a separate -->
<!-- Surefire execution step later. -->
<includes>
<include>**/ITest*.java</include>
</includes>
<excludes>
<exclude>**/ITestS3AContractRootDir.java</exclude>
<exclude>**/ITestS3AFileContextStatistics.java</exclude>
<exclude>**/ITestS3AEncryptionSSEC*.java</exclude>
<exclude>**/ITestS3AHuge*.java</exclude>
<!-- Terasort MR jobs spawn enough processes that they use up all RAM -->
<exclude>**/ITestTerasort*.java</exclude>
<!-- Root marker tool tests -->
<exclude>**/ITestMarkerToolRootOperations.java</exclude>
<!-- leave this until the end for better statistics -->
<exclude>**/ITestAggregateIOStatistics.java</exclude>
</excludes>
</configuration>
</execution>
<execution>
<id>sequential-integration-tests</id>
<goals>
<goal>integration-test</goal>
<goal>verify</goal>
</goals>
<configuration>
<forkedProcessTimeoutInSeconds>${fs.s3a.scale.test.timeout}</forkedProcessTimeoutInSeconds>
<trimStackTrace>false</trimStackTrace>
<systemPropertyVariables>
<!-- Tell tests that they are being executed sequentially -->
<test.parallel.execution>false</test.parallel.execution>
<!-- Propagate scale parameters -->
<fs.s3a.scale.test.enabled>${fs.s3a.scale.test.enabled}</fs.s3a.scale.test.enabled>
<fs.s3a.scale.test.huge.filesize>${fs.s3a.scale.test.huge.filesize}</fs.s3a.scale.test.huge.filesize>
<fs.s3a.scale.test.huge.huge.partitionsize>${fs.s3a.scale.test.huge.partitionsize}</fs.s3a.scale.test.huge.huge.partitionsize>
<fs.s3a.scale.test.timeout>${fs.s3a.scale.test.timeout}</fs.s3a.scale.test.timeout>
<!-- Markers-->
<fs.s3a.directory.marker.retention>${fs.s3a.directory.marker.retention}</fs.s3a.directory.marker.retention>
<fs.s3a.directory.marker.audit>${fs.s3a.directory.marker.audit}</fs.s3a.directory.marker.audit>
</systemPropertyVariables>
<!-- Do a sequential run for tests that cannot handle -->
<!-- parallel execution. -->
<includes>
<include>**/ITestS3AFileContextStatistics.java</include>
<!-- large uploads consuming all bandwidth -->
<include>**/ITestS3AHuge*.java</include>
<!-- SSE encrypted files confuse everything else -->
<include>**/ITestS3AEncryptionSSEC*.java</include>
<!-- the terasort tests both work with a file in the same path in -->
<!-- the local FS. Running them sequentially guarantees isolation -->
<!-- and that they don't conflict with the other MR jobs for RAM -->
<include>**/ITestTerasort*.java</include>
<!-- Root marker tool tests -->
<!-- MUST be run before the other root ops so there's
more likelihood of files in the bucket -->
<include>**/ITestMarkerToolRootOperations.java</include>
<!-- operations on the root dir -->
<include>**/ITestS3AContractRootDir.java</include>
<!-- leave this until the end for better statistics -->
<include>**/ITestAggregateIOStatistics.java</include>
</includes>
</configuration>
</execution>
</executions>
</plugin>
</plugins>
</build>
</profile>
<profile>
<id>sequential-tests</id>
<activation>
<property>
<name>!parallel-tests</name>
</property>
</activation>
<build>
<plugins>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-failsafe-plugin</artifactId>
<executions>
<execution>
<goals>
<goal>integration-test</goal>
<goal>verify</goal>
</goals>
<configuration>
<systemPropertyVariables>
<!-- Propagate scale parameters -->
<fs.s3a.scale.test.enabled>${fs.s3a.scale.test.enabled}</fs.s3a.scale.test.enabled>
<fs.s3a.scale.test.huge.filesize>${fs.s3a.scale.test.huge.filesize}</fs.s3a.scale.test.huge.filesize>
<fs.s3a.scale.test.timeout>${fs.s3a.scale.test.timeout}</fs.s3a.scale.test.timeout>
<!-- Markers-->
<fs.s3a.directory.marker.retention>${fs.s3a.directory.marker.retention}</fs.s3a.directory.marker.retention>
<fs.s3a.directory.marker.audit>${fs.s3a.directory.marker.audit}</fs.s3a.directory.marker.audit>
</systemPropertyVariables>
<forkedProcessTimeoutInSeconds>${fs.s3a.scale.test.timeout}</forkedProcessTimeoutInSeconds>
</configuration>
</execution>
</executions>
</plugin>
</plugins>
</build>
</profile>
<!-- Turn on scale tests-->
<profile>
<id>scale</id>
<activation>
<property>
<name>scale</name>
</property>
</activation>
<properties >
<fs.s3a.scale.test.enabled>true</fs.s3a.scale.test.enabled>
</properties>
</profile>
<!-- Directory marker retention options, all from the -Dmarkers value-->
<profile>
<id>keep-markers</id>
<activation>
<property>
<name>markers</name>
<value>keep</value>
</property>
</activation>
<properties >
<fs.s3a.directory.marker.retention>keep</fs.s3a.directory.marker.retention>
</properties>
</profile>
<profile>
<id>delete-markers</id>
<activation>
<property>
<name>markers</name>
<value>delete</value>
</property>
</activation>
<properties >
<fs.s3a.directory.marker.retention>delete</fs.s3a.directory.marker.retention>
</properties>
</profile>
<profile>
<id>auth-markers</id>
<activation>
<property>
<name>markers</name>
<value>authoritative</value>
</property>
</activation>
<properties >
<fs.s3a.directory.marker.retention>authoritative</fs.s3a.directory.marker.retention>
</properties>
</profile>
</profiles>
<build>
<plugins>
<plugin>
<groupId>com.github.spotbugs</groupId>
<artifactId>spotbugs-maven-plugin</artifactId>
<configuration>
<xmlOutput>true</xmlOutput>
<excludeFilterFile>${basedir}/dev-support/findbugs-exclude.xml
</excludeFilterFile>
<effort>Max</effort>
</configuration>
</plugin>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-surefire-plugin</artifactId>
<configuration>
<forkedProcessTimeoutInSeconds>3600</forkedProcessTimeoutInSeconds>
<systemPropertyVariables>
<test.default.timeout>${test.integration.timeout}</test.default.timeout>
</systemPropertyVariables>
</configuration>
</plugin>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-dependency-plugin</artifactId>
<executions>
<execution>
<id>deplist1</id>
<phase>compile</phase>
<goals>
<goal>list</goal>
</goals>
<configuration>
<!-- build a shellprofile for hadoop-aws optional tools -->
<outputFile>${project.basedir}/target/hadoop-tools-deps/${project.artifactId}.tools-optional.txt</outputFile>
</configuration>
</execution>
<execution>
<id>copy</id>
<phase>test-compile</phase>
<goals>
<goal>copy-dependencies</goal>
</goals>
<configuration>
<includeScope>test</includeScope>
<includeTypes>so,dll,dylib</includeTypes>
<outputDirectory>${project.build.directory}/native-libs</outputDirectory>
</configuration>
</execution>
<execution>
<phase>package</phase>
<goals>
<goal>copy-dependencies</goal>
</goals>
<configuration>
<outputDirectory>${project.build.directory}/lib</outputDirectory>
</configuration>
</execution>
<execution>
<id>deplist2</id>
<phase>compile</phase>
<goals>
<goal>list</goal>
</goals>
<configuration>
<!-- referenced by the s3guard command -->
<outputFile>${project.basedir}/target/hadoop-tools-deps/${project.artifactId}.tools-builtin.txt</outputFile>
</configuration>
</execution>
</executions>
</plugin>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-enforcer-plugin</artifactId>
<executions>
<execution>
<id>banned-illegal-imports</id>
<phase>process-sources</phase>
<goals>
<goal>enforce</goal>
</goals>
<configuration>
<rules>
<restrictImports>
<includeTestCode>false</includeTestCode>
<reason>Restrict mapreduce imports to committer code</reason>
<exclusions>
<exclusion>org.apache.hadoop.fs.s3a.commit.AbstractS3ACommitter</exclusion>
<exclusion>org.apache.hadoop.fs.s3a.commit.AbstractS3ACommitterFactory</exclusion>
<exclusion>org.apache.hadoop.fs.s3a.commit.S3ACommitterFactory</exclusion>
<exclusion>org.apache.hadoop.fs.s3a.commit.impl.*</exclusion>
<exclusion>org.apache.hadoop.fs.s3a.commit.magic.*</exclusion>
<exclusion>org.apache.hadoop.fs.s3a.commit.staging.*</exclusion>
</exclusions>
<bannedImports>
<bannedImport>org.apache.hadoop.mapreduce.**</bannedImport>
<bannedImport>org.apache.hadoop.mapred.**</bannedImport>
</bannedImports>
</restrictImports>
</rules>
</configuration>
</execution>
</executions>
</plugin>
</plugins>
</build>
<dependencies>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-common</artifactId>
<scope>provided</scope>
<exclusions>
<exclusion>
<groupId>javax.servlet</groupId>
<artifactId>servlet-api</artifactId>
</exclusion>
<exclusion>
<groupId>javax.enterprise</groupId>
<artifactId>cdi-api</artifactId>
</exclusion>
</exclusions>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-common</artifactId>
<scope>test</scope>
<type>test-jar</type>
</dependency>
<dependency>
<groupId>com.amazonaws</groupId>
<artifactId>aws-java-sdk-bundle</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>org.assertj</groupId>
<artifactId>assertj-core</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.wildfly.openssl</groupId>
<artifactId>wildfly-openssl</artifactId>
<scope>runtime</scope>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.mockito</groupId>
<artifactId>mockito-core</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-mapreduce-client-core</artifactId>
<scope>provided</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-yarn-server-tests</artifactId>
<scope>test</scope>
<type>test-jar</type>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-mapreduce-client-hs</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-mapreduce-examples</artifactId>
<scope>test</scope>
<type>jar</type>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-hdfs</artifactId>
<scope>test</scope>
<type>test-jar</type>
<exclusions>
<exclusion>
<groupId>org.ow2.asm</groupId>
<artifactId>asm-commons</artifactId>
</exclusion>
</exclusions>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-distcp</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-distcp</artifactId>
<scope>test</scope>
<type>test-jar</type>
</dependency>
<!-- artifacts needed to bring up a Mini MR Yarn cluster-->
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-mapreduce-client-app</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-mapreduce-client-app</artifactId>
<type>test-jar</type>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-mapreduce-client-jobclient</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-mapreduce-client-jobclient</artifactId>
<scope>test</scope>
<type>test-jar</type>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-minikdc</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.hamcrest</groupId>
<artifactId>hamcrest-library</artifactId>
<scope>test</scope>
</dependency>
<!-- Used to create SSL certs for a secure Keystore -->
<dependency>
<groupId>org.bouncycastle</groupId>
<artifactId>bcprov-jdk15on</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.bouncycastle</groupId>
<artifactId>bcpkix-jdk15on</artifactId>
<scope>test</scope>
</dependency>
</dependencies>
</project>