From a1b538122cdee5aba6c494c4941cf5bb55182d17 Mon Sep 17 00:00:00 2001 From: Andy Bristol Date: Wed, 23 May 2018 10:37:57 -0700 Subject: [PATCH 01/12] [test] java tests for archive packaging (#30734) Ports the first couple tests for archive distributions from the old bats project to the new java project that includes windows platforms, consolidating them into one test method that tests that the distributions can be extracted and their contents verified. Includes the zip distributions which were not tested in the bats project. --- TESTING.asciidoc | 28 +- Vagrantfile | 1 + .../gradle/vagrant/VagrantTestPlugin.groovy | 23 +- qa/vagrant/build.gradle | 4 +- .../packaging/PackagingTests.java | 23 +- .../elasticsearch/packaging/VMTestRunner.java | 40 +++ .../packaging/test/ArchiveTestCase.java | 65 +++++ .../packaging/test/DefaultTarTests.java | 30 +++ .../packaging/test/DefaultZipTests.java | 30 +++ .../packaging/test/OssTarTests.java | 30 +++ .../packaging/test/OssZipTests.java | 30 +++ .../packaging/util/Archives.java | 239 ++++++++++++++++++ .../elasticsearch/packaging/util/Cleanup.java | 121 +++++++++ .../packaging/util/Distribution.java | 76 ++++++ .../packaging/util/FileMatcher.java | 137 ++++++++++ .../packaging/util/FileUtils.java | 134 ++++++++++ .../packaging/util/Installation.java | 58 +++++ .../packaging/util/Platforms.java | 68 +++++ .../elasticsearch/packaging/util/Shell.java | 193 ++++++++++++++ 19 files changed, 1313 insertions(+), 17 deletions(-) create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/VMTestRunner.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/test/ArchiveTestCase.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/test/DefaultTarTests.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/test/DefaultZipTests.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/test/OssTarTests.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/test/OssZipTests.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Archives.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Cleanup.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Distribution.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/util/FileMatcher.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/util/FileUtils.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Installation.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Platforms.java create mode 100644 qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Shell.java diff --git a/TESTING.asciidoc b/TESTING.asciidoc index 1e984a17f3c..5e5207b279e 100644 --- a/TESTING.asciidoc +++ b/TESTING.asciidoc @@ -512,7 +512,9 @@ into it vagrant ssh debian-9 -------------------------------------------- -Now inside the VM, to run the https://github.com/sstephenson/bats[bats] packaging tests +Now inside the VM, start the packaging tests from the terminal. There are two packaging +test projects. The old ones are written with https://github.com/sstephenson/bats[bats] +and only run on linux. To run them do -------------------------------------------- cd $PACKAGING_ARCHIVES @@ -524,18 +526,36 @@ sudo bats $BATS_TESTS/*.bats sudo bats $BATS_TESTS/20_tar_package.bats $BATS_TESTS/25_tar_plugins.bats -------------------------------------------- -To run the Java packaging tests, again inside the VM +The new packaging tests are written in Java and run on both linux and windows. On +linux (again, inside the VM) -------------------------------------------- -bash $PACKAGING_TESTS/run-tests.sh +# run the full suite +sudo bash $PACKAGING_TESTS/run-tests.sh + +# run specific test cases +sudo bash $PACKAGING_TESTS/run-tests.sh \ + org.elasticsearch.packaging.test.DefaultZipTests \ + org.elasticsearch.packaging.test.OssZipTests -------------------------------------------- -or on Windows +or on Windows, from a terminal running as Administrator -------------------------------------------- +# run the full suite powershell -File $Env:PACKAGING_TESTS/run-tests.ps1 + +# run specific test cases +powershell -File $Env:PACKAGING_TESTS/run-tests.ps1 ` + org.elasticsearch.packaging.test.DefaultZipTests ` + org.elasticsearch.packaging.test.OssZipTests -------------------------------------------- +Note that on Windows boxes when running from inside the GUI, you may have to log out and +back in to the `vagrant` user (password `vagrant`) for the environment variables that +locate the packaging tests and distributions to take effect, due to how vagrant provisions +Windows machines. + When you've made changes you want to test, keep the VM up and reload the tests and distributions inside by running (on the host) diff --git a/Vagrantfile b/Vagrantfile index 66ec60820ea..d53c80754e6 100644 --- a/Vagrantfile +++ b/Vagrantfile @@ -237,6 +237,7 @@ def linux_common(config, config.vm.provision 'markerfile', type: 'shell', inline: <<-SHELL touch /etc/is_vagrant_vm + touch /is_vagrant_vm # for consistency between linux and windows SHELL # This prevents leftovers from previous tests using the diff --git a/buildSrc/src/main/groovy/org/elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy b/buildSrc/src/main/groovy/org/elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy index 2e02911b7a7..6e8a5fd15ed 100644 --- a/buildSrc/src/main/groovy/org/elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy +++ b/buildSrc/src/main/groovy/org/elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy @@ -52,6 +52,8 @@ class VagrantTestPlugin implements Plugin { static final List DISTRIBUTIONS = unmodifiableList([ 'archives:tar', 'archives:oss-tar', + 'archives:zip', + 'archives:oss-zip', 'packages:rpm', 'packages:oss-rpm', 'packages:deb', @@ -242,13 +244,27 @@ class VagrantTestPlugin implements Plugin { Task createLinuxRunnerScript = project.tasks.create('createLinuxRunnerScript', FileContentsTask) { dependsOn copyPackagingTests file "${testsDir}/run-tests.sh" - contents "java -cp \"\$PACKAGING_TESTS/*\" org.junit.runner.JUnitCore ${-> project.extensions.esvagrant.testClass}" + contents """\ + if [ "\$#" -eq 0 ]; then + test_args=( "${-> project.extensions.esvagrant.testClass}" ) + else + test_args=( "\$@" ) + fi + java -cp "\$PACKAGING_TESTS/*" org.elasticsearch.packaging.VMTestRunner "\${test_args[@]}" + """ } Task createWindowsRunnerScript = project.tasks.create('createWindowsRunnerScript', FileContentsTask) { dependsOn copyPackagingTests file "${testsDir}/run-tests.ps1" + // the use of $args rather than param() here is deliberate because the syntax for array (multivalued) parameters is likely + // a little trappy for those unfamiliar with powershell contents """\ - java -cp "\$Env:PACKAGING_TESTS/*" org.junit.runner.JUnitCore ${-> project.extensions.esvagrant.testClass} + if (\$args.Count -eq 0) { + \$testArgs = @("${-> project.extensions.esvagrant.testClass}") + } else { + \$testArgs = \$args + } + java -cp "\$Env:PACKAGING_TESTS/*" org.elasticsearch.packaging.VMTestRunner @testArgs exit \$LASTEXITCODE """ } @@ -525,9 +541,10 @@ class VagrantTestPlugin implements Plugin { if (LINUX_BOXES.contains(box)) { javaPackagingTest.command = 'ssh' - javaPackagingTest.args = ['--command', 'bash "$PACKAGING_TESTS/run-tests.sh"'] + javaPackagingTest.args = ['--command', 'sudo bash "$PACKAGING_TESTS/run-tests.sh"'] } else { javaPackagingTest.command = 'winrm' + // winrm commands run as administrator javaPackagingTest.args = ['--command', 'powershell -File "$Env:PACKAGING_TESTS/run-tests.ps1"'] } diff --git a/qa/vagrant/build.gradle b/qa/vagrant/build.gradle index 52a6bb1efb5..a8dfe89b678 100644 --- a/qa/vagrant/build.gradle +++ b/qa/vagrant/build.gradle @@ -29,9 +29,9 @@ plugins { dependencies { compile "junit:junit:${versions.junit}" compile "org.hamcrest:hamcrest-core:${versions.hamcrest}" + compile "org.hamcrest:hamcrest-library:${versions.hamcrest}" - // needs to be on the classpath for JarHell - testRuntime project(':libs:elasticsearch-core') + compile project(':libs:elasticsearch-core') // pulls in the jar built by this project and its dependencies packagingTest project(path: project.path, configuration: 'runtime') diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/PackagingTests.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/PackagingTests.java index 0b5e7a3b6e0..fa7f8e8ef78 100644 --- a/qa/vagrant/src/main/java/org/elasticsearch/packaging/PackagingTests.java +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/PackagingTests.java @@ -19,13 +19,20 @@ package org.elasticsearch.packaging; -import org.junit.Test; +import org.elasticsearch.packaging.test.OssTarTests; +import org.elasticsearch.packaging.test.OssZipTests; +import org.elasticsearch.packaging.test.DefaultTarTests; +import org.elasticsearch.packaging.test.DefaultZipTests; -/** - * This class doesn't have any tests yet - */ -public class PackagingTests { +import org.junit.runner.RunWith; +import org.junit.runners.Suite; +import org.junit.runners.Suite.SuiteClasses; - @Test - public void testDummy() {} -} +@RunWith(Suite.class) +@SuiteClasses({ + DefaultTarTests.class, + DefaultZipTests.class, + OssTarTests.class, + OssZipTests.class +}) +public class PackagingTests {} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/VMTestRunner.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/VMTestRunner.java new file mode 100644 index 00000000000..a8fd2c27707 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/VMTestRunner.java @@ -0,0 +1,40 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging; + +import org.junit.runner.JUnitCore; + +import java.nio.file.Files; +import java.nio.file.Paths; + +/** + * Ensures that the current JVM is running on a virtual machine before delegating to {@link JUnitCore}. We just check for the existence + * of a special file that we create during VM provisioning. + */ +public class VMTestRunner { + public static void main(String[] args) { + if (Files.exists(Paths.get("/is_vagrant_vm"))) { + JUnitCore.main(args); + } else { + throw new RuntimeException("This filesystem does not have an expected marker file indicating it's a virtual machine. These " + + "tests should only run in a virtual machine because they're destructive."); + } + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/ArchiveTestCase.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/ArchiveTestCase.java new file mode 100644 index 00000000000..f683cb9c145 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/ArchiveTestCase.java @@ -0,0 +1,65 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.test; + +import org.junit.Before; +import org.junit.BeforeClass; +import org.junit.FixMethodOrder; +import org.junit.Test; +import org.junit.runners.MethodSorters; + +import org.elasticsearch.packaging.util.Distribution; +import org.elasticsearch.packaging.util.Installation; + +import static org.elasticsearch.packaging.util.Cleanup.cleanEverything; +import static org.elasticsearch.packaging.util.Archives.installArchive; +import static org.elasticsearch.packaging.util.Archives.verifyArchiveInstallation; +import static org.hamcrest.CoreMatchers.is; +import static org.junit.Assume.assumeThat; + +/** + * Tests that apply to the archive distributions (tar, zip). To add a case for a distribution, subclass and + * override {@link ArchiveTestCase#distribution()}. These tests should be the same across all archive distributions + */ +@FixMethodOrder(MethodSorters.NAME_ASCENDING) +public abstract class ArchiveTestCase { + + private static Installation installation; + + /** The {@link Distribution} that should be tested in this case */ + protected abstract Distribution distribution(); + + @BeforeClass + public static void cleanup() { + installation = null; + cleanEverything(); + } + + @Before + public void onlyCompatibleDistributions() { + assumeThat(distribution().packaging.compatible, is(true)); + } + + @Test + public void test10Install() { + installation = installArchive(distribution()); + verifyArchiveInstallation(installation, distribution()); + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/DefaultTarTests.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/DefaultTarTests.java new file mode 100644 index 00000000000..9b359a329c1 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/DefaultTarTests.java @@ -0,0 +1,30 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.test; + +import org.elasticsearch.packaging.util.Distribution; + +public class DefaultTarTests extends ArchiveTestCase { + + @Override + protected Distribution distribution() { + return Distribution.DEFAULT_TAR; + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/DefaultZipTests.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/DefaultZipTests.java new file mode 100644 index 00000000000..d9a6353a8c6 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/DefaultZipTests.java @@ -0,0 +1,30 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.test; + +import org.elasticsearch.packaging.util.Distribution; + +public class DefaultZipTests extends ArchiveTestCase { + + @Override + protected Distribution distribution() { + return Distribution.DEFAULT_ZIP; + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/OssTarTests.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/OssTarTests.java new file mode 100644 index 00000000000..86637fc9d48 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/OssTarTests.java @@ -0,0 +1,30 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.test; + +import org.elasticsearch.packaging.util.Distribution; + +public class OssTarTests extends ArchiveTestCase { + + @Override + protected Distribution distribution() { + return Distribution.OSS_TAR; + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/OssZipTests.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/OssZipTests.java new file mode 100644 index 00000000000..b6cd1e596a0 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/test/OssZipTests.java @@ -0,0 +1,30 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.test; + +import org.elasticsearch.packaging.util.Distribution; + +public class OssZipTests extends ArchiveTestCase { + + @Override + protected Distribution distribution() { + return Distribution.OSS_ZIP; + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Archives.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Archives.java new file mode 100644 index 00000000000..4a00570bf30 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Archives.java @@ -0,0 +1,239 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.util; + +import java.nio.file.Files; +import java.nio.file.Path; +import java.util.List; +import java.util.stream.Stream; + +import static org.elasticsearch.packaging.util.FileMatcher.Fileness.Directory; +import static org.elasticsearch.packaging.util.FileMatcher.Fileness.File; +import static org.elasticsearch.packaging.util.FileMatcher.file; +import static org.elasticsearch.packaging.util.FileMatcher.p644; +import static org.elasticsearch.packaging.util.FileMatcher.p660; +import static org.elasticsearch.packaging.util.FileMatcher.p755; +import static org.elasticsearch.packaging.util.FileUtils.getCurrentVersion; +import static org.elasticsearch.packaging.util.FileUtils.getDefaultArchiveInstallPath; +import static org.elasticsearch.packaging.util.FileUtils.getPackagingArchivesDir; +import static org.elasticsearch.packaging.util.FileUtils.lsGlob; + +import static org.elasticsearch.packaging.util.FileUtils.mv; +import static org.elasticsearch.packaging.util.Platforms.isDPKG; +import static org.hamcrest.MatcherAssert.assertThat; +import static org.hamcrest.core.Is.is; +import static org.hamcrest.collection.IsEmptyCollection.empty; +import static org.hamcrest.collection.IsCollectionWithSize.hasSize; + +/** + * Installation and verification logic for archive distributions + */ +public class Archives { + + public static Installation installArchive(Distribution distribution) { + return installArchive(distribution, getDefaultArchiveInstallPath(), getCurrentVersion()); + } + + public static Installation installArchive(Distribution distribution, Path fullInstallPath, String version) { + final Shell sh = new Shell(); + + final Path distributionFile = getPackagingArchivesDir().resolve(distribution.filename(version)); + final Path baseInstallPath = fullInstallPath.getParent(); + final Path extractedPath = baseInstallPath.resolve("elasticsearch-" + version); + + assertThat("distribution file must exist", Files.exists(distributionFile), is(true)); + assertThat("elasticsearch must not already be installed", lsGlob(baseInstallPath, "elasticsearch*"), empty()); + + if (distribution.packaging == Distribution.Packaging.TAR) { + + if (Platforms.LINUX) { + sh.run("tar", "-C", baseInstallPath.toString(), "-xzpf", distributionFile.toString()); + } else { + throw new RuntimeException("Distribution " + distribution + " is not supported on windows"); + } + + } else if (distribution.packaging == Distribution.Packaging.ZIP) { + + if (Platforms.LINUX) { + sh.run("unzip", distributionFile.toString(), "-d", baseInstallPath.toString()); + } else { + sh.run("powershell.exe", "-Command", + "Add-Type -AssemblyName 'System.IO.Compression.Filesystem'; " + + "[IO.Compression.ZipFile]::ExtractToDirectory('" + distributionFile + "', '" + baseInstallPath + "')"); + } + + } else { + throw new RuntimeException("Distribution " + distribution + " is not a known archive type"); + } + + assertThat("archive was extracted", Files.exists(extractedPath), is(true)); + + mv(extractedPath, fullInstallPath); + + assertThat("extracted archive moved to install location", Files.exists(fullInstallPath)); + final List installations = lsGlob(baseInstallPath, "elasticsearch*"); + assertThat("only the intended installation exists", installations, hasSize(1)); + assertThat("only the intended installation exists", installations.get(0), is(fullInstallPath)); + + if (Platforms.LINUX) { + setupArchiveUsersLinux(fullInstallPath); + } + + return new Installation(fullInstallPath); + } + + private static void setupArchiveUsersLinux(Path installPath) { + final Shell sh = new Shell(); + + if (sh.runIgnoreExitCode("getent", "group", "elasticsearch").isSuccess() == false) { + if (isDPKG()) { + sh.run("addgroup", "--system", "elasticsearch"); + } else { + sh.run("groupadd", "-r", "elasticsearch"); + } + } + + if (sh.runIgnoreExitCode("id", "elasticsearch").isSuccess() == false) { + if (isDPKG()) { + sh.run("adduser", + "--quiet", + "--system", + "--no-create-home", + "--ingroup", "elasticsearch", + "--disabled-password", + "--shell", "/bin/false", + "elasticsearch"); + } else { + sh.run("useradd", + "--system", + "-M", + "--gid", "elasticsearch", + "--shell", "/sbin/nologin", + "--comment", "elasticsearch user", + "elasticsearch"); + } + } + sh.run("chown", "-R", "elasticsearch:elasticsearch", installPath.toString()); + } + + public static void verifyArchiveInstallation(Installation installation, Distribution distribution) { + // on Windows for now we leave the installation owned by the vagrant user that the tests run as. Since the vagrant account + // is a local administrator, the files really end up being owned by the local administrators group. In the future we'll + // install and run elasticesearch with a role user on Windows + final String owner = Platforms.WINDOWS + ? "BUILTIN\\Administrators" + : "elasticsearch"; + + verifyOssInstallation(installation, distribution, owner); + if (distribution.flavor == Distribution.Flavor.DEFAULT) { + verifyDefaultInstallation(installation, distribution, owner); + } + } + + private static void verifyOssInstallation(Installation es, Distribution distribution, String owner) { + Stream.of( + es.home, + es.config, + es.plugins, + es.modules, + es.logs + ).forEach(dir -> assertThat(dir, file(Directory, owner, owner, p755))); + + assertThat(Files.exists(es.data), is(false)); + assertThat(Files.exists(es.scripts), is(false)); + + assertThat(es.home.resolve("bin"), file(Directory, owner, owner, p755)); + assertThat(es.home.resolve("lib"), file(Directory, owner, owner, p755)); + assertThat(Files.exists(es.config.resolve("elasticsearch.keystore")), is(false)); + + Stream.of( + "bin/elasticsearch", + "bin/elasticsearch-env", + "bin/elasticsearch-keystore", + "bin/elasticsearch-plugin", + "bin/elasticsearch-translog" + ).forEach(executable -> { + + assertThat(es.home.resolve(executable), file(File, owner, owner, p755)); + + if (distribution.packaging == Distribution.Packaging.ZIP) { + assertThat(es.home.resolve(executable + ".bat"), file(File, owner)); + } + }); + + if (distribution.packaging == Distribution.Packaging.ZIP) { + Stream.of( + "bin/elasticsearch-service.bat", + "bin/elasticsearch-service-mgr.exe", + "bin/elasticsearch-service-x64.exe" + ).forEach(executable -> assertThat(es.home.resolve(executable), file(File, owner))); + } + + Stream.of( + "elasticsearch.yml", + "jvm.options", + "log4j2.properties" + ).forEach(config -> assertThat(es.config.resolve(config), file(File, owner, owner, p660))); + + Stream.of( + "NOTICE.txt", + "LICENSE.txt", + "README.textile" + ).forEach(doc -> assertThat(es.home.resolve(doc), file(File, owner, owner, p644))); + } + + private static void verifyDefaultInstallation(Installation es, Distribution distribution, String owner) { + + Stream.of( + "bin/elasticsearch-certgen", + "bin/elasticsearch-certutil", + "bin/elasticsearch-croneval", + "bin/elasticsearch-migrate", + "bin/elasticsearch-saml-metadata", + "bin/elasticsearch-setup-passwords", + "bin/elasticsearch-sql-cli", + "bin/elasticsearch-syskeygen", + "bin/elasticsearch-users", + "bin/x-pack-env", + "bin/x-pack-security-env", + "bin/x-pack-watcher-env" + ).forEach(executable -> { + + assertThat(es.home.resolve(executable), file(File, owner, owner, p755)); + + if (distribution.packaging == Distribution.Packaging.ZIP) { + assertThat(es.home.resolve(executable + ".bat"), file(File, owner)); + } + }); + + // at this time we only install the current version of archive distributions, but if that changes we'll need to pass + // the version through here + assertThat(es.home.resolve("bin/elasticsearch-sql-cli-" + getCurrentVersion() + ".jar"), file(File, owner, owner, p755)); + + Stream.of( + "users", + "users_roles", + "roles.yml", + "role_mapping.yml", + "log4j2.properties" + ).forEach(config -> assertThat(es.config.resolve(config), file(File, owner, owner, p660))); + } + +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Cleanup.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Cleanup.java new file mode 100644 index 00000000000..9e9150c9c18 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Cleanup.java @@ -0,0 +1,121 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.util; + +import java.nio.file.Files; +import java.nio.file.Paths; +import java.util.Arrays; +import java.util.Collections; +import java.util.List; + +import static org.elasticsearch.packaging.util.FileUtils.getTempDir; +import static org.elasticsearch.packaging.util.FileUtils.lsGlob; +import static org.elasticsearch.packaging.util.Platforms.isAptGet; +import static org.elasticsearch.packaging.util.Platforms.isDPKG; +import static org.elasticsearch.packaging.util.Platforms.isRPM; +import static org.elasticsearch.packaging.util.Platforms.isSystemd; +import static org.elasticsearch.packaging.util.Platforms.isYUM; + +public class Cleanup { + + private static final List ELASTICSEARCH_FILES_LINUX = Arrays.asList( + "/usr/share/elasticsearch", + "/etc/elasticsearch", + "/var/lib/elasticsearch", + "/var/log/elasticsearch", + "/etc/default/elasticsearch", + "/etc/sysconfig/elasticsearch", + "/var/run/elasticsearch", + "/usr/share/doc/elasticsearch", + "/usr/lib/systemd/system/elasticsearch.conf", + "/usr/lib/tmpfiles.d/elasticsearch.conf", + "/usr/lib/sysctl.d/elasticsearch.conf" + ); + + // todo + private static final List ELASTICSEARCH_FILES_WINDOWS = Collections.emptyList(); + + public static void cleanEverything() { + final Shell sh = new Shell(); + + // kill elasticsearch processes + if (Platforms.WINDOWS) { + + // the view of processes returned by Get-Process doesn't expose command line arguments, so we use WMI here + sh.runIgnoreExitCode("powershell.exe", "-Command", + "Get-WmiObject Win32_Process | " + + "Where-Object { $_.CommandLine -Match 'org.elasticsearch.bootstrap.Elasticsearch' } | " + + "ForEach-Object { $_.Terminate() }"); + + } else { + + sh.runIgnoreExitCode("pkill", "-u", "elasticsearch"); + sh.runIgnoreExitCode("bash", "-c", + "ps aux | grep -i 'org.elasticsearch.bootstrap.Elasticsearch' | awk {'print $2'} | xargs kill -9"); + + } + + if (Platforms.LINUX) { + purgePackagesLinux(); + } + + // remove elasticsearch users + if (Platforms.LINUX) { + sh.runIgnoreExitCode("userdel", "elasticsearch"); + sh.runIgnoreExitCode("groupdel", "elasticsearch"); + } + + // delete files that may still exist + lsGlob(getTempDir(), "elasticsearch*").forEach(FileUtils::rm); + final List filesToDelete = Platforms.WINDOWS + ? ELASTICSEARCH_FILES_WINDOWS + : ELASTICSEARCH_FILES_LINUX; + filesToDelete.stream() + .map(Paths::get) + .filter(Files::exists) + .forEach(FileUtils::rm); + + // disable elasticsearch service + // todo add this for windows when adding tests for service intallation + if (Platforms.LINUX && isSystemd()) { + sh.run("systemctl", "unmask", "systemd-sysctl.service"); + } + } + + private static void purgePackagesLinux() { + final Shell sh = new Shell(); + + if (isRPM()) { + sh.runIgnoreExitCode("rpm", "--quiet", "-e", "elasticsearch", "elasticsearch-oss"); + } + + if (isYUM()) { + sh.runIgnoreExitCode("yum", "remove", "-y", "elasticsearch", "elasticsearch-oss"); + } + + if (isDPKG()) { + sh.runIgnoreExitCode("dpkg", "--purge", "elasticsearch", "elasticsearch-oss"); + } + + if (isAptGet()) { + sh.runIgnoreExitCode("apt-get", "--quiet", "--yes", "purge", "elasticsearch", "elasticsearch-oss"); + } + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Distribution.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Distribution.java new file mode 100644 index 00000000000..4f0c8751ca4 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Distribution.java @@ -0,0 +1,76 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.util; + +public enum Distribution { + + OSS_TAR(Packaging.TAR, Flavor.OSS), + OSS_ZIP(Packaging.ZIP, Flavor.OSS), + OSS_DEB(Packaging.DEB, Flavor.OSS), + OSS_RPM(Packaging.RPM, Flavor.OSS), + + DEFAULT_TAR(Packaging.TAR, Flavor.DEFAULT), + DEFAULT_ZIP(Packaging.ZIP, Flavor.DEFAULT), + DEFAULT_DEB(Packaging.DEB, Flavor.DEFAULT), + DEFAULT_RPM(Packaging.RPM, Flavor.DEFAULT); + + public final Packaging packaging; + public final Flavor flavor; + + Distribution(Packaging packaging, Flavor flavor) { + this.packaging = packaging; + this.flavor = flavor; + } + + public String filename(String version) { + return flavor.name + "-" + version + packaging.extension; + } + + public enum Packaging { + + TAR(".tar.gz", Platforms.LINUX), + ZIP(".zip", true), + DEB(".deb", Platforms.isDPKG()), + RPM(".rpm", Platforms.isRPM()); + + /** The extension of this distribution's file */ + public final String extension; + + /** Whether the distribution is intended for use on the platform the current JVM is running on */ + public final boolean compatible; + + Packaging(String extension, boolean compatible) { + this.extension = extension; + this.compatible = compatible; + } + } + + public enum Flavor { + + OSS("elasticsearch-oss"), + DEFAULT("elasticsearch"); + + public final String name; + + Flavor(String name) { + this.name = name; + } + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/FileMatcher.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/FileMatcher.java new file mode 100644 index 00000000000..9fdf6d60081 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/FileMatcher.java @@ -0,0 +1,137 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.util; + +import org.hamcrest.Description; +import org.hamcrest.TypeSafeMatcher; + +import java.nio.file.Files; +import java.nio.file.Path; +import java.nio.file.attribute.BasicFileAttributes; +import java.nio.file.attribute.PosixFileAttributes; +import java.nio.file.attribute.PosixFilePermission; +import java.util.Objects; +import java.util.Set; + +import static org.elasticsearch.packaging.util.FileUtils.getBasicFileAttributes; +import static org.elasticsearch.packaging.util.FileUtils.getFileOwner; +import static org.elasticsearch.packaging.util.FileUtils.getPosixFileAttributes; + +import static java.nio.file.attribute.PosixFilePermissions.fromString; + +/** + * Asserts that a file at a path matches its status as Directory/File, and its owner. If on a posix system, also matches the permission + * set is what we expect. + * + * This class saves information about its failed matches in instance variables and so instances should not be reused + */ +public class FileMatcher extends TypeSafeMatcher { + + public enum Fileness { File, Directory } + + public static final Set p755 = fromString("rwxr-xr-x"); + public static final Set p660 = fromString("rw-rw----"); + public static final Set p644 = fromString("rw-r--r--"); + + private final Fileness fileness; + private final String owner; + private final String group; + private final Set posixPermissions; + + private String mismatch; + + public FileMatcher(Fileness fileness, String owner, String group, Set posixPermissions) { + this.fileness = Objects.requireNonNull(fileness); + this.owner = Objects.requireNonNull(owner); + this.group = group; + this.posixPermissions = posixPermissions; + } + + @Override + protected boolean matchesSafely(Path path) { + if (Files.exists(path) == false) { + mismatch = "Does not exist"; + return false; + } + + if (Platforms.WINDOWS) { + final BasicFileAttributes attributes = getBasicFileAttributes(path); + final String attributeViewOwner = getFileOwner(path); + + if (fileness.equals(Fileness.Directory) != attributes.isDirectory()) { + mismatch = "Is " + (attributes.isDirectory() ? "a directory" : "a file"); + return false; + } + + if (attributeViewOwner.contains(owner) == false) { + mismatch = "Owned by " + attributeViewOwner; + return false; + } + } else { + final PosixFileAttributes attributes = getPosixFileAttributes(path); + + if (fileness.equals(Fileness.Directory) != attributes.isDirectory()) { + mismatch = "Is " + (attributes.isDirectory() ? "a directory" : "a file"); + return false; + } + + if (owner.equals(attributes.owner().getName()) == false) { + mismatch = "Owned by " + attributes.owner().getName(); + return false; + } + + if (group != null && group.equals(attributes.group().getName()) == false) { + mismatch = "Owned by group " + attributes.group().getName(); + return false; + } + + if (posixPermissions != null && posixPermissions.equals(attributes.permissions()) == false) { + mismatch = "Has permissions " + attributes.permissions(); + return false; + } + } + + return true; + } + + @Override + public void describeMismatchSafely(Path path, Description description) { + description.appendText("path ").appendValue(path); + if (mismatch != null) { + description.appendText(mismatch); + } + } + + @Override + public void describeTo(Description description) { + description.appendValue("file/directory: ").appendValue(fileness) + .appendText(" with owner ").appendValue(owner) + .appendText(" with group ").appendValue(group) + .appendText(" with posix permissions ").appendValueList("[", ",", "]", posixPermissions); + } + + public static FileMatcher file(Fileness fileness, String owner) { + return file(fileness, owner, null, null); + } + + public static FileMatcher file(Fileness fileness, String owner, String group, Set permissions) { + return new FileMatcher(fileness, owner, group, permissions); + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/FileUtils.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/FileUtils.java new file mode 100644 index 00000000000..ad826675244 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/FileUtils.java @@ -0,0 +1,134 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.util; + +import org.elasticsearch.core.internal.io.IOUtils; + +import java.io.IOException; +import java.nio.file.DirectoryStream; +import java.nio.file.Files; +import java.nio.file.Path; +import java.nio.file.Paths; +import java.nio.file.attribute.BasicFileAttributes; +import java.nio.file.attribute.FileOwnerAttributeView; +import java.nio.file.attribute.PosixFileAttributes; +import java.util.ArrayList; +import java.util.List; + +import static org.hamcrest.MatcherAssert.assertThat; +import static org.hamcrest.core.IsNot.not; +import static org.hamcrest.text.IsEmptyString.isEmptyOrNullString; + +/** + * Wrappers and convenience methods for common filesystem operations + */ +public class FileUtils { + + public static List lsGlob(Path directory, String glob) { + List paths = new ArrayList<>(); + try (DirectoryStream stream = Files.newDirectoryStream(directory, glob)) { + + for (Path path : stream) { + paths.add(path); + } + return paths; + + } catch (IOException e) { + throw new RuntimeException(e); + } + } + + public static void rm(Path... paths) { + try { + IOUtils.rm(paths); + } catch (IOException e) { + throw new RuntimeException(e); + } + } + + public static Path mv(Path source, Path target) { + try { + return Files.move(source, target); + } catch (IOException e) { + throw new RuntimeException(e); + } + } + + public static String slurp(Path file) { + try { + return String.join("\n", Files.readAllLines(file)); + } catch (IOException e) { + throw new RuntimeException(e); + } + } + + /** + * Gets the owner of a file in a way that should be supported by all filesystems that have a concept of file owner + */ + public static String getFileOwner(Path path) { + try { + FileOwnerAttributeView view = Files.getFileAttributeView(path, FileOwnerAttributeView.class); + return view.getOwner().getName(); + } catch (IOException e) { + throw new RuntimeException(e); + } + } + + /** + * Gets attributes that are supported by all filesystems + */ + public static BasicFileAttributes getBasicFileAttributes(Path path) { + try { + return Files.readAttributes(path, BasicFileAttributes.class); + } catch (IOException e) { + throw new RuntimeException(e); + } + } + + /** + * Gets attributes that are supported by posix filesystems + */ + public static PosixFileAttributes getPosixFileAttributes(Path path) { + try { + return Files.readAttributes(path, PosixFileAttributes.class); + } catch (IOException e) { + throw new RuntimeException(e); + } + } + + // vagrant creates /tmp for us in windows so we use that to avoid long paths + public static Path getTempDir() { + return Paths.get("/tmp"); + } + + public static Path getDefaultArchiveInstallPath() { + return getTempDir().resolve("elasticsearch"); + } + + public static String getCurrentVersion() { + return slurp(getPackagingArchivesDir().resolve("version")); + } + + public static Path getPackagingArchivesDir() { + String fromEnv = System.getenv("PACKAGING_ARCHIVES"); + assertThat(fromEnv, not(isEmptyOrNullString())); + return Paths.get(fromEnv); + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Installation.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Installation.java new file mode 100644 index 00000000000..d231762d062 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Installation.java @@ -0,0 +1,58 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.util; + +import java.nio.file.Path; + +/** + * Represents an installation of Elasticsearch + */ +public class Installation { + + public final Path home; + public final Path config; + public final Path data; + public final Path logs; + public final Path plugins; + public final Path modules; + public final Path scripts; + + public Installation(Path home, Path config, Path data, Path logs, Path plugins, Path modules, Path scripts) { + this.home = home; + this.config = config; + this.data = data; + this.logs = logs; + this.plugins = plugins; + this.modules = modules; + this.scripts = scripts; + } + + public Installation(Path home) { + this( + home, + home.resolve("config"), + home.resolve("data"), + home.resolve("logs"), + home.resolve("plugins"), + home.resolve("modules"), + home.resolve("scripts") + ); + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Platforms.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Platforms.java new file mode 100644 index 00000000000..230af8efc2d --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Platforms.java @@ -0,0 +1,68 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.util; + +public class Platforms { + public static final String OS_NAME = System.getProperty("os.name"); + public static final boolean LINUX = OS_NAME.startsWith("Linux"); + public static final boolean WINDOWS = OS_NAME.startsWith("Windows"); + + public static boolean isDPKG() { + if (WINDOWS) { + return false; + } + return new Shell().runIgnoreExitCode("which", "dpkg").isSuccess(); + } + + public static boolean isAptGet() { + if (WINDOWS) { + return false; + } + return new Shell().runIgnoreExitCode("which", "apt-get").isSuccess(); + } + + public static boolean isRPM() { + if (WINDOWS) { + return false; + } + return new Shell().runIgnoreExitCode("which", "rpm").isSuccess(); + } + + public static boolean isYUM() { + if (WINDOWS) { + return false; + } + return new Shell().runIgnoreExitCode("which", "yum").isSuccess(); + } + + public static boolean isSystemd() { + if (WINDOWS) { + return false; + } + return new Shell().runIgnoreExitCode("which", "systemctl").isSuccess(); + } + + public static boolean isSysVInit() { + if (WINDOWS) { + return false; + } + return new Shell().runIgnoreExitCode("which", "service").isSuccess(); + } +} diff --git a/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Shell.java b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Shell.java new file mode 100644 index 00000000000..3adc0b62e04 --- /dev/null +++ b/qa/vagrant/src/main/java/org/elasticsearch/packaging/util/Shell.java @@ -0,0 +1,193 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.packaging.util; + +import org.elasticsearch.common.SuppressForbidden; + +import java.io.BufferedReader; +import java.io.IOException; +import java.io.InputStream; +import java.io.InputStreamReader; +import java.nio.file.Path; +import java.util.HashMap; +import java.util.Map; +import java.util.Objects; + +import static java.util.Collections.emptyMap; + +/** + * Wrapper to run shell commands and collect their outputs in a less verbose way + */ +public class Shell { + + final Map env; + final Path workingDirectory; + + public Shell() { + this(emptyMap(), null); + } + + public Shell(Map env) { + this(env, null); + } + + public Shell(Path workingDirectory) { + this(emptyMap(), workingDirectory); + } + + public Shell(Map env, Path workingDirectory) { + this.env = new HashMap<>(env); + this.workingDirectory = workingDirectory; + } + + public Result run(String... command) { + Result result = runIgnoreExitCode(command); + if (result.isSuccess() == false) { + throw new RuntimeException("Command was not successful: [" + String.join(" ", command) + "] result: " + result.toString()); + } + return result; + } + + public Result runIgnoreExitCode(String... command) { + ProcessBuilder builder = new ProcessBuilder(); + builder.command(command); + + if (workingDirectory != null) { + setWorkingDirectory(builder, workingDirectory); + } + + if (env != null && env.isEmpty() == false) { + for (Map.Entry entry : env.entrySet()) { + builder.environment().put(entry.getKey(), entry.getValue()); + } + } + + try { + + Process process = builder.start(); + + StringBuilder stdout = new StringBuilder(); + StringBuilder stderr = new StringBuilder(); + + Thread stdoutThread = new Thread(new StreamCollector(process.getInputStream(), stdout)); + Thread stderrThread = new Thread(new StreamCollector(process.getErrorStream(), stderr)); + + stdoutThread.start(); + stderrThread.start(); + + stdoutThread.join(); + stderrThread.join(); + + int exitCode = process.waitFor(); + + return new Result(exitCode, stdout.toString(), stderr.toString()); + + } catch (IOException | InterruptedException e) { + throw new RuntimeException(e); + } + } + + @SuppressForbidden(reason = "ProcessBuilder expects java.io.File") + private static void setWorkingDirectory(ProcessBuilder builder, Path path) { + builder.directory(path.toFile()); + } + + public String toString() { + return new StringBuilder() + .append("<") + .append(this.getClass().getName()) + .append(" ") + .append("env = [") + .append(env) + .append("]") + .append("workingDirectory = [") + .append(workingDirectory) + .append("]") + .append(">") + .toString(); + } + + public static class Result { + public final int exitCode; + public final String stdout; + public final String stderr; + + public Result(int exitCode, String stdout, String stderr) { + this.exitCode = exitCode; + this.stdout = stdout; + this.stderr = stderr; + } + + public boolean isSuccess() { + return exitCode == 0; + } + + public String toString() { + return new StringBuilder() + .append("<") + .append(this.getClass().getName()) + .append(" ") + .append("exitCode = [") + .append(exitCode) + .append("]") + .append(" ") + .append("stdout = [") + .append(stdout) + .append("]") + .append(" ") + .append("stderr = [") + .append(stderr) + .append("]") + .append(">") + .toString(); + } + } + + private static class StreamCollector implements Runnable { + private final InputStream input; + private final Appendable appendable; + + StreamCollector(InputStream input, Appendable appendable) { + this.input = Objects.requireNonNull(input); + this.appendable = Objects.requireNonNull(appendable); + } + + public void run() { + try { + + BufferedReader reader = new BufferedReader(reader(input)); + String line; + + while ((line = reader.readLine()) != null) { + appendable.append(line); + appendable.append("\n"); + } + + } catch (IOException e) { + throw new RuntimeException(e); + } + } + + @SuppressForbidden(reason = "the system's default character set is a best guess of what subprocesses will use") + private static InputStreamReader reader(InputStream inputStream) { + return new InputStreamReader(inputStream); + } + } +} From 4b6915976ce7e3d688711c49179a85942716dbab Mon Sep 17 00:00:00 2001 From: Igor Motov Date: Wed, 23 May 2018 15:15:19 -0400 Subject: [PATCH 02/12] Add support for indexed shape routing in geo_shape query (#30760) Adds ability to specify the routing value for the indexed shape in the geo_shape query. Closes #7663 --- .../query-dsl/geo-shape-query.asciidoc | 1 + .../index/query/GeoShapeQueryBuilder.java | 47 ++++++++++++++++++- .../query/GeoShapeQueryBuilderTests.java | 7 +++ .../search/geo/GeoShapeIntegrationIT.java | 38 +++++++++++++++ 4 files changed, 92 insertions(+), 1 deletion(-) diff --git a/docs/reference/query-dsl/geo-shape-query.asciidoc b/docs/reference/query-dsl/geo-shape-query.asciidoc index 08b504951e1..4e00a2f49b4 100644 --- a/docs/reference/query-dsl/geo-shape-query.asciidoc +++ b/docs/reference/query-dsl/geo-shape-query.asciidoc @@ -93,6 +93,7 @@ to 'shapes'. * `type` - Index type where the pre-indexed shape is. * `path` - The field specified as path containing the pre-indexed shape. Defaults to 'shape'. +* `routing` - The routing of the shape document if required. The following is an example of using the Filter with a pre-indexed shape: diff --git a/server/src/main/java/org/elasticsearch/index/query/GeoShapeQueryBuilder.java b/server/src/main/java/org/elasticsearch/index/query/GeoShapeQueryBuilder.java index d7d9797dad8..03fb2098180 100644 --- a/server/src/main/java/org/elasticsearch/index/query/GeoShapeQueryBuilder.java +++ b/server/src/main/java/org/elasticsearch/index/query/GeoShapeQueryBuilder.java @@ -29,6 +29,7 @@ import org.apache.lucene.spatial.prefix.RecursivePrefixTreeStrategy; import org.apache.lucene.spatial.query.SpatialArgs; import org.apache.lucene.spatial.query.SpatialOperation; import org.apache.lucene.util.SetOnce; +import org.elasticsearch.Version; import org.elasticsearch.action.ActionListener; import org.elasticsearch.action.get.GetRequest; import org.elasticsearch.action.get.GetResponse; @@ -77,6 +78,7 @@ public class GeoShapeQueryBuilder extends AbstractQueryBuilder supplier = new SetOnce<>(); queryRewriteContext.registerAsyncAction((client, listener) -> { GetRequest getRequest = new GetRequest(indexedShapeIndex, indexedShapeType, indexedShapeId); + getRequest.routing(indexedShapeRouting); fetch(client, getRequest, indexedShapePath, ActionListener.wrap(builder-> { supplier.set(builder); listener.onResponse(null); diff --git a/server/src/test/java/org/elasticsearch/index/query/GeoShapeQueryBuilderTests.java b/server/src/test/java/org/elasticsearch/index/query/GeoShapeQueryBuilderTests.java index 3282077ba6a..6356b2122ed 100644 --- a/server/src/test/java/org/elasticsearch/index/query/GeoShapeQueryBuilderTests.java +++ b/server/src/test/java/org/elasticsearch/index/query/GeoShapeQueryBuilderTests.java @@ -59,6 +59,7 @@ public class GeoShapeQueryBuilderTests extends AbstractQueryTestCase Date: Wed, 23 May 2018 13:36:58 -0700 Subject: [PATCH 03/12] Painless: Types Section Clean Up (#30283) Clean up of types section, casting section, and a large number of examples. --- docs/painless/painless-casting.asciidoc | 532 +++++++++++++---- docs/painless/painless-comments.asciidoc | 12 +- docs/painless/painless-identifiers.asciidoc | 8 +- docs/painless/painless-keywords.asciidoc | 6 +- docs/painless/painless-lang-spec.asciidoc | 2 +- docs/painless/painless-literals.asciidoc | 52 +- docs/painless/painless-operators.asciidoc | 47 +- docs/painless/painless-types.asciidoc | 599 +++++++++++++------- docs/painless/painless-variables.asciidoc | 165 ++++-- 9 files changed, 998 insertions(+), 425 deletions(-) diff --git a/docs/painless/painless-casting.asciidoc b/docs/painless/painless-casting.asciidoc index ec4f9919bd0..a3624f90831 100644 --- a/docs/painless/painless-casting.asciidoc +++ b/docs/painless/painless-casting.asciidoc @@ -1,172 +1,456 @@ [[painless-casting]] === Casting -Casting is the conversion of one type to another. Implicit casts are casts that -occur automatically, such as during an assignment operation. Explicit casts are -casts where you use the casting operator to explicitly convert one type to -another. This is necessary during operations where the cast cannot be inferred. +A cast converts the value of an original type to the equivalent value of a +target type. An implicit cast infers the target type and automatically occurs +during certain <>. An explicit cast specifies +the target type and forcefully occurs as its own operation. Use the *cast +operator* to specify an explicit cast. -To cast to a new type, precede the expression by the new type enclosed in -parentheses, for example -`(int)x`. +*Errors* -The following sections specify the implicit casts that can be performed and the -explicit casts that are allowed. The only other permitted cast is casting -a single character `String` to a `char`. +* If during a cast there exists no equivalent value for the target type. +* If an implicit cast is given, but an explicit cast is required. -*Grammar:* +*Grammar* [source,ANTLR4] ---- cast: '(' TYPE ')' expression ---- -[[numeric-casting]] -==== Numeric Casting +*Examples* -The following table shows the allowed implicit and explicit casts between -numeric types. Read the table by row. To find out if you need to explicitly -cast from type A to type B, find the row for type A and scan across to the -column for type B. +* Valid casts. ++ +[source,Painless] +---- +<1> int i = (int)5L; +<2> Map m = new HashMap(); +<3> HashMap hm = (HashMap)m; +---- ++ +<1> declare `int i`; + explicit cast `long 5` to `int 5` -> `int 5`; + assign `int 5` to `i` +<2> declare `Map m`; + allocate `HashMap` instance -> `HashMap reference`; + implicit cast `HashMap reference` to `Map reference` -> `Map reference`; + assign `Map reference` to `m` +<3> declare `HashMap hm`; + access `m` -> `Map reference`; + explicit cast `Map reference` to `HashMap reference` -> `HashMap reference`; + assign `HashMap reference` to `hm` -IMPORTANT: Explicit casts between numeric types can result in some data loss. A -smaller numeric type cannot necessarily accommodate the value from a larger -numeric type. You might also lose precision when casting from integer types -to floating point types. +[[numeric-type-casting]] +==== Numeric Type Casting + +A <> cast converts the value of an original +numeric type to the equivalent value of a target numeric type. A cast between +two numeric type values results in data loss when the value of the original +numeric type is larger than the target numeric type can accommodate. A cast +between an integer type value and a floating point type value can result in +precision loss. + +The allowed casts for values of each numeric type are shown as a row in the +following table: |==== -| | byte | short | char | int | long | float | double -| byte | | implicit | implicit | implicit | implicit | implicit | implicit -| short | explicit | | explicit | implicit | implicit | implicit | implicit -| char | explicit | explicit | | implicit | implicit | implicit | implicit -| int | explicit | explicit | explicit | | implicit | implicit | implicit -| long | explicit | explicit | explicit | explicit | | implicit | implicit -| float | explicit | explicit | explicit | explicit | explicit | | implicit +| | byte | short | char | int | long | float | double +| byte | | implicit | implicit | implicit | implicit | implicit | implicit +| short | explicit | | explicit | implicit | implicit | implicit | implicit +| char | explicit | explicit | | implicit | implicit | implicit | implicit +| int | explicit | explicit | explicit | | implicit | implicit | implicit +| long | explicit | explicit | explicit | explicit | | implicit | implicit +| float | explicit | explicit | explicit | explicit | explicit | | implicit | double | explicit | explicit | explicit | explicit | explicit | explicit | |==== +*Examples* -Example(s) -[source,Java] +* Valid numeric type casts. ++ +[source,Painless] ---- -int a = 1; // Declare int variable a and set it to the literal - // value 1 -long b = a; // Declare long variable b and set it to int variable - // a with an implicit cast to convert from int to long -short c = (short)b; // Declare short variable c, explicitly cast b to a - // short, and assign b to c -byte d = a; // ERROR: Casting an int to a byte requires an explicit - // cast -double e = (double)a; // Explicitly cast int variable a to a double and assign - // it to the double variable e. The explicit cast is - // allowed, but it is not necessary. +<1> int a = 1; +<2> long b = a; +<3> short c = (short)b; +<4> double e = (double)a; ---- - -[[reference-casting]] -==== Reference Casting - -A reference type can be implicitly cast to another reference type as long as -the type being cast _from_ is a descendant of the type being cast _to_. A -reference type can be explicitly cast _to_ if the type being cast to is a -descendant of the type being cast _from_. - -*Examples:* -[source,Java] ++ +<1> declare `int a`; + assign `int 1` to `a` +<2> declare `long b`; + access `a` -> `int 1`; + implicit cast `int 1` to `long 1` -> `long 1`; + assign `long 1` to `b` +<3> declare `short c`; + access `b` -> `long 1`; + explicit cast `long 1` to `short 1` -> `short 1`; + assign `short 1` value to `c` +<4> declare `double e`; + access `a` -> `int 1`; + explicit cast `int 1` to `double 1.0`; + assign `double 1.0` to `e`; + (note the explicit cast is extraneous since an implicit cast is valid) ++ +* Invalid numeric type casts resulting in errors. ++ +[source,Painless] ---- -List x; // Declare List variable x -ArrayList y = new ArrayList(); // Declare ArrayList variable y and assign it a - // newly allocated ArrayList [1] -x = y; // Assign Arraylist y to List x using an - // implicit cast -y = (ArrayList)x; // Explicitly cast List x to an ArrayList and - // assign it to ArrayList y -x = (List)y; // Set List x to ArrayList y using an explicit - // cast (the explicit cast is not necessary) -y = x; // ERROR: List x cannot be implicitly cast to - // an ArrayList, an explicit cast is required -Map m = y; // ERROR: Cannot implicitly or explicitly cast [2] - // an ArrayList to a Map, no relationship - // exists between the two types. +<1> int a = 1.0; // error +<2> int b = 2; +<3> byte c = b; // error ---- -[1] `ArrayList` is a descendant of the `List` type. -[2] `Map` is unrelated to the `List` and `ArrayList` types. ++ +<1> declare `int i`; + *error* -> cannot implicit cast `double 1.0` to `int 1`; + (note an explicit cast is valid) +<2> declare `int b`; + assign `int 2` to `b` +<3> declare byte `c`; + access `b` -> `int 2`; + *error* -> cannot implicit cast `int 2` to `byte 2`; + (note an explicit cast is valid) -[[def-type-casting]] -==== def Type Casting -All primitive and reference types can always be implicitly cast to -`def`. While it is possible to explicitly cast to `def`, it is not necessary. +[[reference-type-casting]] +==== Reference Type Casting -However, it is not always possible to implicitly cast a `def` to other -primitive and reference types. An explicit cast is required if an explicit -cast would normally be required between the non-def types. +A <> cast converts the value of an original +reference type to the equivalent value of a target reference type. An implicit +cast between two reference type values is allowed when the original reference +type is a descendant of the target type. An explicit cast between two reference +type values is allowed when the original type is a descendant of the target type +or the target type is a descendant of the original type. +*Examples* -*Examples:* -[source,Java] +* Valid reference type casts. ++ +[source,Painless] ---- -def x; // Declare def variable x and set it to null -x = 3; // Set the def variable x to the literal 3 with an implicit - // cast from int to def -double a = x; // Declare double variable a and set it to def variable x, - // which contains a double -int b = x; // ERROR: Results in a run-time error because an explicit cast is - // required to cast from a double to an int -int c = (int)x; // Declare int variable c, explicitly cast def variable x to an - // int, and assign x to c +<1> List x; +<2> ArrayList y = new ArrayList(); +<3> x = y; +<4> y = (ArrayList)x; +<5> x = (List)y; ---- ++ +<1> declare `List x`; + assign default value `null` to `x` +<2> declare `ArrayList y`; + allocate `ArrayList` instance -> `ArrayList reference`; + assign `ArrayList reference` to `y`; +<3> access `y` -> `ArrayList reference`; + implicit cast `ArrayList reference` to `List reference` -> `List reference`; + assign `List reference` to `x`; + (note `ArrayList` is a descendant of `List`) +<4> access `x` -> `List reference`; + explicit cast `List reference` to `ArrayList reference` + -> `ArrayList reference`; + assign `ArrayList reference` to `y`; +<5> access `y` -> `ArrayList reference`; + explicit cast `ArrayList reference` to `List reference` -> `List reference`; + assign `List reference` to `x`; + (note the explicit cast is extraneous, and an implicit cast is valid) ++ +* Invalid reference type casts resulting in errors. ++ +[source,Painless] +---- +<1> List x = new ArrayList(); +<2> ArrayList y = x; // error +<3> Map m = (Map)x; // error +---- ++ +<1> declare `List x`; + allocate `ArrayList` instance -> `ArrayList reference`; + implicit cast `ArrayList reference` to `List reference` -> `List reference`; + assign `List reference` to `x` +<2> declare `ArrayList y`; + access `x` -> `List reference`; + *error* -> cannot implicit cast `List reference` to `ArrayList reference`; + (note an explicit cast is valid since `ArrayList` is a descendant of `List`) +<3> declare `ArrayList y`; + access `x` -> `List reference`; + *error* -> cannot explicit cast `List reference` to `Map reference`; + (note no cast would be valid since neither `List` nor `Map` is a descendant + of the other) + +[[dynamic-type-casting]] +==== Dynamic Type Casting + +A <> cast converts the value of an original +`def` type to the equivalent value of any target type or converts the value of +any original type to the equivalent value of a target `def` type. + +An implicit cast from any original type value to a `def` type value is always +allowed. An explicit cast from any original type value to a `def` type value is +always allowed but never necessary. + +An implicit or explicit cast from an original `def` type value to +any target type value is allowed if and only if the cast is normally allowed +based on the current type value the `def` type value represents. + +*Examples* + +* Valid dynamic type casts with any original type to a target `def` type. ++ +[source,Painless] +---- +<1> def d0 = 3; +<2> d0 = new ArrayList(); +<3> Object o = new HashMap(); +<4> def d1 = o; +<5> int i = d1.size(); +---- ++ +<1> declare `def d0`; + implicit cast `int 3` to `def`; + assign `int 3` to `d0` +<2> allocate `ArrayList` instance -> `ArrayList reference`; + implicit cast `ArrayList reference` to `def` -> `def`; + assign `def` to `d0` +<3> declare `Object o`; + allocate `HashMap` instance -> `HashMap reference`; + implicit cast `HashMap reference` to `Object reference` + -> `Object reference`; + assign `Object reference` to `o` +<4> declare `def d1`; + access `o` -> `Object reference`; + implicit cast `Object reference` to `def` -> `def`; + assign `def` to `d1` +<5> declare `int i`; + access `d1` -> `def`; + implicit cast `def` to `HashMap reference` -> HashMap reference`; + call `size` on `HashMap reference` -> `int 0`; + assign `int 0` to `i`; + (note `def` was implicit cast to `HashMap reference` since `HashMap` is the + child-most descendant type value that the `def` type value + represents) ++ +* Valid dynamic type casts with an original `def` type to any target type. ++ +[source,Painless] +---- +<1> def d = 1.0; +<2> int i = (int)d; +<3> d = 1; +<4> float f = d; +<5> d = new ArrayList(); +<6> List l = d; +---- ++ +<1> declare `def d`; + implicit cast `double 1.0` to `def` -> `def`; + assign `def` to `d` +<2> declare `int i`; + access `d` -> `def`; + implicit cast `def` to `double 1.0` -> `double 1.0`; + explicit cast `double 1.0` to `int 1` -> `int 1`; + assign `int 1` to `i`; + (note the explicit cast is necessary since a `double` value cannot be + converted to an `int` value implicitly) +<3> assign `int 1` to `d`; + (note the switch in the type `d` represents from `double` to `int`) +<4> declare `float i`; + access `d` -> `def`; + implicit cast `def` to `int 1` -> `int 1`; + implicit cast `int 1` to `float 1.0` -> `float 1.0`; + assign `float 1.0` to `f` +<5> allocate `ArrayList` instance -> `ArrayList reference`; + assign `ArrayList reference` to `d`; + (note the switch in the type `d` represents from `int` to `ArrayList`) +<6> declare `List l`; + access `d` -> `def`; + implicit cast `def` to `ArrayList reference` -> `ArrayList reference`; + implicit cast `ArrayList reference` to `List reference` -> `List reference`; + assign `List reference` to `l` ++ +* Invalid dynamic type casts resulting in errors. ++ +[source,Painless] +---- +<1> def d = 1; +<2> short s = d; // error +<3> d = new HashMap(); +<4> List l = d; // error +---- +<1> declare `def d`; + implicit cast `int 1` to `def` -> `def`; + assign `def` to `d` +<2> declare `short s`; + access `d` -> `def`; + implicit cast `def` to `int 1` -> `int 1`; + *error* -> cannot implicit cast `int 1` to `short 1`; + (note an explicit cast is valid) +<3> allocate `HashMap` instance -> `HashMap reference`; + implicit cast `HashMap reference` to `def` -> `def`; + assign `def` to `d` +<4> declare `List l`; + access `d` -> `def`; + implicit cast `def` to `HashMap reference`; + *error* -> cannot implicit cast `HashMap reference` to `List reference`; + (note no cast would be valid since neither `HashMap` nor `List` is a + descendant of the other) + +[[string-character-casting]] +==== String to Character Casting + +Use the *cast operator* to convert a <> value into a +<> value. + +*Errors* + +* If the `String` type value isn't one character in length. +* If the `String` type value is `null`. + +*Examples* + +* Casting string literals into `char` type values. ++ +[source,Painless] +---- +<1> char c = (char)"C" +<2> c = (char)'c' +---- ++ +<1> declare `char c`; + explicit cast `String "C"` to `char C` -> `char C`; + assign `char C` to `c` +<2> explicit cast `String 'c'` to `char c` -> `char c`; + assign `char c` to `c` ++ +* Casting a `String` reference into a `char` value. ++ +[source,Painless] +---- +<1> String s = "s"; +<2> char c = (char)s; +---- +<1> declare `String s`; + assign `String "s"` to `s`; +<2> declare `char c` + access `s` -> `String "s"`; + explicit cast `String "s"` to `char s` -> `char s`; + assign `char s` to `c` [[boxing-unboxing]] ==== Boxing and Unboxing -Boxing is where a cast is used to convert a primitive type to its corresponding -reference type. Unboxing is the reverse, converting a reference type to the -corresponding primitive type. +Boxing is a special type of cast used to convert a primitive type to its +corresponding reference type. Unboxing is the reverse used to convert a +reference type to its corresponding primitive type. -There are two places Painless performs implicit boxing and unboxing: +Implicit boxing/unboxing occurs during the following operations: -* When you call methods, Painless automatically boxes and unboxes arguments -so you can specify either primitive types or their corresponding reference -types. -* When you use the `def` type, Painless automatically boxes and unboxes as -needed when converting to and from `def`. +* Conversions between a `def` type and a primitive type will be implicitly + boxed/unboxed as necessary, though this is referred to as an implicit cast + throughout the documentation. +* Method/function call arguments will be implicitly boxed/unboxed as necessary. +* A primitive type value will be implicitly boxed when a reference type method + call is invoked on it. -The casting operator does not support any way to explicitly box a primitive -type or unbox a reference type. +Explicit boxing/unboxing is not allowed. Use the reference type API to +explicitly convert a primitive type value to its respective reference type +value and vice versa. -If a primitive type needs to be converted to a reference type, the Painless -reference type API supports methods that can do that. However, under normal -circumstances this should not be necessary. +*Errors* -*Examples:* -[source,Java] +* If an explicit cast is made to box/unbox a primitive type. + +*Examples* + +* Uses of implicit boxing/unboxing. ++ +[source,Painless] ---- -Integer x = 1; // ERROR: not a legal implicit cast -Integer y = (Integer)1; // ERROR: not a legal explicit cast -int a = new Integer(1); // ERROR: not a legal implicit cast -int b = (int)new Integer(1); // ERROR: not a legal explicit cast +<1> List l = new ArrayList(); +<2> l.add(1); +<3> Integer I = Integer.valueOf(0); +<4> int i = l.get(i); ---- ++ +<1> declare `List l`; + allocate `ArrayList` instance -> `ArrayList reference`; + assign `ArrayList reference` to `l`; +<2> access `l` -> `List reference`; + implicit cast `int 1` to `def` -> `def`; + call `add` on `List reference` with arguments (`def`); + (note internally `int 1` is boxed to `Integer 1` to store as a `def` type + value) +<3> declare `Integer I`; + call `valueOf` on `Integer` with arguments of (`int 0`) -> `Integer 0`; + assign `Integer 0` to `I`; +<4> declare `int i`; + access `I` -> `Integer 0`; + unbox `Integer 0` -> `int 0`; + access `l` -> `List reference`; + call `get` on `List reference` with arguments (`int 0`) -> `def`; + implicit cast `def` to `int 1` -> `int 1`; + assign `int 1` to `i`; + (note internally `int 1` is unboxed from `Integer 1` when loaded from a + `def` type value) ++ +* Uses of invalid boxing/unboxing resulting in errors. ++ +[source,Painless] +---- +<1> Integer x = 1; // error +<2> Integer y = (Integer)1; // error +<3> int a = Integer.valueOf(1); // error +<4> int b = (int)Integer.valueOf(1); // error +---- ++ +<1> declare `Integer x`; + *error* -> cannot implicit box `int 1` to `Integer 1` during assignment +<2> declare `Integer y`; + *error* -> cannot explicit box `int 1` to `Integer 1` during assignment +<3> declare `int a`; + call `valueOf` on `Integer` with arguments of (`int 1`) -> `Integer 1`; + *error* -> cannot implicit unbox `Integer 1` to `int 1` during assignment +<4> declare `int a`; + call `valueOf` on `Integer` with arguments of (`int 1`) -> `Integer 1`; + *error* -> cannot explicit unbox `Integer 1` to `int 1` during assignment [[promotion]] ==== Promotion -Promotion is where certain operations require types to be either a minimum -numerical type or for two (or more) types to be equivalent. -The documentation for each operation that has these requirements -includes promotion tables that describe how this is handled. +Promotion is when a single value is implicitly cast to a certain type or +multiple values are implicitly cast to the same type as required for evaluation +by certain operations. Each operation that requires promotion has a promotion +table that shows all required implicit casts based on the type(s) of value(s). A +value can be promoted to a `def` type at compile-time; however, the promoted +type value is derived from what the `def` type value represents at run-time. -When an operation promotes a type or types, the resultant type -of the operation is the promoted type. Types can be promoted to def -at compile-time; however, at run-time, the resultant type will be the -promotion of the types the `def` is representing. +*Errors* -*Examples:* -[source,Java] +* If a specific operation cannot find an allowed promotion type for the type(s) + of value(s) given. + +*Examples* + +* Uses of promotion. ++ +[source,Painless] ---- -2 + 2.0 // Add the literal int 2 and the literal double 2.0. The literal - // 2 is promoted to a double and the resulting value is a double. - -def x = 1; // Declare def variable x and set it to the literal int 1 through - // an implicit cast -x + 2.0F // Add def variable x and the literal float 2.0. - // At compile-time the types are promoted to def. - // At run-time the types are promoted to float. +<1> double d = 2 + 2.0; +<2> def x = 1; +<3> float f = x + 2.0F; ---- +<1> declare `double d`; + promote `int 2` and `double 2.0 @0` -> `double 2.0 @0`; + implicit cast `int 2` to `double 2.0 @1` -> `double 2.0 @1`; + add `double 2.0 @1` and `double 2.0 @0` -> `double 4.0`; + assign `double 4.0` to `d` +<2> declare `def x`; + implicit cast `int 1` to `def` -> `def`; + assign `def` to `x`; +<3> declare `float f`; + access `x` -> `def`; + implicit cast `def` to `int 1` -> `int 1`; + promote `int 1` and `float 2.0` -> `float 2.0`; + implicit cast `int 1` to `float 1.0` -> `float `1.0`; + add `float 1.0` and `float 2.0` -> `float 3.0`; + assign `float 3.0` to `f`; + (note this example illustrates promotion done at run-time as promotion + done at compile-time would have resolved to a `def` type value) diff --git a/docs/painless/painless-comments.asciidoc b/docs/painless/painless-comments.asciidoc index 588e464d97f..bde30e37e04 100644 --- a/docs/painless/painless-comments.asciidoc +++ b/docs/painless/painless-comments.asciidoc @@ -1,12 +1,12 @@ [[painless-comments]] === Comments -Use the `//` token anywhere on a line to specify a single-line comment. All -characters from the `//` token to the end of the line are ignored. Use an -opening `/*` token and a closing `*/` token to specify a multi-line comment. -Multi-line comments can start anywhere on a line, and all characters in between -the `/*` token and `*/` token are ignored. Comments can be included anywhere -within a script. +Use a comment to annotate or explain code within a script. Use the `//` token +anywhere on a line to specify a single-line comment. All characters from the +`//` token to the end of the line are ignored. Use an opening `/*` token and a +closing `*/` token to specify a multi-line comment. Multi-line comments can +start anywhere on a line, and all characters in between the `/*` token and `*/` +token are ignored. Comments can be included anywhere within a script. *Grammar* [source,ANTLR4] diff --git a/docs/painless/painless-identifiers.asciidoc b/docs/painless/painless-identifiers.asciidoc index 17073e3d4c4..7762f56cb7b 100644 --- a/docs/painless/painless-identifiers.asciidoc +++ b/docs/painless/painless-identifiers.asciidoc @@ -1,10 +1,10 @@ [[painless-identifiers]] === Identifiers -Specify identifiers to <>, <>, and -<> variables, <>, and -<>. <> and -<> cannot be used as identifiers. +Use an identifier as a named token to specify a +<>, <>, +<>, <>, or function. +<> cannot be used as identifiers. *Grammar* [source,ANTLR4] diff --git a/docs/painless/painless-keywords.asciidoc b/docs/painless/painless-keywords.asciidoc index cb3bafbd20f..39a2201fd2b 100644 --- a/docs/painless/painless-keywords.asciidoc +++ b/docs/painless/painless-keywords.asciidoc @@ -1,9 +1,9 @@ [[painless-keywords]] === Keywords -The keywords in the table below are reserved for built-in language -features. These keywords cannot be used as -<> or <>. +Keywords are reserved tokens for built-in language features and cannot be used +as <> within a script. The following are +keywords: [cols="^1,^1,^1,^1,^1"] |==== diff --git a/docs/painless/painless-lang-spec.asciidoc b/docs/painless/painless-lang-spec.asciidoc index ba6595000ae..5e6b84d8c57 100644 --- a/docs/painless/painless-lang-spec.asciidoc +++ b/docs/painless/painless-lang-spec.asciidoc @@ -6,7 +6,7 @@ Painless syntax is similar to Java syntax along with some additional features such as dynamic typing, Map and List accessor shortcuts, and array initializers. As a direct comparison to Java, there are some important differences, especially related to the casting model. For more detailed -conceptual information about the basic constructs that Java and Painless share, +conceptual information about the basic constructs that Painless and Java share, refer to the corresponding topics in the https://docs.oracle.com/javase/specs/jls/se8/html/index.html[Java Language Specification]. diff --git a/docs/painless/painless-literals.asciidoc b/docs/painless/painless-literals.asciidoc index 441cb264f1e..ebf7eaa07b6 100644 --- a/docs/painless/painless-literals.asciidoc +++ b/docs/painless/painless-literals.asciidoc @@ -1,18 +1,19 @@ [[painless-literals]] === Literals -Use literals to specify different types of values directly in a script. +Use a literal to specify a value directly in an +<>. [[integers]] ==== Integers -Use integer literals to specify an integer value in decimal, octal, or hex -notation of the <> `int`, `long`, `float`, +Use an integer literal to specify an integer type value in decimal, octal, or +hex notation of a <> `int`, `long`, `float`, or `double`. Use the following single letter designations to specify the -<>: `l` or `L` for `long`, `f` or `F` for -`float`, and `d` or `D` for `double`. If not specified, the type defaults to -`int`. Use `0` as a prefix to specify an integer literal as octal, and use -`0x` or `0X` as a prefix to specify an integer literal as hex. +primitive type: `l` or `L` for `long`, `f` or `F` for `float`, and `d` or `D` +for `double`. If not specified, the type defaults to `int`. Use `0` as a prefix +to specify an integer literal as octal, and use `0x` or `0X` as a prefix to +specify an integer literal as hex. *Grammar* [source,ANTLR4] @@ -46,11 +47,10 @@ HEX: '-'? '0' [xX] [0-9a-fA-F]+ [lL]?; [[floats]] ==== Floats -Use floating point literals to specify a floating point value of the -<> `float` or `double`. Use the following -single letter designations to specify the <>: -`f` or `F` for `float` and `d` or `D` for `double`. If not specified, the type defaults -to `double`. +Use a floating point literal to specify a floating point type value of a +<> `float` or `double`. Use the following +single letter designations to specify the primitive type: `f` or `F` for `float` +and `d` or `D` for `double`. If not specified, the type defaults to `double`. *Grammar* [source,ANTLR4] @@ -81,7 +81,7 @@ EXPONENT: ( [eE] [+\-]? [0-9]+ ); [[strings]] ==== Strings -Use string literals to specify <> values with +Use a string literal to specify a <> value with either single-quotes or double-quotes. Use a `\"` token to include a double-quote as part of a double-quoted string literal. Use a `\'` token to include a single-quote as part of a single-quoted string literal. Use a `\\` @@ -117,26 +117,6 @@ STRING: ( '"' ( '\\"' | '\\\\' | ~[\\"] )*? '"' ) [[characters]] ==== Characters -Use the <> to convert string literals or -<> values into <> values. -<> values converted into -<> values must be exactly one character in length -or an error will occur. - -*Examples* - -* Casting string literals into <> values. -+ -[source,Painless] ----- -(char)"C" -(char)'c' ----- -+ -* Casting a <> value into a <> value. -+ -[source,Painless] ----- -String s = "s"; -char c = (char)s; ----- +A character literal cannot be specified directly. Instead, use the +<> to convert a `String` type value +into a `char` type value. diff --git a/docs/painless/painless-operators.asciidoc b/docs/painless/painless-operators.asciidoc index 915d811fa44..8329686f663 100644 --- a/docs/painless/painless-operators.asciidoc +++ b/docs/painless/painless-operators.asciidoc @@ -240,6 +240,7 @@ operator. See Function Calls [MARK] for more information. The brackets operator `[]` is used to create and access arrays, lists, and maps. The braces operator `{}` is used to intialize arrays. +[[array-initialization]] ===== Creating and Initializing Arrays You create and initialize arrays using the brackets `[]` and braces `{}` @@ -248,9 +249,49 @@ initialize each dimension with are specified as a comma-separated list enclosed in braces. For example, `new int[] {1, 2, 3}` creates a one dimensional `int` array with a size of 3 and the values 1, 2, and 3. -For more information about allocating and initializing arrays, see <>. +To allocate an array, you use the `new` keyword followed by the type and a +set of brackets for each dimension. You can explicitly define the size of each dimension by specifying an expression within the brackets, or initialize each +dimension with the desired number of values. The allocated size of each +dimension is its permanent size. +To initialize an array, specify the values you want to initialize +each dimension with as a comma-separated list of expressions enclosed in braces. +For example, `new int[] {1, 2, 3}` creates a one-dimensional `int` array with a +size of 3 and the values 1, 2, and 3. + +When you initialize an array, the order of the expressions is maintained. Each expression used as part of the initialization is converted to the +array's type. An error occurs if the types do not match. + +*Grammar:* +[source,ANTLR4] +---- +declare_array: TYPE ('[' ']')+; + +array_initialization: 'new' TYPE '[' ']' '{' expression (',' expression) '}' + | 'new' TYPE '[' ']' '{' '}'; +---- + +*Examples:* +[source,Java] +---- +int[] x = new int[5]; // Declare int array x and assign it a newly + // allocated int array with a size of 5 +def[][] y = new def[5][5]; // Declare the 2-dimensional def array y and + // assign it a newly allocated 2-dimensional + // array where both dimensions have a size of 5 +int[] x = new int[] {1, 2, 3}; // Declare int array x and set it to an int + // array with values 1, 2, 3 and a size of 3 +int i = 1; +long l = 2L; +float f = 3.0F; +double d = 4.0; +String s = "5"; +def[] da = new def[] {i, l, f*d, s}; // Declare def array da and set it to + // a def array with a size of 4 and the + // values i, l, f*d, and s +---- + +[[array-access]] ===== Accessing Array Elements Elements in an array are stored and accessed using the brackets `[]` operator. @@ -298,6 +339,7 @@ return d[z]; // Access the 1st element of array d using the NOTE: The use of the `def` type in the second example means that the types cannot be resolved until runtime. +[[array-length]] ===== Array Length Arrays contain a special member known as 'length' that is a read-only value that contains the size of the array. This member can be accessed from an array using the dot operator. @@ -727,6 +769,7 @@ def e; // declares the def variable e e = new HashMap(m); // sets e to a newly allocated HashMap using the constructor with a single argument m ---- +[[new-array]] ==== New Array An array type instance can be allocated using the new operator. The format starts with the new operator followed by the type followed by a series of opening and closing braces each containing an expression for the size of the dimension. diff --git a/docs/painless/painless-types.asciidoc b/docs/painless/painless-types.asciidoc index 9d575a2069a..a897b8e8a04 100644 --- a/docs/painless/painless-types.asciidoc +++ b/docs/painless/painless-types.asciidoc @@ -1,269 +1,466 @@ [[painless-types]] === Types -Painless supports both dynamic and static types. Static types are split into -_primitive types_ and _reference types_. - -[[dynamic-types]] -==== Dynamic Types - -Painless supports one dynamic type: `def`. The `def` type can represent any -primitive or reference type. When you use the `def` type, it mimics the exact -behavior of whatever type it represents at runtime. The default value for the -def type is `null.` - -Internally, if the `def` type represents a primitive type, it is converted to the -corresponding reference type. It still behaves like the primitive type, however, -including within the casting model. The `def` type can be assigned to different -types during the course of script execution. - -IMPORTANT: Because a `def` type variable can be assigned to different types -during execution, type conversion errors that occur when using the `def` type -happen at runtime. - -Using the `def` type can have a slight impact on performance. If performance is -critical, it's better to declare static types. - -*Examples:* -[source,Java] ----- -def x = 1; // Declare def variable x and set it to the - // literal int 1 -def l = new ArrayList(); // Declare def variable l and set it a newly - // allocated ArrayList ----- +A type is a classification of data used to define the properties of a value. +These properties specify what data a value represents and the rules for how a +value is evaluated during an <>. Each type +belongs to one of the following categories: <>, +<>, or <>. [[primitive-types]] ==== Primitive Types -Primitive types are allocated directly onto the stack according to the standard -Java memory model. +A primitive type represents basic data built natively into the JVM and is +allocated to non-heap memory. Declare a primitive type +<>, and assign it a primitive type value for +evaluation during later operations. The default value for a newly-declared +primitive type variable is listed as part of the definitions below. A primitive +type value is copied during an assignment or as an argument for a +method/function call. -Primitive types can behave as their corresponding (<>) -reference type. This means any piece of a reference type can be accessed or -called through the primitive type. Operations performed in this manner convert -the primitive type to its corresponding reference type at runtime and perform -the field access or method call without needing to perform any other -operations. +A primitive type has a corresponding reference type (also known as a boxed +type). Use the <> or +<> on a primitive type value to force +evaluation as its corresponding reference type value. -Painless supports the following primitive types. +The following primitive types are available: -byte:: -An 8-bit, signed, two's complement integer. -Range: [-128, 127]. -Default value: 0. -Reference type: Byte. +[horizontal] +`byte`:: +8-bit, signed, two's complement integer +* range: [`-128`, `127`] +* default value: `0` +* reference type: `Byte` -short:: -A 16-bit, signed, two's complement integer. -Range: [-32768, 32767]. -Default value: 0. -Reference type: Short. +`short`:: +16-bit, signed, two's complement integer +* range: [`-32768`, `32767`] +* default value: `0` +* reference type: `Short` -char:: -A 16-bit Unicode character. -Range: [0, 65535]. -Default value: 0 or `\u0000`. -Reference type: Character. +`char`:: +16-bit, unsigned, Unicode character +* range: [`0`, `65535`] +* default value: `0` or `\u0000` +* reference type: `Character` -int:: -A 32-bit, signed, two's complement integer. -Range: [-2^32, 2^32-1]. -Default value: 0. -Reference type: Integer. +`int`:: +32-bit, signed, two's complement integer +* range: [`-2^32`, `2^32-1`] +* default value: `0` +* reference type: `Integer` -long:: -A 64-bit, signed, two's complement integer. -Range: [-2^64, 2^64-1]. -Default value: 0. -Reference type: Long. +`long`:: +64-bit, signed, two's complement integer +* range: [`-2^64`, `2^64-1`] +* default value: `0` +* reference type: `Long` -float:: -A 32-bit, single-precision, IEEE 754 floating point number. -Range: Depends on multiple factors. -Default value: 0.0. -Reference type: Float. +`float`:: +32-bit, signed, single-precision, IEEE 754 floating point number +* default value: `0.0` +* reference type: `Float` -double:: -A 64-bit, double-precision, IEEE 754 floating point number. -Range: Depends on multiple factors. -Default value: 0.0. -Reference type: Double. +`double`:: +64-bit, signed, double-precision, IEEE 754 floating point number +* default value: `0.0` +* reference type: `Double` -boolean:: -A logical quanity with two possible values: true and false. -Range: true/false. -Default value: false. -Reference type: Boolean. +`boolean`:: +logical quantity with two possible values of `true` and `false` +* default value: `false` +* reference type: `Boolean` +*Examples* -*Examples:* -[source,Java] +* Primitive types used in declaration, declaration and assignment. ++ +[source,Painless] ---- -int i = 1; // Declare variable i as an int and set it to the - // literal 1 -double d; // Declare variable d as a double and set it to the - // default value of 0.0 -boolean b = true; // Declare variable b as a boolean and set it to true +<1> int i = 1; +<2> double d; +<3> boolean b = true; ---- - -Using methods from the corresponding reference type on a primitive type. - -[source,Java] ++ +<1> declare `int i`; + assign `int 1` to `i` +<2> declare `double d`; + assign default `double 0.0` to `d` +<3> declare `boolean b`; + assign `boolean true` to `b` ++ +* Method call on a primitive type using the corresponding reference type. ++ +[source,Painless] ---- -int i = 1; // Declare variable i as an int and set it to the - // literal 1 -i.toString(); // Invokes the Integer method toString on variable i +<1> int i = 1; +<2> i.toString(); ---- ++ +<1> declare `int i`; + assign `int 1` to `i` +<2> access `i` -> `int 1`; + box `int 1` -> `Integer 1 reference`; + call `toString` on `Integer 1 reference` -> `String '1'` [[reference-types]] ==== Reference Types -Reference types are similar to Java classes and can contain multiple pieces -known as _members_. However, reference types do not support access modifiers. -You allocate reference type instances on the heap using the `new` operator. +A reference type is a named construct (object), potentially representing +multiple pieces of data (member fields) and logic to manipulate that data +(member methods), defined as part of the application programming interface +(API) for scripts. -Reference types can have both static and non-static members: +A reference type instance is a single set of data for one reference type +object allocated to the heap. Use the +<> to allocate a reference type +instance. Use a reference type instance to load from, store to, and manipulate +complex data. -* Static members are shared by all instances of the same reference type and -can be accessed without allocating an instance of the reference type. For -example `Integer.MAX_VALUE`. -* Non-static members are specific to an instance of the reference type -and can only be accessed through the allocated instance. +A reference type value refers to a reference type instance, and multiple +reference type values may refer to the same reference type instance. A change to +a reference type instance will affect all reference type values referring to +that specific instance. -The default value for a reference type is `null`, indicating that no memory has -been allocated for it. When you assign `null` to a reference type, its previous -value is discarded and garbage collected in accordance with the Java memory -model as long as there are no other references to that value. +Declare a reference type <>, and assign it a +reference type value for evaluation during later operations. The default value +for a newly-declared reference type variable is `null`. A reference type value +is shallow-copied during an assignment or as an argument for a method/function +call. Assign `null` to a reference type variable to indicate the reference type +value refers to no reference type instance. The JVM will garbage collect a +reference type instance when it is no longer referred to by any reference type +values. Pass `null` as an argument to a method/function call to indicate the +argument refers to no reference type instance. -A reference type can contain: +A reference type object defines zero-to-many of each of the following: -* Zero to many primitive types. Primitive type members can be static or -non-static and read-only or read-write. -* Zero to many reference types. Reference type members can be static or -non-static and read-only or read-write. -* Methods that call an internal function to return a value and/or manipulate -the primitive or reference type members. Method members can be static or -non-static. -* Constructors that call an internal function to return a newly-allocated -reference type instance. Constructors are non-static methods that can -optionally manipulate the primitive and reference type members. +static member field:: -Reference types support a Java-style inheritance model. Consider types A and B. -Type A is considered to be a parent of B, and B a child of A, if B inherits -(is able to access as its own) all of A's fields and methods. Type B is +A static member field is a named and typed piece of data. Each reference type +*object* contains one set of data representative of its static member fields. +Use the <> in correspondence with the +reference type object name to access a static member field for loading and +storing to a specific reference type *object*. No reference type instance +allocation is necessary to use a static member field. + +non-static member field:: + +A non-static member field is a named and typed piece of data. Each reference +type *instance* contains one set of data representative of its reference type +object's non-static member fields. Use the +<> for loading and storing to a non-static +member field of a specific reference type *instance*. An allocated reference +type instance is required to use a non-static member field. + +static member method:: + +A static member method is a function called on a reference type *object*. Use +the <> in correspondence with the reference +type object name to call a static member method. No reference type instance +allocation is necessary to use a static member method. + +non-static member method:: + +A non-static member method is a function called on a reference type *instance*. +A non-static member method called on a reference type instance can load from and +store to non-static member fields of that specific reference type instance. Use +the <> in correspondence with a specific +reference type instance to call a non-static member method. An allocated +reference type instance is required to use a non-static member method. + +constructor:: + +A constructor is a special type of function used to allocate a reference type +*instance* defined by a specific reference type *object*. Use the +<> to allocate a reference type +instance. + +A reference type object follows a basic inheritance model. Consider types A and +B. Type A is considered to be a parent of B, and B a child of A, if B inherits +(is able to access as its own) all of A's non-static members. Type B is considered a descendant of A if there exists a recursive parent-child relationship from B to A with none to many types in between. In this case, B -inherits all of A's fields and methods along with all of the fields and -methods of the types in between. Type B is also considered to be a type A -in both relationships. +inherits all of A's non-static members along with all of the non-static members +of the types in between. Type B is also considered to be a type A in both +relationships. -For the complete list of Painless reference types and their supported methods, -see the https://www.elastic.co/guide/en/elasticsearch/reference/current/painless-api-reference.html[Painless API Reference]. +*Examples* -For more information about working with reference types, see -<> and <>. - -*Examples:* -[source,Java] +* Reference types evaluated in several different operations. ++ +[source,Painless] ---- -ArrayList al = new ArrayList(); // Declare variable al as an ArrayList and - // set it to a newly allocated ArrayList -List l = new ArrayList(); // Declare variable l as a List and set - // it to a newly allocated ArrayList, which is - // allowed because ArrayList inherits from List -Map m; // Declare variable m as a Map and set it - // to the default value of null +<1> List l = new ArrayList(); +<2> l.add(1); +<3> int i = l.get(0) + 2; ---- ++ +<1> declare `List l`; + allocate `ArrayList` instance -> `ArrayList reference`; + implicit cast `ArrayList reference` to `List reference` -> `List reference`; + assign `List reference` to `l` +<2> access `l` -> `List reference`; + implicit cast `int 1` to `def` -> `def` + call `add` on `List reference` with arguments (`def`) +<3> declare `int i`; + access `l` -> `List reference`; + call `get` on `List reference` with arguments (`int 0`) -> `def`; + implicit cast `def` to `int 1` -> `int 1`; + add `int 1` and `int 2` -> `int 3`; + assign `int 3` to `i` ++ +* Sharing a reference type instance. ++ +[source,Painless] +---- +<1> List l0 = new ArrayList(); +<2> List l1 = l0; +<3> l0.add(1); +<4> l1.add(2); +<5> int i = l1.get(0) + l0.get(1); +---- ++ +<1> declare `List l0`; + allocate `ArrayList` instance -> `ArrayList reference`; + implicit cast `ArrayList reference` to `List reference` -> `List reference`; + assign `List reference` to `l0` +<2> declare `List l1`; + access `l0` -> `List reference`; + assign `List reference` to `l1` + (note `l0` and `l1` refer to the same instance known as a shallow-copy) +<3> access `l0` -> `List reference`; + implicit cast `int 1` to `def` -> `def` + call `add` on `List reference` with arguments (`def`) +<4> access `l1` -> `List reference`; + implicit cast `int 2` to `def` -> `def` + call `add` on `List reference` with arguments (`def`) +<5> declare `int i`; + access `l0` -> `List reference`; + call `get` on `List reference` with arguments (`int 0`) -> `def @0`; + implicit cast `def @0` to `int 1` -> `int 1`; + access `l1` -> `List reference`; + call `get` on `List reference` with arguments (`int 1`) -> `def @1`; + implicit cast `def @1` to `int 2` -> `int 2`; + add `int 1` and `int 2` -> `int 3`; + assign `int 3` to `i`; ++ +* Using the static members of a reference type. ++ +[source,Painless] +---- +<1> int i = Integer.MAX_VALUE; +<2> long l = Long.parseLong("123L"); +---- ++ +<1> declare `int i`; + access `MAX_VALUE` on `Integer` -> `int 2147483647`; + assign `int 2147483647` to `i` +<2> declare `long l`; + call `parseLong` on `Long` with arguments (`long 123`) -> `long 123`; + assign `long 123` to `l` -Directly accessing static pieces of a reference type. +[[dynamic-types]] +==== Dynamic Types -[source,Java] +A dynamic type value can represent the value of any primitive type or +reference type using a single type name `def`. A `def` type value mimics +the behavior of whatever value it represents at run-time and will always +represent the child-most descendant type value of any type value when evaluated +during operations. + +Declare a `def` type <>, and assign it +any type of value for evaluation during later operations. The default value +for a newly-declared `def` type variable is `null`. A `def` type variable or +method/function parameter can change the type it represents during the +compilation and evaluation of a script. + +Using the `def` type can have a slight impact on performance. Use only primitive +types and reference types directly when performance is critical. + +*Errors* + +* If a `def` type value represents an inappropriate type for evaluation of an + operation at run-time. + +*Examples* + +* General uses of the `def` type. ++ +[source,Painless] ---- -Integer.MAX_VALUE // a static field access -Long.parseLong("123L") // a static function call +<1> def dp = 1; +<2> def dr = new ArrayList(); +<3> dr = dp; ---- ++ +<1> declare `def dp`; + implicit cast `int 1` to `def` -> `def`; + assign `def` to `dp` +<2> declare `def dr`; + allocate `ArrayList` instance -> `ArrayList reference`; + implicit cast `ArrayList reference` to `def` -> `def`; + assign `def` to `dr` +<3> access `dp` -> `def`; + assign `def` to `dr`; + (note the switch in the type `dr` represents from `ArrayList` to `int`) ++ +* A `def` type value representing the child-most descendant of a value. ++ +[source,Painless] +---- +<1> Object l = new ArrayList(); +<2> def d = l; +<3> d.ensureCapacity(10); +---- ++ +<1> declare `Object l`; + allocate `ArrayList` instance -> `ArrayList reference`; + implicit cast `ArrayList reference` to `Object reference` + -> `Object reference`; + assign `Object reference` to `l` +<2> declare `def d`; + access `l` -> `Object reference`; + implicit cast `Object reference` to `def` -> `def`; + assign `def` to `d`; +<3> access `d` -> `def`; + implicit cast `def` to `ArrayList reference` -> `ArrayList reference`; + call `ensureCapacity` on `ArrayList reference` with arguments (`int 10`); + (note `def` was implicit cast to `ArrayList reference` + since ArrayList` is the child-most descendant type value that the + `def` type value represents) [[string-type]] ==== String Type -A `String` is a specialized reference type that is immutable and does not have -to be explicitly allocated. You can directly assign to a `String` without first -allocating it with the `new` keyword. (Strings can be allocated with the `new` -keyword, but it's not required.) +The `String` type is a specialized reference type that does not require +explicit allocation. Use a <> to directly evaluate a +`String` type value. While not required, the +<> can allocate `String` type +instances. -When assigning a value to a `String`, you must enclose the text in single or -double quotes. Strings are allocated according to the standard Java Memory Model. -The default value for a `String` is `null.` +*Examples* -*Examples:* -[source,Java] +* General use of the `String` type. ++ +[source,Painless] ---- -String r = "some text"; // Declare String r and set it to the - // String "some text" -String s = 'some text'; // Declare String s and set it to the - // String 'some text' -String t = new String("some text"); // Declare String t and set it to the - // String "some text" -String u; // Declare String u and set it to the - // default value null +<1> String r = "some text"; +<2> String s = 'some text'; +<3> String t = new String("some text"); +<4> String u; ---- ++ +<1> declare `String r`; + assign `String "some text"` to `r` +<2> declare `String s`; + assign `String 'some text'` to `s` +<3> declare `String t`; + allocate `String` instance with arguments (`String "some text"`) + -> `String "some text"`; + assign `String "some text"` to `t` +<4> declare `String u`; + assign default `null` to `u` [[void-type]] ==== void Type -The `void` type represents the concept of no type. In Painless, `void` declares -that a function has no return value. +The `void` type represents the concept of a lack of type. Use the `void` type to +indicate a function returns no value. + +*Examples* + +* Use of the `void` type in a function. ++ +[source,Painless] +---- +void addToList(List l, def d) { + l.add(d); +} +---- [[array-type]] ==== Array Type -Arrays contain a series of elements of the same type that can be allocated -simultaneously. Painless supports both single and multi-dimensional arrays for -all types except void (including `def`). +An array type is a specialized reference type where an array type instance +represents a series of values allocated to the heap. All values in an array +type instance are of the same type. Each value is assigned an index from within +the range `[0, length)` where length is the total number of values allocated for +the array type instance. -You declare an array by specifying a type followed by a series of empty brackets, -where each set of brackets represents a dimension. Declared arrays have a default -value of `null` and are themselves a reference type. +Use the <> or the +<> to allocate an array +type instance. Declare an array type <>, and +assign it an array type value for evaluation during later operations. The +default value for a newly-declared array type variable is `null`. An array type +value is shallow-copied during an assignment or as an argument for a +method/function call. Assign `null` to an array type variable to indicate the +array type value refers to no array type instance. The JVM will garbage collect +an array type instance when it is no longer referred to by any array type +values. Pass `null` as an argument to a method/function call to indicate the +argument refers to no array type instance. -To allocate an array, you use the `new` keyword followed by the type and a -set of brackets for each dimension. You can explicitly define the size of each dimension by specifying an expression within the brackets, or initialize each -dimension with the desired number of values. The allocated size of each -dimension is its permanent size. +Use the <> to retrieve the length of an +array type value as an int type value. Use the +<> to load from and store to individual +values within an array type value. -To initialize an array, specify the values you want to initialize -each dimension with as a comma-separated list of expressions enclosed in braces. -For example, `new int[] {1, 2, 3}` creates a one-dimensional `int` array with a -size of 3 and the values 1, 2, and 3. +When an array type instance is allocated with multiple dimensions using the +range `[2, d]` where `d >= 2`, each dimension in the range `[1, d-1]` is also +an array type. The array type of each dimension, `n`, is an array type with the +number of dimensions equal to `d-n`. For example, consider `int[][][]` with 3 +dimensions. The 3rd dimension, `d-3`, is the primitive type `int`. The 2nd +dimension, `d-2`, is the array type `int[]`. And the 1st dimension, `d-1` is +the array type `int[][]`. -When you initialize an array, the order of the expressions is maintained. Each expression used as part of the initialization is converted to the -array's type. An error occurs if the types do not match. +*Examples* -*Grammar:* -[source,ANTLR4] +* General use of single-dimensional arrays. ++ +[source,Painless] ---- -declare_array: TYPE ('[' ']')+; - -array_initialization: 'new' TYPE '[' ']' '{' expression (',' expression) '}' - | 'new' TYPE '[' ']' '{' '}'; +<1> int[] x; +<2> float[] y = new float[10]; +<3> def z = new float[5]; +<4> y[9] = 1.0F; +<5> z[0] = y[9]; ---- - -*Examples:* -[source,Java] ++ +<1> declare `int[] x`; + assign default `null` to `x` +<2> declare `float[] y`; + allocate `1-d float array` instance with `length [10]` + -> `1-d float array reference`; + assign `1-d float array reference` to `y` +<3> declare `def z`; + allocate `1-d float array` instance with `length [5]` + -> `1-d float array reference`; + implicit cast `1-d float array reference` to `def` -> `def`; + assign `def` to `z` +<4> access `y` -> `1-d float array reference`; + assign `float 1.0` to `index [9]` of `1-d float array reference` +<5> access `y` -> `1-d float array reference @0`; + access `index [9]` of `1-d float array reference @0` -> `float 1.0`; + access `z` -> `def`; + implicit cast `def` to `1-d float array reference @1` + -> `1-d float array reference @1`; + assign `float 1.0` to `index [0]` of `1-d float array reference @1` ++ +* Use of a multi-dimensional array. ++ +[source,Painless] ---- -int[] x = new int[5]; // Declare int array x and assign it a newly - // allocated int array with a size of 5 -def[][] y = new def[5][5]; // Declare the 2-dimensional def array y and - // assign it a newly allocated 2-dimensional - // array where both dimensions have a size of 5 -int[] x = new int[] {1, 2, 3}; // Declare int array x and set it to an int - // array with values 1, 2, 3 and a size of 3 -int i = 1; -long l = 2L; -float f = 3.0F; -double d = 4.0; -String s = "5"; -def[] da = new def[] {i, l, f*d, s}; // Declare def array da and set it to - // a def array with a size of 4 and the - // values i, l, f*d, and s +<1> int[][][] ia3 = new int[2][3][4]; +<2> ia3[1][2][3] = 99; +<3> int i = ia3[1][2][3]; ---- ++ +<1> declare `int[][][] ia`; + allocate `3-d int array` instance with length `[2, 3, 4]` + -> `3-d int array reference`; + assign `3-d int array reference` to `ia3` +<2> access `ia3` -> `3-d int array reference`; + assign `int 99` to `index [1, 2, 3]` of `3-d int array reference` +<3> declare `int i`; + access `ia3` -> `3-d int array reference`; + access `index [1, 2, 3]` of `3-d int array reference` -> `int 99`; + assign `int 99` to `i` diff --git a/docs/painless/painless-variables.asciidoc b/docs/painless/painless-variables.asciidoc index 9756676a08b..8b8782b1511 100644 --- a/docs/painless/painless-variables.asciidoc +++ b/docs/painless/painless-variables.asciidoc @@ -1,29 +1,31 @@ [[painless-variables]] === Variables -<> variables to <> values for -<> in expressions. Specify variables as a -<>, <>, or -<>. Variable operations follow the structure of a -standard JVM in relation to instruction execution and memory usage. +A variable loads and stores a value for evaluation during +<>. [[declaration]] ==== Declaration -Declare variables before use with the format of <> -<>. Specify a comma-separated list of -<> following the <> -to declare multiple variables in a single statement. Use an -<> statement combined with a declaration statement to -immediately assign a value to a variable. Variables not immediately assigned a -value will have a default value assigned implicitly based on the -<>. +Declare a variable before use with the format of <> +followed by <>. Declare an +<> variable using an opening `[` token and a closing `]` +token for each dimension directly after the identifier. Specify a +comma-separated list of identifiers following the type to declare multiple +variables in a single statement. Use an <> +combined with a declaration to immediately assign a value to a variable. +A variable not immediately assigned a value will have a default value assigned +implicitly based on the type. + +*Errors* + +* If a variable is used prior to or without declaration. *Grammar* [source,ANTLR4] ---- declaration : type ID assignment? (',' ID assignment?)*; -type: ID ('[' ']')*; +type: ID ('.' ID)* ('[' ']')*; assignment: '=' expression; ---- @@ -35,27 +37,43 @@ assignment: '=' expression; ---- <1> int x; <2> List y; -<3> int x, y, z; -<4> def[] d; +<3> int x, y = 5, z; +<4> def d; <5> int i = 10; +<6> float[] f; +<7> Map[][] m; ---- + -<1> declare a variable of type `int` and identifier `x` -<2> declare a variable of type `List` and identifier `y` -<3> declare three variables of type `int` and identifiers `x`, `y`, `z` -<4> declare a variable of type `def[]` and identifier `d` -<5> declare a variable of type `int` and identifier `i`; - assign the integer literal `10` to `i` +<1> declare `int x`; + assign default `null` to `x` +<2> declare `List y`; + assign default `null` to `y` +<3> declare `int x`; + assign default `int 0` to `x`; + declare `int y`; + assign `int 5` to `y`; + declare `int z`; + assign default `int 0` to `z`; +<4> declare `def d`; + assign default `null` to `d` +<5> declare `int i`; + assign `int 10` to `i` +<6> declare `float[] f`; + assign default `null` to `f` +<7> declare `Map[][] m`; + assign default `null` to `m` [[assignment]] ==== Assignment -Use the `equals` operator (`=`) to assign a value to a variable. Any expression +Use the *assignment operator* to store a value in a variable. Any operation that produces a value can be assigned to any variable as long as the -<> are the same or the resultant -<> can be implicitly <> to -the variable <>. Otherwise, an error will occur. -<> values are shallow-copied when assigned. +<> are the same or the resultant type can be +<> to the variable type. + +*Errors* + +* If the type of value is unable to match the type of variable. *Grammar* [source,ANTLR4] @@ -65,7 +83,7 @@ assignment: ID '=' expression *Examples* -* Variable assignment with an <>. +* Variable assignment with an integer literal. + [source,Painless] ---- @@ -73,10 +91,11 @@ assignment: ID '=' expression <2> i = 10; ---- + -<1> declare `int i` -<2> assign `10` to `i` +<1> declare `int i`; + assign default `int 0` to `i` +<2> assign `int 10` to `i` + -* <> combined with immediate variable assignment. +* Declaration combined with immediate assignment. + [source,Painless] ---- @@ -84,11 +103,12 @@ assignment: ID '=' expression <2> double j = 2.0; ---- + -<1> declare `int i`; assign `10` to `i` -<2> declare `double j`; assign `2.0` to `j` +<1> declare `int i`; + assign `int 10` to `i` +<2> declare `double j`; + assign `double 2.0` to `j` + -* Assignment of one variable to another using -<>. +* Assignment of one variable to another using primitive types. + [source,Painless] ---- @@ -96,11 +116,13 @@ assignment: ID '=' expression <2> int j = i; ---- + -<1> declare `int i`; assign `10` to `i` -<2> declare `int j`; assign `j` to `i` +<1> declare `int i`; + assign `int 10` to `i` +<2> declare `int j`; + access `i` -> `int 10`; + assign `int 10` to `j` + -* Assignment with <> using the -<>. +* Assignment with reference types using the *new instance operator*. + [source,Painless] ---- @@ -108,12 +130,15 @@ assignment: ID '=' expression <2> Map m = new HashMap(); ---- + -<1> declare `ArrayList l`; assign a newly-allocated `Arraylist` to `l` -<2> declare `Map m`; assign a newly-allocated `HashMap` to `m` - with an implicit cast to `Map` +<1> declare `ArrayList l`; + allocate `ArrayList` instance -> `ArrayList reference`; + assign `ArrayList reference` to `l` +<2> declare `Map m`; + allocate `HashMap` instance -> `HashMap reference`; + implicit cast `HashMap reference` to `Map reference` -> `Map reference`; + assign `Map reference` to `m` + -* Assignment of one variable to another using -<>. +* Assignment of one variable to another using reference types. + [source,Painless] ---- @@ -123,8 +148,52 @@ assignment: ID '=' expression <4> m = k; ---- + -<1> declare `List l`; assign a newly-allocated `Arraylist` to `l` - with an implicit cast to `List` -<2> declare `List k`; assign a shallow-copy of `l` to `k` +<1> declare `List l`; + allocate `ArrayList` instance -> `ArrayList reference`; + implicit cast `ArrayList reference` to `List reference` -> `List reference`; + assign `List reference` to `l` +<2> declare `List k`; + access `l` -> `List reference`; + assign `List reference` to `k`; + (note `l` and `k` refer to the same instance known as a shallow-copy) <3> declare `List m`; -<4> assign a shallow-copy of `k` to `m` + assign default `null` to `m` +<4> access `k` -> `List reference`; + assign `List reference` to `m`; + (note `l`, `k`, and `m` refer to the same instance) ++ +* Assignment with an array type variable using the *new array operator*. ++ +[source,Painless] +---- +<1> int[] ia1; +<2> ia1 = new int[2]; +<3> ia1[0] = 1; +<4> int[] ib1 = ia1; +<5> int[][] ic2 = new int[2][5]; +<6> ic2[1][3] = 2; +<7> ic2[0] = ia1; +---- ++ +<1> declare `int[] ia1`; + assign default `null` to `ia1` +<2> allocate `1-d int array` instance with `length [2]` + -> `1-d int array reference`; + assign `1-d int array reference` to `ia1` +<3> access `ia1` -> `1-d int array reference`; + assign `int 1` to `index [0]` of `1-d int array reference` +<4> declare `int[] ib1`; + access `ia1` -> `1-d int array reference`; + assign `1-d int array reference` to `ib1`; + (note `ia1` and `ib1` refer to the same instance known as a shallow copy) +<5> declare `int[][] ic2`; + allocate `2-d int array` instance with `length [2, 5]` + -> `2-d int array reference`; + assign `2-d int array reference` to `ic2` +<6> access `ic2` -> `2-d int array reference`; + assign `int 2` to `index [1, 3]` of `2-d int array reference` +<7> access `ia1` -> `1-d int array reference`; + access `ic2` -> `2-d int array reference`; + assign `1-d int array reference` to + `index [0]` of `2-d int array reference`; + (note `ia1`, `ib1`, and `index [0]` of `ia2` refer to the same instance) From d7040ad7b41633dfe0b5fee993b59b1a49181813 Mon Sep 17 00:00:00 2001 From: Tim Brooks Date: Wed, 23 May 2018 14:38:52 -0600 Subject: [PATCH 04/12] Reintroduce mandatory http pipelining support (#30820) This commit reintroduces 31251c9 and 63a5799. These commits introduced a memory leak and were reverted. This commit brings those commits back and fixes the memory leak by removing unnecessary retain method calls. --- .../migration/migrate_7_0/settings.asciidoc | 10 +- docs/reference/modules/http.asciidoc | 2 - .../http/netty4/Netty4HttpChannel.java | 39 +-- .../netty4/Netty4HttpPipeliningHandler.java | 102 ++++++ .../http/netty4/Netty4HttpRequestHandler.java | 30 +- .../http/netty4/Netty4HttpResponse.java | 37 +++ .../netty4/Netty4HttpServerTransport.java | 20 +- .../pipelining/HttpPipelinedRequest.java | 88 ----- .../pipelining/HttpPipelinedResponse.java | 94 ------ .../pipelining/HttpPipeliningHandler.java | 144 --------- .../http/netty4/Netty4HttpChannelTests.java | 19 +- .../Netty4HttpPipeliningHandlerTests.java | 57 ++-- .../Netty4HttpServerPipeliningTests.java | 82 +---- ...EnabledIT.java => Netty4PipeliningIT.java} | 12 +- .../http/nio/HttpReadWriteHandler.java | 145 +++++---- .../http/nio/HttpWriteOperation.java | 7 +- .../elasticsearch/http/nio/NettyAdaptor.java | 20 +- .../elasticsearch/http/nio/NettyListener.java | 30 +- .../http/nio/NioHttpChannel.java | 10 +- .../http/nio/NioHttpPipeliningHandler.java | 103 ++++++ .../http/nio/NioHttpResponse.java | 37 +++ .../http/nio/NioHttpServerTransport.java | 15 +- .../org/elasticsearch/NioIntegTestCase.java | 5 +- .../http/nio/HttpReadWriteHandlerTests.java | 11 +- .../nio/NioHttpPipeliningHandlerTests.java | 304 ++++++++++++++++++ .../http/nio/NioPipeliningIT.java | 48 ++- .../common/settings/ClusterSettings.java | 1 - .../http/HttpHandlingSettings.java | 9 +- .../http/HttpPipelinedMessage.java | 37 +++ .../http/HttpPipelinedRequest.java | 33 ++ .../http/HttpPipeliningAggregator.java | 81 +++++ .../http/HttpTransportSettings.java | 2 - .../single/SingleNodeDiscoveryIT.java | 1 - .../elasticsearch/test/ESIntegTestCase.java | 2 +- .../test/InternalTestCluster.java | 5 +- .../test/test/InternalTestClusterTests.java | 23 +- .../audit/index/IndexAuditTrailTests.java | 2 +- .../RemoteIndexAuditTrailStartingTests.java | 2 +- 38 files changed, 1001 insertions(+), 668 deletions(-) create mode 100644 modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpPipeliningHandler.java create mode 100644 modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpResponse.java delete mode 100644 modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipelinedRequest.java delete mode 100644 modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipelinedResponse.java delete mode 100644 modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipeliningHandler.java rename modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/{pipelining => }/Netty4HttpPipeliningHandlerTests.java (83%) rename modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/{Netty4PipeliningEnabledIT.java => Netty4PipeliningIT.java} (87%) create mode 100644 plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpPipeliningHandler.java create mode 100644 plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpResponse.java create mode 100644 plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/NioHttpPipeliningHandlerTests.java rename modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningDisabledIT.java => plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/NioPipeliningIT.java (53%) create mode 100644 server/src/main/java/org/elasticsearch/http/HttpPipelinedMessage.java create mode 100644 server/src/main/java/org/elasticsearch/http/HttpPipelinedRequest.java create mode 100644 server/src/main/java/org/elasticsearch/http/HttpPipeliningAggregator.java diff --git a/docs/reference/migration/migrate_7_0/settings.asciidoc b/docs/reference/migration/migrate_7_0/settings.asciidoc index d62d7e6065d..7826afc05fa 100644 --- a/docs/reference/migration/migrate_7_0/settings.asciidoc +++ b/docs/reference/migration/migrate_7_0/settings.asciidoc @@ -29,6 +29,14 @@ [[remove-http-enabled]] ==== Http enabled setting removed -The setting `http.enabled` previously allowed disabling binding to HTTP, only allowing +* The setting `http.enabled` previously allowed disabling binding to HTTP, only allowing use of the transport client. This setting has been removed, as the transport client will be removed in the future, thus requiring HTTP to always be enabled. + +[[remove-http-pipelining-setting]] +==== Http pipelining setting removed + +* The setting `http.pipelining` previously allowed disabling HTTP pipelining support. +This setting has been removed, as disabling http pipelining support on the server +provided little value. The setting `http.pipelining.max_events` can still be used to +limit the number of pipelined requests in-flight. diff --git a/docs/reference/modules/http.asciidoc b/docs/reference/modules/http.asciidoc index 7f29a9db7f6..dab8e813689 100644 --- a/docs/reference/modules/http.asciidoc +++ b/docs/reference/modules/http.asciidoc @@ -96,8 +96,6 @@ and stack traces in response output. Note: When set to `false` and the `error_tr parameter is specified, an error will be returned; when `error_trace` is not specified, a simple message will be returned. Defaults to `true` -|`http.pipelining` |Enable or disable HTTP pipelining, defaults to `true`. - |`http.pipelining.max_events` |The maximum number of events to be queued up in memory before a HTTP connection is closed, defaults to `10000`. |`http.max_warning_header_count` |The maximum number of warning headers in diff --git a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpChannel.java b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpChannel.java index 6e39a7f50d2..cb31d444544 100644 --- a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpChannel.java +++ b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpChannel.java @@ -42,7 +42,6 @@ import org.elasticsearch.common.lease.Releasable; import org.elasticsearch.common.util.concurrent.ThreadContext; import org.elasticsearch.http.HttpHandlingSettings; import org.elasticsearch.http.netty4.cors.Netty4CorsHandler; -import org.elasticsearch.http.netty4.pipelining.HttpPipelinedRequest; import org.elasticsearch.rest.AbstractRestChannel; import org.elasticsearch.rest.RestResponse; import org.elasticsearch.rest.RestStatus; @@ -59,29 +58,24 @@ final class Netty4HttpChannel extends AbstractRestChannel { private final Netty4HttpServerTransport transport; private final Channel channel; private final FullHttpRequest nettyRequest; - private final HttpPipelinedRequest pipelinedRequest; + private final int sequence; private final ThreadContext threadContext; private final HttpHandlingSettings handlingSettings; /** - * @param transport The corresponding NettyHttpServerTransport where this channel belongs to. - * @param request The request that is handled by this channel. - * @param pipelinedRequest If HTTP pipelining is enabled provide the corresponding pipelined request. May be null if - * HTTP pipelining is disabled. - * @param handlingSettings true iff error messages should include stack traces. - * @param threadContext the thread context for the channel + * @param transport The corresponding NettyHttpServerTransport where this channel belongs to. + * @param request The request that is handled by this channel. + * @param sequence The pipelining sequence number for this request + * @param handlingSettings true if error messages should include stack traces. + * @param threadContext the thread context for the channel */ - Netty4HttpChannel( - final Netty4HttpServerTransport transport, - final Netty4HttpRequest request, - final HttpPipelinedRequest pipelinedRequest, - final HttpHandlingSettings handlingSettings, - final ThreadContext threadContext) { + Netty4HttpChannel(Netty4HttpServerTransport transport, Netty4HttpRequest request, int sequence, HttpHandlingSettings handlingSettings, + ThreadContext threadContext) { super(request, handlingSettings.getDetailedErrorsEnabled()); this.transport = transport; this.channel = request.getChannel(); this.nettyRequest = request.request(); - this.pipelinedRequest = pipelinedRequest; + this.sequence = sequence; this.threadContext = threadContext; this.handlingSettings = handlingSettings; } @@ -129,7 +123,7 @@ final class Netty4HttpChannel extends AbstractRestChannel { final ChannelPromise promise = channel.newPromise(); if (releaseContent) { - promise.addListener(f -> ((Releasable)content).close()); + promise.addListener(f -> ((Releasable) content).close()); } if (releaseBytesStreamOutput) { @@ -140,13 +134,9 @@ final class Netty4HttpChannel extends AbstractRestChannel { promise.addListener(ChannelFutureListener.CLOSE); } - final Object msg; - if (pipelinedRequest != null) { - msg = pipelinedRequest.createHttpResponse(resp, promise); - } else { - msg = resp; - } - channel.writeAndFlush(msg, promise); + Netty4HttpResponse newResponse = new Netty4HttpResponse(sequence, resp); + + channel.writeAndFlush(newResponse, promise); releaseContent = false; releaseBytesStreamOutput = false; } finally { @@ -156,9 +146,6 @@ final class Netty4HttpChannel extends AbstractRestChannel { if (releaseBytesStreamOutput) { bytesOutputOrNull().close(); } - if (pipelinedRequest != null) { - pipelinedRequest.release(); - } } } diff --git a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpPipeliningHandler.java b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpPipeliningHandler.java new file mode 100644 index 00000000000..52dfbff6d3e --- /dev/null +++ b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpPipeliningHandler.java @@ -0,0 +1,102 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.http.netty4; + +import io.netty.channel.ChannelDuplexHandler; +import io.netty.channel.ChannelHandlerContext; +import io.netty.channel.ChannelPromise; +import io.netty.handler.codec.http.LastHttpContent; +import org.apache.logging.log4j.Logger; +import org.elasticsearch.common.collect.Tuple; +import org.elasticsearch.http.HttpPipelinedRequest; +import org.elasticsearch.http.HttpPipeliningAggregator; +import org.elasticsearch.transport.netty4.Netty4Utils; + +import java.nio.channels.ClosedChannelException; +import java.util.Collections; +import java.util.List; + +/** + * Implements HTTP pipelining ordering, ensuring that responses are completely served in the same order as their corresponding requests. + */ +public class Netty4HttpPipeliningHandler extends ChannelDuplexHandler { + + private final Logger logger; + private final HttpPipeliningAggregator aggregator; + + /** + * Construct a new pipelining handler; this handler should be used downstream of HTTP decoding/aggregation. + * + * @param logger for logging unexpected errors + * @param maxEventsHeld the maximum number of channel events that will be retained prior to aborting the channel connection; this is + * required as events cannot queue up indefinitely + */ + public Netty4HttpPipeliningHandler(Logger logger, final int maxEventsHeld) { + this.logger = logger; + this.aggregator = new HttpPipeliningAggregator<>(maxEventsHeld); + } + + @Override + public void channelRead(final ChannelHandlerContext ctx, final Object msg) { + if (msg instanceof LastHttpContent) { + HttpPipelinedRequest pipelinedRequest = aggregator.read(((LastHttpContent) msg)); + ctx.fireChannelRead(pipelinedRequest); + } else { + ctx.fireChannelRead(msg); + } + } + + @Override + public void write(final ChannelHandlerContext ctx, final Object msg, final ChannelPromise promise) { + assert msg instanceof Netty4HttpResponse : "Message must be type: " + Netty4HttpResponse.class; + Netty4HttpResponse response = (Netty4HttpResponse) msg; + boolean success = false; + try { + List> readyResponses = aggregator.write(response, promise); + for (Tuple readyResponse : readyResponses) { + ctx.write(readyResponse.v1().getResponse(), readyResponse.v2()); + } + success = true; + } catch (IllegalStateException e) { + ctx.channel().close(); + } finally { + if (success == false) { + promise.setFailure(new ClosedChannelException()); + } + } + } + + @Override + public void close(ChannelHandlerContext ctx, ChannelPromise promise) { + List> inflightResponses = aggregator.removeAllInflightResponses(); + + if (inflightResponses.isEmpty() == false) { + ClosedChannelException closedChannelException = new ClosedChannelException(); + for (Tuple inflightResponse : inflightResponses) { + try { + inflightResponse.v2().setFailure(closedChannelException); + } catch (RuntimeException e) { + logger.error("unexpected error while releasing pipelined http responses", e); + } + } + } + ctx.close(promise); + } +} diff --git a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpRequestHandler.java b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpRequestHandler.java index 74429c8dda9..c3a010226a4 100644 --- a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpRequestHandler.java +++ b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpRequestHandler.java @@ -30,41 +30,30 @@ import io.netty.handler.codec.http.FullHttpRequest; import io.netty.handler.codec.http.HttpHeaders; import org.elasticsearch.common.util.concurrent.ThreadContext; import org.elasticsearch.http.HttpHandlingSettings; -import org.elasticsearch.http.netty4.pipelining.HttpPipelinedRequest; +import org.elasticsearch.http.HttpPipelinedRequest; import org.elasticsearch.rest.RestRequest; import org.elasticsearch.transport.netty4.Netty4Utils; import java.util.Collections; @ChannelHandler.Sharable -class Netty4HttpRequestHandler extends SimpleChannelInboundHandler { +class Netty4HttpRequestHandler extends SimpleChannelInboundHandler> { private final Netty4HttpServerTransport serverTransport; private final HttpHandlingSettings handlingSettings; - private final boolean httpPipeliningEnabled; private final ThreadContext threadContext; Netty4HttpRequestHandler(Netty4HttpServerTransport serverTransport, HttpHandlingSettings handlingSettings, ThreadContext threadContext) { this.serverTransport = serverTransport; - this.httpPipeliningEnabled = serverTransport.pipelining; this.handlingSettings = handlingSettings; this.threadContext = threadContext; } @Override - protected void channelRead0(ChannelHandlerContext ctx, Object msg) throws Exception { - final FullHttpRequest request; - final HttpPipelinedRequest pipelinedRequest; - if (this.httpPipeliningEnabled && msg instanceof HttpPipelinedRequest) { - pipelinedRequest = (HttpPipelinedRequest) msg; - request = (FullHttpRequest) pipelinedRequest.last(); - } else { - pipelinedRequest = null; - request = (FullHttpRequest) msg; - } + protected void channelRead0(ChannelHandlerContext ctx, HttpPipelinedRequest msg) throws Exception { + final FullHttpRequest request = msg.getRequest(); - boolean success = false; try { final FullHttpRequest copy = @@ -111,7 +100,7 @@ class Netty4HttpRequestHandler extends SimpleChannelInboundHandler { Netty4HttpChannel innerChannel; try { innerChannel = - new Netty4HttpChannel(serverTransport, httpRequest, pipelinedRequest, handlingSettings, threadContext); + new Netty4HttpChannel(serverTransport, httpRequest, msg.getSequence(), handlingSettings, threadContext); } catch (final IllegalArgumentException e) { if (badRequestCause == null) { badRequestCause = e; @@ -126,7 +115,7 @@ class Netty4HttpRequestHandler extends SimpleChannelInboundHandler { copy, ctx.channel()); innerChannel = - new Netty4HttpChannel(serverTransport, innerRequest, pipelinedRequest, handlingSettings, threadContext); + new Netty4HttpChannel(serverTransport, innerRequest, msg.getSequence(), handlingSettings, threadContext); } channel = innerChannel; } @@ -138,12 +127,9 @@ class Netty4HttpRequestHandler extends SimpleChannelInboundHandler { } else { serverTransport.dispatchRequest(httpRequest, channel); } - success = true; } finally { - // the request is otherwise released in case of dispatch - if (success == false && pipelinedRequest != null) { - pipelinedRequest.release(); - } + // As we have copied the buffer, we can release the request + request.release(); } } diff --git a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpResponse.java b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpResponse.java new file mode 100644 index 00000000000..779c9125a2e --- /dev/null +++ b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpResponse.java @@ -0,0 +1,37 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.http.netty4; + +import io.netty.handler.codec.http.FullHttpResponse; +import org.elasticsearch.http.HttpPipelinedMessage; + +public class Netty4HttpResponse extends HttpPipelinedMessage { + + private final FullHttpResponse response; + + public Netty4HttpResponse(int sequence, FullHttpResponse response) { + super(sequence); + this.response = response; + } + + public FullHttpResponse getResponse() { + return response; + } +} diff --git a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpServerTransport.java b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpServerTransport.java index 8e5bace46aa..45e889797bd 100644 --- a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpServerTransport.java +++ b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/Netty4HttpServerTransport.java @@ -62,7 +62,6 @@ import org.elasticsearch.http.HttpStats; import org.elasticsearch.http.netty4.cors.Netty4CorsConfig; import org.elasticsearch.http.netty4.cors.Netty4CorsConfigBuilder; import org.elasticsearch.http.netty4.cors.Netty4CorsHandler; -import org.elasticsearch.http.netty4.pipelining.HttpPipeliningHandler; import org.elasticsearch.rest.RestUtils; import org.elasticsearch.threadpool.ThreadPool; import org.elasticsearch.transport.netty4.Netty4OpenChannelsHandler; @@ -99,7 +98,6 @@ import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_TCP_NO_D import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_TCP_RECEIVE_BUFFER_SIZE; import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_TCP_REUSE_ADDRESS; import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_TCP_SEND_BUFFER_SIZE; -import static org.elasticsearch.http.HttpTransportSettings.SETTING_PIPELINING; import static org.elasticsearch.http.HttpTransportSettings.SETTING_PIPELINING_MAX_EVENTS; import static org.elasticsearch.http.netty4.cors.Netty4CorsHandler.ANY_ORIGIN; @@ -162,8 +160,6 @@ public class Netty4HttpServerTransport extends AbstractHttpServerTransport { protected final int workerCount; - protected final boolean pipelining; - protected final int pipeliningMaxEvents; /** @@ -204,6 +200,7 @@ public class Netty4HttpServerTransport extends AbstractHttpServerTransport { this.maxChunkSize = SETTING_HTTP_MAX_CHUNK_SIZE.get(settings); this.maxHeaderSize = SETTING_HTTP_MAX_HEADER_SIZE.get(settings); this.maxInitialLineLength = SETTING_HTTP_MAX_INITIAL_LINE_LENGTH.get(settings); + this.pipeliningMaxEvents = SETTING_PIPELINING_MAX_EVENTS.get(settings); this.httpHandlingSettings = new HttpHandlingSettings(Math.toIntExact(maxContentLength.getBytes()), Math.toIntExact(maxChunkSize.getBytes()), Math.toIntExact(maxHeaderSize.getBytes()), @@ -211,7 +208,8 @@ public class Netty4HttpServerTransport extends AbstractHttpServerTransport { SETTING_HTTP_RESET_COOKIES.get(settings), SETTING_HTTP_COMPRESSION.get(settings), SETTING_HTTP_COMPRESSION_LEVEL.get(settings), - SETTING_HTTP_DETAILED_ERRORS_ENABLED.get(settings)); + SETTING_HTTP_DETAILED_ERRORS_ENABLED.get(settings), + pipeliningMaxEvents); this.maxCompositeBufferComponents = SETTING_HTTP_NETTY_MAX_COMPOSITE_BUFFER_COMPONENTS.get(settings); this.workerCount = SETTING_HTTP_WORKER_COUNT.get(settings); @@ -226,14 +224,12 @@ public class Netty4HttpServerTransport extends AbstractHttpServerTransport { ByteSizeValue receivePredictor = SETTING_HTTP_NETTY_RECEIVE_PREDICTOR_SIZE.get(settings); recvByteBufAllocator = new FixedRecvByteBufAllocator(receivePredictor.bytesAsInt()); - this.pipelining = SETTING_PIPELINING.get(settings); - this.pipeliningMaxEvents = SETTING_PIPELINING_MAX_EVENTS.get(settings); this.corsConfig = buildCorsConfig(settings); logger.debug("using max_chunk_size[{}], max_header_size[{}], max_initial_line_length[{}], max_content_length[{}], " + - "receive_predictor[{}], max_composite_buffer_components[{}], pipelining[{}], pipelining_max_events[{}]", - maxChunkSize, maxHeaderSize, maxInitialLineLength, this.maxContentLength, receivePredictor, maxCompositeBufferComponents, - pipelining, pipeliningMaxEvents); + "receive_predictor[{}], max_composite_buffer_components[{}], pipelining_max_events[{}]", + maxChunkSize, maxHeaderSize, maxInitialLineLength, maxContentLength, receivePredictor, maxCompositeBufferComponents, + pipeliningMaxEvents); } public Settings settings() { @@ -452,9 +448,7 @@ public class Netty4HttpServerTransport extends AbstractHttpServerTransport { if (SETTING_CORS_ENABLED.get(transport.settings())) { ch.pipeline().addLast("cors", new Netty4CorsHandler(transport.getCorsConfig())); } - if (transport.pipelining) { - ch.pipeline().addLast("pipelining", new HttpPipeliningHandler(transport.logger, transport.pipeliningMaxEvents)); - } + ch.pipeline().addLast("pipelining", new Netty4HttpPipeliningHandler(transport.logger, transport.pipeliningMaxEvents)); ch.pipeline().addLast("handler", requestHandler); } diff --git a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipelinedRequest.java b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipelinedRequest.java deleted file mode 100644 index be1669c60c2..00000000000 --- a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipelinedRequest.java +++ /dev/null @@ -1,88 +0,0 @@ -/* - * Licensed to Elasticsearch under one or more contributor - * license agreements. See the NOTICE file distributed with - * this work for additional information regarding copyright - * ownership. Elasticsearch 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 - * - * 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. - */ - -package org.elasticsearch.http.netty4.pipelining; - -import io.netty.channel.ChannelPromise; -import io.netty.handler.codec.http.FullHttpResponse; -import io.netty.handler.codec.http.LastHttpContent; -import io.netty.util.ReferenceCounted; - -/** - * Permits downstream channel events to be ordered and signalled as to whether more are to come for - * a given sequence. - */ -public class HttpPipelinedRequest implements ReferenceCounted { - - private final LastHttpContent last; - private final int sequence; - - public HttpPipelinedRequest(final LastHttpContent last, final int sequence) { - this.last = last; - this.sequence = sequence; - } - - public LastHttpContent last() { - return last; - } - - public HttpPipelinedResponse createHttpResponse(final FullHttpResponse response, final ChannelPromise promise) { - return new HttpPipelinedResponse(response, promise, sequence); - } - - @Override - public int refCnt() { - return last.refCnt(); - } - - @Override - public ReferenceCounted retain() { - last.retain(); - return this; - } - - @Override - public ReferenceCounted retain(int increment) { - last.retain(increment); - return this; - } - - @Override - public ReferenceCounted touch() { - last.touch(); - return this; - } - - @Override - public ReferenceCounted touch(Object hint) { - last.touch(hint); - return this; - } - - @Override - public boolean release() { - return last.release(); - } - - @Override - public boolean release(int decrement) { - return last.release(decrement); - } - -} diff --git a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipelinedResponse.java b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipelinedResponse.java deleted file mode 100644 index 6b6db94d69a..00000000000 --- a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipelinedResponse.java +++ /dev/null @@ -1,94 +0,0 @@ -package org.elasticsearch.http.netty4.pipelining; - -/* - * Licensed to Elasticsearch under one or more contributor - * license agreements. See the NOTICE file distributed with - * this work for additional information regarding copyright - * ownership. Elasticsearch 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 - * - * 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. - */ - -import io.netty.channel.ChannelPromise; -import io.netty.handler.codec.http.FullHttpResponse; -import io.netty.util.ReferenceCounted; - -class HttpPipelinedResponse implements Comparable, ReferenceCounted { - - private final FullHttpResponse response; - private final ChannelPromise promise; - private final int sequence; - - HttpPipelinedResponse(FullHttpResponse response, ChannelPromise promise, int sequence) { - this.response = response; - this.promise = promise; - this.sequence = sequence; - } - - public FullHttpResponse response() { - return response; - } - - public ChannelPromise promise() { - return promise; - } - - public int sequence() { - return sequence; - } - - @Override - public int compareTo(HttpPipelinedResponse o) { - return Integer.compare(sequence, o.sequence); - } - - @Override - public int refCnt() { - return response.refCnt(); - } - - @Override - public ReferenceCounted retain() { - response.retain(); - return this; - } - - @Override - public ReferenceCounted retain(int increment) { - response.retain(increment); - return this; - } - - @Override - public ReferenceCounted touch() { - response.touch(); - return this; - } - - @Override - public ReferenceCounted touch(Object hint) { - response.touch(hint); - return this; - } - - @Override - public boolean release() { - return response.release(); - } - - @Override - public boolean release(int decrement) { - return response.release(decrement); - } - -} diff --git a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipeliningHandler.java b/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipeliningHandler.java deleted file mode 100644 index a90027c8148..00000000000 --- a/modules/transport-netty4/src/main/java/org/elasticsearch/http/netty4/pipelining/HttpPipeliningHandler.java +++ /dev/null @@ -1,144 +0,0 @@ -/* - * Licensed to Elasticsearch under one or more contributor - * license agreements. See the NOTICE file distributed with - * this work for additional information regarding copyright - * ownership. Elasticsearch 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 - * - * 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. - */ - -package org.elasticsearch.http.netty4.pipelining; - -import io.netty.channel.ChannelDuplexHandler; -import io.netty.channel.ChannelHandlerContext; -import io.netty.channel.ChannelPromise; -import io.netty.handler.codec.http.LastHttpContent; -import org.apache.logging.log4j.Logger; -import org.elasticsearch.transport.netty4.Netty4Utils; - -import java.nio.channels.ClosedChannelException; -import java.util.Collections; -import java.util.PriorityQueue; - -/** - * Implements HTTP pipelining ordering, ensuring that responses are completely served in the same order as their corresponding requests. - */ -public class HttpPipeliningHandler extends ChannelDuplexHandler { - - // we use a priority queue so that responses are ordered by their sequence number - private final PriorityQueue holdingQueue; - - private final Logger logger; - private final int maxEventsHeld; - - /* - * The current read and write sequence numbers. Read sequence numbers are attached to requests in the order they are read from the - * channel, and then transferred to responses. A response is not written to the channel context until its sequence number matches the - * current write sequence, implying that all preceding messages have been written. - */ - private int readSequence; - private int writeSequence; - - /** - * Construct a new pipelining handler; this handler should be used downstream of HTTP decoding/aggregation. - * - * @param logger for logging unexpected errors - * @param maxEventsHeld the maximum number of channel events that will be retained prior to aborting the channel connection; this is - * required as events cannot queue up indefinitely - */ - public HttpPipeliningHandler(Logger logger, final int maxEventsHeld) { - this.logger = logger; - this.maxEventsHeld = maxEventsHeld; - this.holdingQueue = new PriorityQueue<>(1); - } - - @Override - public void channelRead(final ChannelHandlerContext ctx, final Object msg) throws Exception { - if (msg instanceof LastHttpContent) { - ctx.fireChannelRead(new HttpPipelinedRequest(((LastHttpContent) msg).retain(), readSequence++)); - } else { - ctx.fireChannelRead(msg); - } - } - - @Override - public void write(final ChannelHandlerContext ctx, final Object msg, final ChannelPromise promise) throws Exception { - if (msg instanceof HttpPipelinedResponse) { - final HttpPipelinedResponse current = (HttpPipelinedResponse) msg; - /* - * We attach the promise to the response. When we invoke a write on the channel with the response, we must ensure that we invoke - * the write methods that accept the same promise that we have attached to the response otherwise as the response proceeds - * through the handler pipeline a different promise will be used until reaching this handler. Therefore, we assert here that the - * attached promise is identical to the provided promise as a safety mechanism that we are respecting this. - */ - assert current.promise() == promise; - - boolean channelShouldClose = false; - - synchronized (holdingQueue) { - if (holdingQueue.size() < maxEventsHeld) { - holdingQueue.add(current); - - while (!holdingQueue.isEmpty()) { - /* - * Since the response with the lowest sequence number is the top of the priority queue, we know if its sequence - * number does not match the current write sequence number then we have not processed all preceding responses yet. - */ - final HttpPipelinedResponse top = holdingQueue.peek(); - if (top.sequence() != writeSequence) { - break; - } - holdingQueue.remove(); - /* - * We must use the promise attached to the response; this is necessary since are going to hold a response until all - * responses that precede it in the pipeline are written first. Note that the promise from the method invocation is - * not ignored, it will already be attached to an existing response and consumed when that response is drained. - */ - ctx.write(top.response(), top.promise()); - writeSequence++; - } - } else { - channelShouldClose = true; - } - } - - if (channelShouldClose) { - try { - Netty4Utils.closeChannels(Collections.singletonList(ctx.channel())); - } finally { - current.release(); - promise.setSuccess(); - } - } - } else { - ctx.write(msg, promise); - } - } - - @Override - public void close(ChannelHandlerContext ctx, ChannelPromise promise) throws Exception { - if (holdingQueue.isEmpty() == false) { - ClosedChannelException closedChannelException = new ClosedChannelException(); - HttpPipelinedResponse pipelinedResponse; - while ((pipelinedResponse = holdingQueue.poll()) != null) { - try { - pipelinedResponse.release(); - pipelinedResponse.promise().setFailure(closedChannelException); - } catch (Exception e) { - logger.error("unexpected error while releasing pipelined http responses", e); - } - } - } - ctx.close(promise); - } -} diff --git a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpChannelTests.java b/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpChannelTests.java index 0ef1ea585b1..7c5b35a3229 100644 --- a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpChannelTests.java +++ b/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpChannelTests.java @@ -60,7 +60,6 @@ import org.elasticsearch.http.HttpHandlingSettings; import org.elasticsearch.http.HttpTransportSettings; import org.elasticsearch.http.NullDispatcher; import org.elasticsearch.http.netty4.cors.Netty4CorsHandler; -import org.elasticsearch.http.netty4.pipelining.HttpPipelinedRequest; import org.elasticsearch.indices.breaker.NoneCircuitBreakerService; import org.elasticsearch.rest.BytesRestResponse; import org.elasticsearch.rest.RestResponse; @@ -212,12 +211,12 @@ public class Netty4HttpChannelTests extends ESTestCase { final FullHttpRequest httpRequest = new DefaultFullHttpRequest(HttpVersion.HTTP_1_1, HttpMethod.GET, "/"); httpRequest.headers().add(HttpHeaderNames.ORIGIN, "remote"); final WriteCapturingChannel writeCapturingChannel = new WriteCapturingChannel(); - Netty4HttpRequest request = new Netty4HttpRequest(xContentRegistry(), httpRequest, writeCapturingChannel); + final Netty4HttpRequest request = new Netty4HttpRequest(xContentRegistry(), httpRequest, writeCapturingChannel); HttpHandlingSettings handlingSettings = httpServerTransport.httpHandlingSettings; // send a response Netty4HttpChannel channel = - new Netty4HttpChannel(httpServerTransport, request, null, handlingSettings, threadPool.getThreadContext()); + new Netty4HttpChannel(httpServerTransport, request, 1, handlingSettings, threadPool.getThreadContext()); TestResponse resp = new TestResponse(); final String customHeader = "custom-header"; final String customHeaderValue = "xyz"; @@ -227,7 +226,7 @@ public class Netty4HttpChannelTests extends ESTestCase { // inspect what was written List writtenObjects = writeCapturingChannel.getWrittenObjects(); assertThat(writtenObjects.size(), is(1)); - HttpResponse response = (HttpResponse) writtenObjects.get(0); + HttpResponse response = ((Netty4HttpResponse) writtenObjects.get(0)).getResponse(); assertThat(response.headers().get("non-existent-header"), nullValue()); assertThat(response.headers().get(customHeader), equalTo(customHeaderValue)); assertThat(response.headers().get(HttpHeaderNames.CONTENT_LENGTH), equalTo(Integer.toString(resp.content().length()))); @@ -243,10 +242,9 @@ public class Netty4HttpChannelTests extends ESTestCase { final FullHttpRequest httpRequest = new DefaultFullHttpRequest(HttpVersion.HTTP_1_1, HttpMethod.GET, "/"); final EmbeddedChannel embeddedChannel = new EmbeddedChannel(); final Netty4HttpRequest request = new Netty4HttpRequest(registry, httpRequest, embeddedChannel); - final HttpPipelinedRequest pipelinedRequest = randomBoolean() ? new HttpPipelinedRequest(request.request(), 1) : null; HttpHandlingSettings handlingSettings = httpServerTransport.httpHandlingSettings; final Netty4HttpChannel channel = - new Netty4HttpChannel(httpServerTransport, request, pipelinedRequest, handlingSettings, threadPool.getThreadContext()); + new Netty4HttpChannel(httpServerTransport, request, 1, handlingSettings, threadPool.getThreadContext()); final TestResponse response = new TestResponse(bigArrays); assertThat(response.content(), instanceOf(Releasable.class)); embeddedChannel.close(); @@ -263,10 +261,9 @@ public class Netty4HttpChannelTests extends ESTestCase { final FullHttpRequest httpRequest = new DefaultFullHttpRequest(HttpVersion.HTTP_1_1, HttpMethod.GET, "/"); final EmbeddedChannel embeddedChannel = new EmbeddedChannel(); final Netty4HttpRequest request = new Netty4HttpRequest(registry, httpRequest, embeddedChannel); - final HttpPipelinedRequest pipelinedRequest = randomBoolean() ? new HttpPipelinedRequest(request.request(), 1) : null; HttpHandlingSettings handlingSettings = httpServerTransport.httpHandlingSettings; final Netty4HttpChannel channel = - new Netty4HttpChannel(httpServerTransport, request, pipelinedRequest, handlingSettings, threadPool.getThreadContext()); + new Netty4HttpChannel(httpServerTransport, request, 1, handlingSettings, threadPool.getThreadContext()); final BytesRestResponse response = new BytesRestResponse(RestStatus.INTERNAL_SERVER_ERROR, JsonXContent.contentBuilder().startObject().endObject()); assertThat(response.content(), not(instanceOf(Releasable.class))); @@ -312,7 +309,7 @@ public class Netty4HttpChannelTests extends ESTestCase { assertTrue(embeddedChannel.isOpen()); HttpHandlingSettings handlingSettings = httpServerTransport.httpHandlingSettings; final Netty4HttpChannel channel = - new Netty4HttpChannel(httpServerTransport, request, null, handlingSettings, threadPool.getThreadContext()); + new Netty4HttpChannel(httpServerTransport, request, 1, handlingSettings, threadPool.getThreadContext()); final TestResponse resp = new TestResponse(); channel.sendResponse(resp); assertThat(embeddedChannel.isOpen(), equalTo(!close)); @@ -340,13 +337,13 @@ public class Netty4HttpChannelTests extends ESTestCase { HttpHandlingSettings handlingSettings = httpServerTransport.httpHandlingSettings; Netty4HttpChannel channel = - new Netty4HttpChannel(httpServerTransport, request, null, handlingSettings, threadPool.getThreadContext()); + new Netty4HttpChannel(httpServerTransport, request, 1, handlingSettings, threadPool.getThreadContext()); channel.sendResponse(new TestResponse()); // get the response List writtenObjects = writeCapturingChannel.getWrittenObjects(); assertThat(writtenObjects.size(), is(1)); - return (FullHttpResponse) writtenObjects.get(0); + return ((Netty4HttpResponse) writtenObjects.get(0)).getResponse(); } } diff --git a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/pipelining/Netty4HttpPipeliningHandlerTests.java b/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpPipeliningHandlerTests.java similarity index 83% rename from modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/pipelining/Netty4HttpPipeliningHandlerTests.java rename to modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpPipeliningHandlerTests.java index ffb6c8fb356..21151304424 100644 --- a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/pipelining/Netty4HttpPipeliningHandlerTests.java +++ b/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpPipeliningHandlerTests.java @@ -17,7 +17,7 @@ * under the License. */ -package org.elasticsearch.http.netty4.pipelining; +package org.elasticsearch.http.netty4; import io.netty.buffer.ByteBuf; import io.netty.buffer.ByteBufUtil; @@ -37,6 +37,7 @@ import io.netty.handler.codec.http.HttpVersion; import io.netty.handler.codec.http.LastHttpContent; import io.netty.handler.codec.http.QueryStringDecoder; import org.elasticsearch.common.Randomness; +import org.elasticsearch.http.HttpPipelinedRequest; import org.elasticsearch.test.ESTestCase; import org.junit.After; @@ -62,7 +63,8 @@ import static org.hamcrest.core.Is.is; public class Netty4HttpPipeliningHandlerTests extends ESTestCase { - private final ExecutorService executorService = Executors.newFixedThreadPool(randomIntBetween(4, 8)); + private final ExecutorService handlerService = Executors.newFixedThreadPool(randomIntBetween(4, 8)); + private final ExecutorService eventLoopService = Executors.newFixedThreadPool(1); private final Map waitingRequests = new ConcurrentHashMap<>(); private final Map finishingRequests = new ConcurrentHashMap<>(); @@ -79,15 +81,19 @@ public class Netty4HttpPipeliningHandlerTests extends ESTestCase { } private void shutdownExecutorService() throws InterruptedException { - if (!executorService.isShutdown()) { - executorService.shutdown(); - executorService.awaitTermination(10, TimeUnit.SECONDS); + if (!handlerService.isShutdown()) { + handlerService.shutdown(); + handlerService.awaitTermination(10, TimeUnit.SECONDS); + } + if (!eventLoopService.isShutdown()) { + eventLoopService.shutdown(); + eventLoopService.awaitTermination(10, TimeUnit.SECONDS); } } public void testThatPipeliningWorksWithFastSerializedRequests() throws InterruptedException { final int numberOfRequests = randomIntBetween(2, 128); - final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new HttpPipeliningHandler(logger, numberOfRequests), + final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new Netty4HttpPipeliningHandler(logger, numberOfRequests), new WorkEmulatorHandler()); for (int i = 0; i < numberOfRequests; i++) { @@ -114,7 +120,7 @@ public class Netty4HttpPipeliningHandlerTests extends ESTestCase { public void testThatPipeliningWorksWhenSlowRequestsInDifferentOrder() throws InterruptedException { final int numberOfRequests = randomIntBetween(2, 128); - final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new HttpPipeliningHandler(logger, numberOfRequests), + final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new Netty4HttpPipeliningHandler(logger, numberOfRequests), new WorkEmulatorHandler()); for (int i = 0; i < numberOfRequests; i++) { @@ -147,7 +153,7 @@ public class Netty4HttpPipeliningHandlerTests extends ESTestCase { final EmbeddedChannel embeddedChannel = new EmbeddedChannel( new AggregateUrisAndHeadersHandler(), - new HttpPipeliningHandler(logger, numberOfRequests), + new Netty4HttpPipeliningHandler(logger, numberOfRequests), new WorkEmulatorHandler()); for (int i = 0; i < numberOfRequests; i++) { @@ -176,7 +182,7 @@ public class Netty4HttpPipeliningHandlerTests extends ESTestCase { public void testThatPipeliningClosesConnectionWithTooManyEvents() throws InterruptedException { final int numberOfRequests = randomIntBetween(2, 128); - final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new HttpPipeliningHandler(logger, numberOfRequests), + final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new Netty4HttpPipeliningHandler(logger, numberOfRequests), new WorkEmulatorHandler()); for (int i = 0; i < 1 + numberOfRequests + 1; i++) { @@ -184,7 +190,7 @@ public class Netty4HttpPipeliningHandlerTests extends ESTestCase { } final List latches = new ArrayList<>(); - final List requests = IntStream.range(1, numberOfRequests + 1).mapToObj(r -> r).collect(Collectors.toList()); + final List requests = IntStream.range(1, numberOfRequests + 1).boxed().collect(Collectors.toList()); Randomness.shuffle(requests); for (final Integer request : requests) { @@ -205,25 +211,26 @@ public class Netty4HttpPipeliningHandlerTests extends ESTestCase { public void testPipeliningRequestsAreReleased() throws InterruptedException { final int numberOfRequests = 10; final EmbeddedChannel embeddedChannel = - new EmbeddedChannel(new HttpPipeliningHandler(logger, numberOfRequests + 1)); + new EmbeddedChannel(new Netty4HttpPipeliningHandler(logger, numberOfRequests + 1)); for (int i = 0; i < numberOfRequests; i++) { embeddedChannel.writeInbound(createHttpRequest("/" + i)); } - HttpPipelinedRequest inbound; - ArrayList requests = new ArrayList<>(); + HttpPipelinedRequest inbound; + ArrayList> requests = new ArrayList<>(); while ((inbound = embeddedChannel.readInbound()) != null) { requests.add(inbound); } ArrayList promises = new ArrayList<>(); for (int i = 1; i < requests.size(); ++i) { - final DefaultFullHttpResponse httpResponse = new DefaultFullHttpResponse(HTTP_1_1, OK); + final FullHttpResponse httpResponse = new DefaultFullHttpResponse(HTTP_1_1, OK); ChannelPromise promise = embeddedChannel.newPromise(); promises.add(promise); - HttpPipelinedResponse response = requests.get(i).createHttpResponse(httpResponse, promise); - embeddedChannel.writeAndFlush(response, promise); + int sequence = requests.get(i).getSequence(); + Netty4HttpResponse resp = new Netty4HttpResponse(sequence, httpResponse); + embeddedChannel.writeAndFlush(resp, promise); } for (ChannelPromise promise : promises) { @@ -260,14 +267,14 @@ public class Netty4HttpPipeliningHandlerTests extends ESTestCase { } - private class WorkEmulatorHandler extends SimpleChannelInboundHandler { + private class WorkEmulatorHandler extends SimpleChannelInboundHandler> { @Override - protected void channelRead0(final ChannelHandlerContext ctx, final HttpPipelinedRequest pipelinedRequest) throws Exception { + protected void channelRead0(final ChannelHandlerContext ctx, HttpPipelinedRequest pipelinedRequest) { + LastHttpContent request = pipelinedRequest.getRequest(); final QueryStringDecoder decoder; - if (pipelinedRequest.last() instanceof FullHttpRequest) { - final FullHttpRequest fullHttpRequest = (FullHttpRequest) pipelinedRequest.last(); - decoder = new QueryStringDecoder(fullHttpRequest.uri()); + if (request instanceof FullHttpRequest) { + decoder = new QueryStringDecoder(((FullHttpRequest)request).uri()); } else { decoder = new QueryStringDecoder(AggregateUrisAndHeadersHandler.QUEUE_URI.poll()); } @@ -282,12 +289,14 @@ public class Netty4HttpPipeliningHandlerTests extends ESTestCase { final CountDownLatch finishingLatch = new CountDownLatch(1); finishingRequests.put(uri, finishingLatch); - executorService.submit(() -> { + handlerService.submit(() -> { try { waitingLatch.await(1000, TimeUnit.SECONDS); final ChannelPromise promise = ctx.newPromise(); - ctx.write(pipelinedRequest.createHttpResponse(httpResponse, promise), promise); - finishingLatch.countDown(); + eventLoopService.submit(() -> { + ctx.write(new Netty4HttpResponse(pipelinedRequest.getSequence(), httpResponse), promise); + finishingLatch.countDown(); + }); } catch (InterruptedException e) { fail(e.toString()); } diff --git a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpServerPipeliningTests.java b/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpServerPipeliningTests.java index 0eb14a8a76e..f2b28b90918 100644 --- a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpServerPipeliningTests.java +++ b/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4HttpServerPipeliningTests.java @@ -38,9 +38,9 @@ import org.elasticsearch.common.transport.TransportAddress; import org.elasticsearch.common.util.MockBigArrays; import org.elasticsearch.common.util.MockPageCacheRecycler; import org.elasticsearch.common.util.concurrent.ThreadContext; +import org.elasticsearch.http.HttpPipelinedRequest; import org.elasticsearch.http.HttpServerTransport; import org.elasticsearch.http.NullDispatcher; -import org.elasticsearch.http.netty4.pipelining.HttpPipelinedRequest; import org.elasticsearch.indices.breaker.NoneCircuitBreakerService; import org.elasticsearch.test.ESTestCase; import org.elasticsearch.threadpool.TestThreadPool; @@ -52,16 +52,11 @@ import java.nio.charset.StandardCharsets; import java.util.ArrayList; import java.util.Collection; import java.util.Collections; -import java.util.HashSet; import java.util.List; -import java.util.Set; import java.util.concurrent.ExecutorService; import java.util.concurrent.Executors; -import static org.elasticsearch.test.hamcrest.RegexMatcher.matches; -import static org.hamcrest.CoreMatchers.equalTo; import static org.hamcrest.Matchers.contains; -import static org.hamcrest.Matchers.hasSize; /** * This test just tests, if he pipelining works in general with out any connection the Elasticsearch handler @@ -85,9 +80,8 @@ public class Netty4HttpServerPipeliningTests extends ESTestCase { } } - public void testThatHttpPipeliningWorksWhenEnabled() throws Exception { + public void testThatHttpPipeliningWorks() throws Exception { final Settings settings = Settings.builder() - .put("http.pipelining", true) .put("http.port", "0") .build(); try (HttpServerTransport httpServerTransport = new CustomNettyHttpServerTransport(settings)) { @@ -112,48 +106,6 @@ public class Netty4HttpServerPipeliningTests extends ESTestCase { } } - public void testThatHttpPipeliningCanBeDisabled() throws Exception { - final Settings settings = Settings.builder() - .put("http.pipelining", false) - .put("http.port", "0") - .build(); - try (HttpServerTransport httpServerTransport = new CustomNettyHttpServerTransport(settings)) { - httpServerTransport.start(); - final TransportAddress transportAddress = randomFrom(httpServerTransport.boundAddress().boundAddresses()); - - final int numberOfRequests = randomIntBetween(4, 16); - final Set slowIds = new HashSet<>(); - final List requests = new ArrayList<>(numberOfRequests); - for (int i = 0; i < numberOfRequests; i++) { - if (rarely()) { - requests.add("/slow/" + i); - slowIds.add(i); - } else { - requests.add("/" + i); - } - } - - try (Netty4HttpClient nettyHttpClient = new Netty4HttpClient()) { - Collection responses = nettyHttpClient.get(transportAddress.address(), requests.toArray(new String[]{})); - List responseBodies = new ArrayList<>(Netty4HttpClient.returnHttpResponseBodies(responses)); - // we can not be sure about the order of the responses, but the slow ones should come last - assertThat(responseBodies, hasSize(numberOfRequests)); - for (int i = 0; i < numberOfRequests - slowIds.size(); i++) { - assertThat(responseBodies.get(i), matches("/\\d+")); - } - - final Set ids = new HashSet<>(); - for (int i = 0; i < slowIds.size(); i++) { - final String response = responseBodies.get(numberOfRequests - slowIds.size() + i); - assertThat(response, matches("/slow/\\d+" )); - assertTrue(ids.add(Integer.parseInt(response.split("/")[2]))); - } - - assertThat(slowIds, equalTo(ids)); - } - } - } - class CustomNettyHttpServerTransport extends Netty4HttpServerTransport { private final ExecutorService executorService = Executors.newCachedThreadPool(); @@ -196,7 +148,7 @@ public class Netty4HttpServerPipeliningTests extends ESTestCase { } - class PossiblySlowUpstreamHandler extends SimpleChannelInboundHandler { + class PossiblySlowUpstreamHandler extends SimpleChannelInboundHandler> { private final ExecutorService executorService; @@ -205,7 +157,7 @@ public class Netty4HttpServerPipeliningTests extends ESTestCase { } @Override - protected void channelRead0(ChannelHandlerContext ctx, Object msg) throws Exception { + protected void channelRead0(ChannelHandlerContext ctx, HttpPipelinedRequest msg) throws Exception { executorService.submit(new PossiblySlowRunnable(ctx, msg)); } @@ -220,26 +172,18 @@ public class Netty4HttpServerPipeliningTests extends ESTestCase { class PossiblySlowRunnable implements Runnable { private ChannelHandlerContext ctx; - private HttpPipelinedRequest pipelinedRequest; + private HttpPipelinedRequest pipelinedRequest; private FullHttpRequest fullHttpRequest; - PossiblySlowRunnable(ChannelHandlerContext ctx, Object msg) { + PossiblySlowRunnable(ChannelHandlerContext ctx, HttpPipelinedRequest msg) { this.ctx = ctx; - if (msg instanceof HttpPipelinedRequest) { - this.pipelinedRequest = (HttpPipelinedRequest) msg; - } else if (msg instanceof FullHttpRequest) { - this.fullHttpRequest = (FullHttpRequest) msg; - } + this.pipelinedRequest = msg; + this.fullHttpRequest = pipelinedRequest.getRequest(); } @Override public void run() { - final String uri; - if (pipelinedRequest != null && pipelinedRequest.last() instanceof FullHttpRequest) { - uri = ((FullHttpRequest) pipelinedRequest.last()).uri(); - } else { - uri = fullHttpRequest.uri(); - } + final String uri = fullHttpRequest.uri(); final ByteBuf buffer = Unpooled.copiedBuffer(uri, StandardCharsets.UTF_8); @@ -258,13 +202,7 @@ public class Netty4HttpServerPipeliningTests extends ESTestCase { } final ChannelPromise promise = ctx.newPromise(); - final Object msg; - if (pipelinedRequest != null) { - msg = pipelinedRequest.createHttpResponse(httpResponse, promise); - } else { - msg = httpResponse; - } - ctx.writeAndFlush(msg, promise); + ctx.writeAndFlush(new Netty4HttpResponse(pipelinedRequest.getSequence(), httpResponse), promise); } } diff --git a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningEnabledIT.java b/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningIT.java similarity index 87% rename from modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningEnabledIT.java rename to modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningIT.java index 9723ee93faf..ebb91d9663e 100644 --- a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningEnabledIT.java +++ b/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningIT.java @@ -21,8 +21,6 @@ package org.elasticsearch.http.netty4; import io.netty.handler.codec.http.FullHttpResponse; import org.elasticsearch.ESNetty4IntegTestCase; -import org.elasticsearch.common.network.NetworkModule; -import org.elasticsearch.common.settings.Settings; import org.elasticsearch.common.transport.TransportAddress; import org.elasticsearch.http.HttpServerTransport; import org.elasticsearch.test.ESIntegTestCase.ClusterScope; @@ -35,21 +33,13 @@ import static org.hamcrest.Matchers.hasSize; import static org.hamcrest.Matchers.is; @ClusterScope(scope = Scope.TEST, supportsDedicatedMasters = false, numDataNodes = 1) -public class Netty4PipeliningEnabledIT extends ESNetty4IntegTestCase { +public class Netty4PipeliningIT extends ESNetty4IntegTestCase { @Override protected boolean addMockHttpTransport() { return false; // enable http } - @Override - protected Settings nodeSettings(int nodeOrdinal) { - return Settings.builder() - .put(super.nodeSettings(nodeOrdinal)) - .put("http.pipelining", true) - .build(); - } - public void testThatNettyHttpServerSupportsPipelining() throws Exception { String[] requests = new String[]{"/", "/_nodes/stats", "/", "/_cluster/state", "/"}; diff --git a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/HttpReadWriteHandler.java b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/HttpReadWriteHandler.java index f1d18ddacbd..e3481e3c254 100644 --- a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/HttpReadWriteHandler.java +++ b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/HttpReadWriteHandler.java @@ -25,20 +25,21 @@ import io.netty.handler.codec.ByteToMessageDecoder; import io.netty.handler.codec.http.DefaultFullHttpRequest; import io.netty.handler.codec.http.DefaultHttpHeaders; import io.netty.handler.codec.http.FullHttpRequest; -import io.netty.handler.codec.http.FullHttpResponse; import io.netty.handler.codec.http.HttpContentCompressor; import io.netty.handler.codec.http.HttpContentDecompressor; import io.netty.handler.codec.http.HttpHeaders; import io.netty.handler.codec.http.HttpObjectAggregator; import io.netty.handler.codec.http.HttpRequestDecoder; import io.netty.handler.codec.http.HttpResponseEncoder; +import org.elasticsearch.common.util.BigArrays; import org.elasticsearch.common.util.concurrent.ThreadContext; import org.elasticsearch.common.xcontent.NamedXContentRegistry; import org.elasticsearch.http.HttpHandlingSettings; +import org.elasticsearch.http.HttpPipelinedRequest; import org.elasticsearch.nio.FlushOperation; import org.elasticsearch.nio.InboundChannelBuffer; -import org.elasticsearch.nio.ReadWriteHandler; import org.elasticsearch.nio.NioSocketChannel; +import org.elasticsearch.nio.ReadWriteHandler; import org.elasticsearch.nio.SocketChannelContext; import org.elasticsearch.nio.WriteOperation; import org.elasticsearch.rest.RestRequest; @@ -77,6 +78,7 @@ public class HttpReadWriteHandler implements ReadWriteHandler { if (settings.isCompression()) { handlers.add(new HttpContentCompressor(settings.getCompressionLevel())); } + handlers.add(new NioHttpPipeliningHandler(transport.getLogger(), settings.getPipeliningMaxEvents())); adaptor = new NettyAdaptor(handlers.toArray(new ChannelHandler[0])); adaptor.addCloseListener((v, e) -> nioChannel.close()); @@ -95,9 +97,9 @@ public class HttpReadWriteHandler implements ReadWriteHandler { @Override public WriteOperation createWriteOperation(SocketChannelContext context, Object message, BiConsumer listener) { - assert message instanceof FullHttpResponse : "This channel only supports messages that are of type: " + FullHttpResponse.class - + ". Found type: " + message.getClass() + "."; - return new HttpWriteOperation(context, (FullHttpResponse) message, listener); + assert message instanceof NioHttpResponse : "This channel only supports messages that are of type: " + + NioHttpResponse.class + ". Found type: " + message.getClass() + "."; + return new HttpWriteOperation(context, (NioHttpResponse) message, listener); } @Override @@ -125,76 +127,85 @@ public class HttpReadWriteHandler implements ReadWriteHandler { } } + @SuppressWarnings("unchecked") private void handleRequest(Object msg) { - final FullHttpRequest request = (FullHttpRequest) msg; + final HttpPipelinedRequest pipelinedRequest = (HttpPipelinedRequest) msg; + FullHttpRequest request = pipelinedRequest.getRequest(); - final FullHttpRequest copiedRequest = - new DefaultFullHttpRequest( - request.protocolVersion(), - request.method(), - request.uri(), - Unpooled.copiedBuffer(request.content()), - request.headers(), - request.trailingHeaders()); + try { + final FullHttpRequest copiedRequest = + new DefaultFullHttpRequest( + request.protocolVersion(), + request.method(), + request.uri(), + Unpooled.copiedBuffer(request.content()), + request.headers(), + request.trailingHeaders()); - Exception badRequestCause = null; + Exception badRequestCause = null; - /* - * We want to create a REST request from the incoming request from Netty. However, creating this request could fail if there - * are incorrectly encoded parameters, or the Content-Type header is invalid. If one of these specific failures occurs, we - * attempt to create a REST request again without the input that caused the exception (e.g., we remove the Content-Type header, - * or skip decoding the parameters). Once we have a request in hand, we then dispatch the request as a bad request with the - * underlying exception that caused us to treat the request as bad. - */ - final NioHttpRequest httpRequest; - { - NioHttpRequest innerHttpRequest; - try { - innerHttpRequest = new NioHttpRequest(xContentRegistry, copiedRequest); - } catch (final RestRequest.ContentTypeHeaderException e) { - badRequestCause = e; - innerHttpRequest = requestWithoutContentTypeHeader(copiedRequest, badRequestCause); - } catch (final RestRequest.BadParameterException e) { - badRequestCause = e; - innerHttpRequest = requestWithoutParameters(copiedRequest); - } - httpRequest = innerHttpRequest; - } - - /* - * We now want to create a channel used to send the response on. However, creating this channel can fail if there are invalid - * parameter values for any of the filter_path, human, or pretty parameters. We detect these specific failures via an - * IllegalArgumentException from the channel constructor and then attempt to create a new channel that bypasses parsing of these - * parameter values. - */ - final NioHttpChannel channel; - { - NioHttpChannel innerChannel; - try { - innerChannel = new NioHttpChannel(nioChannel, transport.getBigArrays(), httpRequest, settings, threadContext); - } catch (final IllegalArgumentException e) { - if (badRequestCause == null) { + /* + * We want to create a REST request from the incoming request from Netty. However, creating this request could fail if there + * are incorrectly encoded parameters, or the Content-Type header is invalid. If one of these specific failures occurs, we + * attempt to create a REST request again without the input that caused the exception (e.g., we remove the Content-Type header, + * or skip decoding the parameters). Once we have a request in hand, we then dispatch the request as a bad request with the + * underlying exception that caused us to treat the request as bad. + */ + final NioHttpRequest httpRequest; + { + NioHttpRequest innerHttpRequest; + try { + innerHttpRequest = new NioHttpRequest(xContentRegistry, copiedRequest); + } catch (final RestRequest.ContentTypeHeaderException e) { badRequestCause = e; - } else { - badRequestCause.addSuppressed(e); + innerHttpRequest = requestWithoutContentTypeHeader(copiedRequest, badRequestCause); + } catch (final RestRequest.BadParameterException e) { + badRequestCause = e; + innerHttpRequest = requestWithoutParameters(copiedRequest); } - final NioHttpRequest innerRequest = - new NioHttpRequest( - xContentRegistry, - Collections.emptyMap(), // we are going to dispatch the request as a bad request, drop all parameters - copiedRequest.uri(), - copiedRequest); - innerChannel = new NioHttpChannel(nioChannel, transport.getBigArrays(), innerRequest, settings, threadContext); + httpRequest = innerHttpRequest; } - channel = innerChannel; - } - if (request.decoderResult().isFailure()) { - transport.dispatchBadRequest(httpRequest, channel, request.decoderResult().cause()); - } else if (badRequestCause != null) { - transport.dispatchBadRequest(httpRequest, channel, badRequestCause); - } else { - transport.dispatchRequest(httpRequest, channel); + /* + * We now want to create a channel used to send the response on. However, creating this channel can fail if there are invalid + * parameter values for any of the filter_path, human, or pretty parameters. We detect these specific failures via an + * IllegalArgumentException from the channel constructor and then attempt to create a new channel that bypasses parsing of + * these parameter values. + */ + final NioHttpChannel channel; + { + NioHttpChannel innerChannel; + int sequence = pipelinedRequest.getSequence(); + BigArrays bigArrays = transport.getBigArrays(); + try { + innerChannel = new NioHttpChannel(nioChannel, bigArrays, httpRequest, sequence, settings, threadContext); + } catch (final IllegalArgumentException e) { + if (badRequestCause == null) { + badRequestCause = e; + } else { + badRequestCause.addSuppressed(e); + } + final NioHttpRequest innerRequest = + new NioHttpRequest( + xContentRegistry, + Collections.emptyMap(), // we are going to dispatch the request as a bad request, drop all parameters + copiedRequest.uri(), + copiedRequest); + innerChannel = new NioHttpChannel(nioChannel, bigArrays, innerRequest, sequence, settings, threadContext); + } + channel = innerChannel; + } + + if (request.decoderResult().isFailure()) { + transport.dispatchBadRequest(httpRequest, channel, request.decoderResult().cause()); + } else if (badRequestCause != null) { + transport.dispatchBadRequest(httpRequest, channel, badRequestCause); + } else { + transport.dispatchRequest(httpRequest, channel); + } + } finally { + // As we have copied the buffer, we can release the request + request.release(); } } diff --git a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/HttpWriteOperation.java b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/HttpWriteOperation.java index c838ae85e9d..8ddce7a5b73 100644 --- a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/HttpWriteOperation.java +++ b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/HttpWriteOperation.java @@ -19,7 +19,6 @@ package org.elasticsearch.http.nio; -import io.netty.handler.codec.http.FullHttpResponse; import org.elasticsearch.nio.SocketChannelContext; import org.elasticsearch.nio.WriteOperation; @@ -28,10 +27,10 @@ import java.util.function.BiConsumer; public class HttpWriteOperation implements WriteOperation { private final SocketChannelContext channelContext; - private final FullHttpResponse response; + private final NioHttpResponse response; private final BiConsumer listener; - HttpWriteOperation(SocketChannelContext channelContext, FullHttpResponse response, BiConsumer listener) { + HttpWriteOperation(SocketChannelContext channelContext, NioHttpResponse response, BiConsumer listener) { this.channelContext = channelContext; this.response = response; this.listener = listener; @@ -48,7 +47,7 @@ public class HttpWriteOperation implements WriteOperation { } @Override - public FullHttpResponse getObject() { + public NioHttpResponse getObject() { return response; } } diff --git a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NettyAdaptor.java b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NettyAdaptor.java index 3344a312641..cf8c92bff90 100644 --- a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NettyAdaptor.java +++ b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NettyAdaptor.java @@ -53,12 +53,7 @@ public class NettyAdaptor implements AutoCloseable { try { ByteBuf message = (ByteBuf) msg; promise.addListener((f) -> message.release()); - NettyListener listener; - if (promise instanceof NettyListener) { - listener = (NettyListener) promise; - } else { - listener = new NettyListener(promise); - } + NettyListener listener = NettyListener.fromChannelPromise(promise); flushOperations.add(new FlushOperation(message.nioBuffers(), listener)); } catch (Exception e) { promise.setFailure(e); @@ -107,18 +102,7 @@ public class NettyAdaptor implements AutoCloseable { } public void write(WriteOperation writeOperation) { - ChannelPromise channelPromise = nettyChannel.newPromise(); - channelPromise.addListener(f -> { - BiConsumer consumer = writeOperation.getListener(); - if (f.cause() == null) { - consumer.accept(null, null); - } else { - ExceptionsHelper.dieOnError(f.cause()); - consumer.accept(null, f.cause()); - } - }); - - nettyChannel.writeAndFlush(writeOperation.getObject(), new NettyListener(channelPromise)); + nettyChannel.writeAndFlush(writeOperation.getObject(), NettyListener.fromBiConsumer(writeOperation.getListener(), nettyChannel)); } public FlushOperation pollOutboundOperation() { diff --git a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NettyListener.java b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NettyListener.java index e806b0d23ce..b907c0f2bc6 100644 --- a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NettyListener.java +++ b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NettyListener.java @@ -23,7 +23,7 @@ import io.netty.channel.Channel; import io.netty.channel.ChannelPromise; import io.netty.util.concurrent.Future; import io.netty.util.concurrent.GenericFutureListener; -import org.elasticsearch.action.ActionListener; +import org.elasticsearch.ExceptionsHelper; import org.elasticsearch.common.util.concurrent.FutureUtils; import java.util.concurrent.ExecutionException; @@ -40,7 +40,7 @@ public class NettyListener implements BiConsumer, ChannelPromis private final ChannelPromise promise; - NettyListener(ChannelPromise promise) { + private NettyListener(ChannelPromise promise) { this.promise = promise; } @@ -211,4 +211,30 @@ public class NettyListener implements BiConsumer, ChannelPromis public ChannelPromise unvoid() { return promise.unvoid(); } + + public static NettyListener fromBiConsumer(BiConsumer biConsumer, Channel channel) { + if (biConsumer instanceof NettyListener) { + return (NettyListener) biConsumer; + } else { + ChannelPromise channelPromise = channel.newPromise(); + channelPromise.addListener(f -> { + if (f.cause() == null) { + biConsumer.accept(null, null); + } else { + ExceptionsHelper.dieOnError(f.cause()); + biConsumer.accept(null, f.cause()); + } + }); + + return new NettyListener(channelPromise); + } + } + + public static NettyListener fromChannelPromise(ChannelPromise channelPromise) { + if (channelPromise instanceof NettyListener) { + return (NettyListener) channelPromise; + } else { + return new NettyListener(channelPromise); + } + } } diff --git a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpChannel.java b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpChannel.java index 672c6d5abad..97eba20a16f 100644 --- a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpChannel.java +++ b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpChannel.java @@ -52,20 +52,23 @@ import java.util.EnumMap; import java.util.List; import java.util.Map; import java.util.Set; +import java.util.function.BiConsumer; public class NioHttpChannel extends AbstractRestChannel { private final BigArrays bigArrays; + private final int sequence; private final ThreadContext threadContext; private final FullHttpRequest nettyRequest; private final NioSocketChannel nioChannel; private final boolean resetCookies; - NioHttpChannel(NioSocketChannel nioChannel, BigArrays bigArrays, NioHttpRequest request, + NioHttpChannel(NioSocketChannel nioChannel, BigArrays bigArrays, NioHttpRequest request, int sequence, HttpHandlingSettings settings, ThreadContext threadContext) { super(request, settings.getDetailedErrorsEnabled()); this.nioChannel = nioChannel; this.bigArrays = bigArrays; + this.sequence = sequence; this.threadContext = threadContext; this.nettyRequest = request.getRequest(); this.resetCookies = settings.isResetCookies(); @@ -117,9 +120,8 @@ public class NioHttpChannel extends AbstractRestChannel { toClose.add(nioChannel::close); } - nioChannel.getContext().sendMessage(resp, (aVoid, throwable) -> { - Releasables.close(toClose); - }); + BiConsumer listener = (aVoid, throwable) -> Releasables.close(toClose); + nioChannel.getContext().sendMessage(new NioHttpResponse(sequence, resp), listener); success = true; } finally { if (success == false) { diff --git a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpPipeliningHandler.java b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpPipeliningHandler.java new file mode 100644 index 00000000000..2b702042ba7 --- /dev/null +++ b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpPipeliningHandler.java @@ -0,0 +1,103 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.http.nio; + +import io.netty.channel.ChannelDuplexHandler; +import io.netty.channel.ChannelHandlerContext; +import io.netty.channel.ChannelPromise; +import io.netty.handler.codec.http.LastHttpContent; +import org.apache.logging.log4j.Logger; +import org.elasticsearch.common.collect.Tuple; +import org.elasticsearch.http.HttpPipelinedRequest; +import org.elasticsearch.http.HttpPipeliningAggregator; +import org.elasticsearch.http.nio.NettyListener; +import org.elasticsearch.http.nio.NioHttpResponse; + +import java.nio.channels.ClosedChannelException; +import java.util.List; + +/** + * Implements HTTP pipelining ordering, ensuring that responses are completely served in the same order as their corresponding requests. + */ +public class NioHttpPipeliningHandler extends ChannelDuplexHandler { + + private final Logger logger; + private final HttpPipeliningAggregator aggregator; + + /** + * Construct a new pipelining handler; this handler should be used downstream of HTTP decoding/aggregation. + * + * @param logger for logging unexpected errors + * @param maxEventsHeld the maximum number of channel events that will be retained prior to aborting the channel connection; this is + * required as events cannot queue up indefinitely + */ + public NioHttpPipeliningHandler(Logger logger, final int maxEventsHeld) { + this.logger = logger; + this.aggregator = new HttpPipeliningAggregator<>(maxEventsHeld); + } + + @Override + public void channelRead(final ChannelHandlerContext ctx, final Object msg) { + if (msg instanceof LastHttpContent) { + HttpPipelinedRequest pipelinedRequest = aggregator.read(((LastHttpContent) msg)); + ctx.fireChannelRead(pipelinedRequest); + } else { + ctx.fireChannelRead(msg); + } + } + + @Override + public void write(final ChannelHandlerContext ctx, final Object msg, final ChannelPromise promise) { + assert msg instanceof NioHttpResponse : "Message must be type: " + NioHttpResponse.class; + NioHttpResponse response = (NioHttpResponse) msg; + boolean success = false; + try { + NettyListener listener = NettyListener.fromChannelPromise(promise); + List> readyResponses = aggregator.write(response, listener); + success = true; + for (Tuple responseToWrite : readyResponses) { + ctx.write(responseToWrite.v1().getResponse(), responseToWrite.v2()); + } + } catch (IllegalStateException e) { + ctx.channel().close(); + } finally { + if (success == false) { + promise.setFailure(new ClosedChannelException()); + } + } + } + + @Override + public void close(ChannelHandlerContext ctx, ChannelPromise promise) { + List> inflightResponses = aggregator.removeAllInflightResponses(); + + if (inflightResponses.isEmpty() == false) { + ClosedChannelException closedChannelException = new ClosedChannelException(); + for (Tuple inflightResponse : inflightResponses) { + try { + inflightResponse.v2().setFailure(closedChannelException); + } catch (RuntimeException e) { + logger.error("unexpected error while releasing pipelined http responses", e); + } + } + } + ctx.close(promise); + } +} diff --git a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpResponse.java b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpResponse.java new file mode 100644 index 00000000000..4b634994b45 --- /dev/null +++ b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpResponse.java @@ -0,0 +1,37 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.http.nio; + +import io.netty.handler.codec.http.FullHttpResponse; +import org.elasticsearch.http.HttpPipelinedMessage; + +public class NioHttpResponse extends HttpPipelinedMessage { + + private final FullHttpResponse response; + + public NioHttpResponse(int sequence, FullHttpResponse response) { + super(sequence); + this.response = response; + } + + public FullHttpResponse getResponse() { + return response; + } +} diff --git a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpServerTransport.java b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpServerTransport.java index 06f581d7ab7..825a023bd51 100644 --- a/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpServerTransport.java +++ b/plugins/transport-nio/src/main/java/org/elasticsearch/http/nio/NioHttpServerTransport.java @@ -20,6 +20,7 @@ package org.elasticsearch.http.nio; import io.netty.handler.timeout.ReadTimeoutException; +import org.apache.logging.log4j.Logger; import org.apache.logging.log4j.message.ParameterizedMessage; import org.apache.logging.log4j.util.Supplier; import org.elasticsearch.ElasticsearchException; @@ -84,6 +85,7 @@ import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_TCP_NO_D import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_TCP_RECEIVE_BUFFER_SIZE; import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_TCP_REUSE_ADDRESS; import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_TCP_SEND_BUFFER_SIZE; +import static org.elasticsearch.http.HttpTransportSettings.SETTING_PIPELINING_MAX_EVENTS; public class NioHttpServerTransport extends AbstractHttpServerTransport { @@ -124,6 +126,7 @@ public class NioHttpServerTransport extends AbstractHttpServerTransport { ByteSizeValue maxChunkSize = SETTING_HTTP_MAX_CHUNK_SIZE.get(settings); ByteSizeValue maxHeaderSize = SETTING_HTTP_MAX_HEADER_SIZE.get(settings); ByteSizeValue maxInitialLineLength = SETTING_HTTP_MAX_INITIAL_LINE_LENGTH.get(settings); + int pipeliningMaxEvents = SETTING_PIPELINING_MAX_EVENTS.get(settings); this.httpHandlingSettings = new HttpHandlingSettings(Math.toIntExact(maxContentLength.getBytes()), Math.toIntExact(maxChunkSize.getBytes()), Math.toIntExact(maxHeaderSize.getBytes()), @@ -131,7 +134,8 @@ public class NioHttpServerTransport extends AbstractHttpServerTransport { SETTING_HTTP_RESET_COOKIES.get(settings), SETTING_HTTP_COMPRESSION.get(settings), SETTING_HTTP_COMPRESSION_LEVEL.get(settings), - SETTING_HTTP_DETAILED_ERRORS_ENABLED.get(settings)); + SETTING_HTTP_DETAILED_ERRORS_ENABLED.get(settings), + pipeliningMaxEvents); this.tcpNoDelay = SETTING_HTTP_TCP_NO_DELAY.get(settings); this.tcpKeepAlive = SETTING_HTTP_TCP_KEEP_ALIVE.get(settings); @@ -140,14 +144,19 @@ public class NioHttpServerTransport extends AbstractHttpServerTransport { this.tcpReceiveBufferSize = Math.toIntExact(SETTING_HTTP_TCP_RECEIVE_BUFFER_SIZE.get(settings).getBytes()); - logger.debug("using max_chunk_size[{}], max_header_size[{}], max_initial_line_length[{}], max_content_length[{}]", - maxChunkSize, maxHeaderSize, maxInitialLineLength, maxContentLength); + logger.debug("using max_chunk_size[{}], max_header_size[{}], max_initial_line_length[{}], max_content_length[{}]," + + " pipelining_max_events[{}]", + maxChunkSize, maxHeaderSize, maxInitialLineLength, maxContentLength, pipeliningMaxEvents); } BigArrays getBigArrays() { return bigArrays; } + public Logger getLogger() { + return logger; + } + @Override protected void doStart() { boolean success = false; diff --git a/plugins/transport-nio/src/test/java/org/elasticsearch/NioIntegTestCase.java b/plugins/transport-nio/src/test/java/org/elasticsearch/NioIntegTestCase.java index e0c8bacca1d..703f7acbf82 100644 --- a/plugins/transport-nio/src/test/java/org/elasticsearch/NioIntegTestCase.java +++ b/plugins/transport-nio/src/test/java/org/elasticsearch/NioIntegTestCase.java @@ -20,6 +20,7 @@ package org.elasticsearch; import org.elasticsearch.common.network.NetworkModule; import org.elasticsearch.common.settings.Settings; +import org.elasticsearch.http.nio.NioHttpServerTransport; import org.elasticsearch.plugins.Plugin; import org.elasticsearch.test.ESIntegTestCase; import org.elasticsearch.transport.nio.NioTransport; @@ -43,11 +44,13 @@ public abstract class NioIntegTestCase extends ESIntegTestCase { @Override protected Settings nodeSettings(int nodeOrdinal) { Settings.Builder builder = Settings.builder().put(super.nodeSettings(nodeOrdinal)); - // randomize netty settings + // randomize nio settings if (randomBoolean()) { builder.put(NioTransport.NIO_WORKER_COUNT.getKey(), random().nextInt(3) + 1); + builder.put(NioHttpServerTransport.NIO_HTTP_WORKER_COUNT.getKey(), random().nextInt(3) + 1); } builder.put(NetworkModule.TRANSPORT_TYPE_KEY, NioTransportPlugin.NIO_TRANSPORT_NAME); + builder.put(NetworkModule.HTTP_TYPE_KEY, NioTransportPlugin.NIO_HTTP_TRANSPORT_NAME); return builder.build(); } diff --git a/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/HttpReadWriteHandlerTests.java b/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/HttpReadWriteHandlerTests.java index dce8319d2fc..cc8eeb77cc2 100644 --- a/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/HttpReadWriteHandlerTests.java +++ b/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/HttpReadWriteHandlerTests.java @@ -61,11 +61,11 @@ import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_MAX_CHUN import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_MAX_HEADER_SIZE; import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_MAX_INITIAL_LINE_LENGTH; import static org.elasticsearch.http.HttpTransportSettings.SETTING_HTTP_RESET_COOKIES; +import static org.elasticsearch.http.HttpTransportSettings.SETTING_PIPELINING_MAX_EVENTS; import static org.mockito.Matchers.any; import static org.mockito.Mockito.mock; import static org.mockito.Mockito.times; import static org.mockito.Mockito.verify; -import static org.mockito.Mockito.verifyZeroInteractions; public class HttpReadWriteHandlerTests extends ESTestCase { @@ -91,7 +91,8 @@ public class HttpReadWriteHandlerTests extends ESTestCase { SETTING_HTTP_RESET_COOKIES.getDefault(settings), SETTING_HTTP_COMPRESSION.getDefault(settings), SETTING_HTTP_COMPRESSION_LEVEL.getDefault(settings), - SETTING_HTTP_DETAILED_ERRORS_ENABLED.getDefault(settings)); + SETTING_HTTP_DETAILED_ERRORS_ENABLED.getDefault(settings), + SETTING_PIPELINING_MAX_EVENTS.getDefault(settings)); ThreadContext threadContext = new ThreadContext(settings); nioSocketChannel = mock(NioSocketChannel.class); handler = new HttpReadWriteHandler(nioSocketChannel, transport, httpHandlingSettings, NamedXContentRegistry.EMPTY, threadContext); @@ -148,7 +149,8 @@ public class HttpReadWriteHandlerTests extends ESTestCase { handler.consumeReads(toChannelBuffer(buf)); - verifyZeroInteractions(transport); + verify(transport, times(0)).dispatchBadRequest(any(), any(), any()); + verify(transport, times(0)).dispatchRequest(any(), any()); List flushOperations = handler.pollFlushOperations(); assertFalse(flushOperations.isEmpty()); @@ -169,9 +171,10 @@ public class HttpReadWriteHandlerTests extends ESTestCase { prepareHandlerForResponse(handler); FullHttpResponse fullHttpResponse = new DefaultFullHttpResponse(HttpVersion.HTTP_1_1, HttpResponseStatus.OK); + NioHttpResponse pipelinedResponse = new NioHttpResponse(0, fullHttpResponse); SocketChannelContext context = mock(SocketChannelContext.class); - HttpWriteOperation writeOperation = new HttpWriteOperation(context, fullHttpResponse, mock(BiConsumer.class)); + HttpWriteOperation writeOperation = new HttpWriteOperation(context, pipelinedResponse, mock(BiConsumer.class)); List flushOperations = handler.writeToBytes(writeOperation); HttpResponse response = responseDecoder.decode(Unpooled.wrappedBuffer(flushOperations.get(0).getBuffersToWrite())); diff --git a/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/NioHttpPipeliningHandlerTests.java b/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/NioHttpPipeliningHandlerTests.java new file mode 100644 index 00000000000..d12c608aeca --- /dev/null +++ b/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/NioHttpPipeliningHandlerTests.java @@ -0,0 +1,304 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ + +package org.elasticsearch.http.nio; + +import io.netty.buffer.ByteBuf; +import io.netty.buffer.ByteBufUtil; +import io.netty.buffer.Unpooled; +import io.netty.channel.ChannelHandlerContext; +import io.netty.channel.ChannelPromise; +import io.netty.channel.SimpleChannelInboundHandler; +import io.netty.channel.embedded.EmbeddedChannel; +import io.netty.handler.codec.http.DefaultFullHttpRequest; +import io.netty.handler.codec.http.DefaultFullHttpResponse; +import io.netty.handler.codec.http.DefaultHttpRequest; +import io.netty.handler.codec.http.FullHttpRequest; +import io.netty.handler.codec.http.FullHttpResponse; +import io.netty.handler.codec.http.HttpMethod; +import io.netty.handler.codec.http.HttpRequest; +import io.netty.handler.codec.http.HttpVersion; +import io.netty.handler.codec.http.LastHttpContent; +import io.netty.handler.codec.http.QueryStringDecoder; +import org.elasticsearch.common.Randomness; +import org.elasticsearch.http.HttpPipelinedRequest; +import org.elasticsearch.test.ESTestCase; +import org.junit.After; + +import java.nio.channels.ClosedChannelException; +import java.nio.charset.StandardCharsets; +import java.util.ArrayList; +import java.util.List; +import java.util.Map; +import java.util.Queue; +import java.util.concurrent.ConcurrentHashMap; +import java.util.concurrent.CountDownLatch; +import java.util.concurrent.ExecutorService; +import java.util.concurrent.Executors; +import java.util.concurrent.LinkedTransferQueue; +import java.util.concurrent.TimeUnit; +import java.util.stream.Collectors; +import java.util.stream.IntStream; + +import static io.netty.handler.codec.http.HttpHeaderNames.CONTENT_LENGTH; +import static io.netty.handler.codec.http.HttpResponseStatus.OK; +import static io.netty.handler.codec.http.HttpVersion.HTTP_1_1; +import static org.hamcrest.core.Is.is; + +public class NioHttpPipeliningHandlerTests extends ESTestCase { + + private final ExecutorService handlerService = Executors.newFixedThreadPool(randomIntBetween(4, 8)); + private final ExecutorService eventLoopService = Executors.newFixedThreadPool(1); + private final Map waitingRequests = new ConcurrentHashMap<>(); + private final Map finishingRequests = new ConcurrentHashMap<>(); + + @After + public void cleanup() throws Exception { + waitingRequests.keySet().forEach(this::finishRequest); + shutdownExecutorService(); + } + + private CountDownLatch finishRequest(String url) { + waitingRequests.get(url).countDown(); + return finishingRequests.get(url); + } + + private void shutdownExecutorService() throws InterruptedException { + if (!handlerService.isShutdown()) { + handlerService.shutdown(); + handlerService.awaitTermination(10, TimeUnit.SECONDS); + } + if (!eventLoopService.isShutdown()) { + eventLoopService.shutdown(); + eventLoopService.awaitTermination(10, TimeUnit.SECONDS); + } + } + + public void testThatPipeliningWorksWithFastSerializedRequests() throws InterruptedException { + final int numberOfRequests = randomIntBetween(2, 128); + final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new NioHttpPipeliningHandler(logger, numberOfRequests), + new WorkEmulatorHandler()); + + for (int i = 0; i < numberOfRequests; i++) { + embeddedChannel.writeInbound(createHttpRequest("/" + String.valueOf(i))); + } + + final List latches = new ArrayList<>(); + for (final String url : waitingRequests.keySet()) { + latches.add(finishRequest(url)); + } + + for (final CountDownLatch latch : latches) { + latch.await(); + } + + embeddedChannel.flush(); + + for (int i = 0; i < numberOfRequests; i++) { + assertReadHttpMessageHasContent(embeddedChannel, String.valueOf(i)); + } + + assertTrue(embeddedChannel.isOpen()); + } + + public void testThatPipeliningWorksWhenSlowRequestsInDifferentOrder() throws InterruptedException { + final int numberOfRequests = randomIntBetween(2, 128); + final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new NioHttpPipeliningHandler(logger, numberOfRequests), + new WorkEmulatorHandler()); + + for (int i = 0; i < numberOfRequests; i++) { + embeddedChannel.writeInbound(createHttpRequest("/" + String.valueOf(i))); + } + + // random order execution + final List urls = new ArrayList<>(waitingRequests.keySet()); + Randomness.shuffle(urls); + final List latches = new ArrayList<>(); + for (final String url : urls) { + latches.add(finishRequest(url)); + } + + for (final CountDownLatch latch : latches) { + latch.await(); + } + + embeddedChannel.flush(); + + for (int i = 0; i < numberOfRequests; i++) { + assertReadHttpMessageHasContent(embeddedChannel, String.valueOf(i)); + } + + assertTrue(embeddedChannel.isOpen()); + } + + public void testThatPipeliningWorksWithChunkedRequests() throws InterruptedException { + final int numberOfRequests = randomIntBetween(2, 128); + final EmbeddedChannel embeddedChannel = + new EmbeddedChannel( + new AggregateUrisAndHeadersHandler(), + new NioHttpPipeliningHandler(logger, numberOfRequests), + new WorkEmulatorHandler()); + + for (int i = 0; i < numberOfRequests; i++) { + final DefaultHttpRequest request = new DefaultFullHttpRequest(HttpVersion.HTTP_1_1, HttpMethod.GET, "/" + i); + embeddedChannel.writeInbound(request); + embeddedChannel.writeInbound(LastHttpContent.EMPTY_LAST_CONTENT); + } + + final List latches = new ArrayList<>(); + for (int i = numberOfRequests - 1; i >= 0; i--) { + latches.add(finishRequest(Integer.toString(i))); + } + + for (final CountDownLatch latch : latches) { + latch.await(); + } + + embeddedChannel.flush(); + + for (int i = 0; i < numberOfRequests; i++) { + assertReadHttpMessageHasContent(embeddedChannel, Integer.toString(i)); + } + + assertTrue(embeddedChannel.isOpen()); + } + + public void testThatPipeliningClosesConnectionWithTooManyEvents() throws InterruptedException { + final int numberOfRequests = randomIntBetween(2, 128); + final EmbeddedChannel embeddedChannel = new EmbeddedChannel(new NioHttpPipeliningHandler(logger, numberOfRequests), + new WorkEmulatorHandler()); + + for (int i = 0; i < 1 + numberOfRequests + 1; i++) { + embeddedChannel.writeInbound(createHttpRequest("/" + Integer.toString(i))); + } + + final List latches = new ArrayList<>(); + final List requests = IntStream.range(1, numberOfRequests + 1).boxed().collect(Collectors.toList()); + Randomness.shuffle(requests); + + for (final Integer request : requests) { + latches.add(finishRequest(request.toString())); + } + + for (final CountDownLatch latch : latches) { + latch.await(); + } + + finishRequest(Integer.toString(numberOfRequests + 1)).await(); + + embeddedChannel.flush(); + + assertFalse(embeddedChannel.isOpen()); + } + + public void testPipeliningRequestsAreReleased() throws InterruptedException { + final int numberOfRequests = 10; + final EmbeddedChannel embeddedChannel = + new EmbeddedChannel(new NioHttpPipeliningHandler(logger, numberOfRequests + 1)); + + for (int i = 0; i < numberOfRequests; i++) { + embeddedChannel.writeInbound(createHttpRequest("/" + i)); + } + + HttpPipelinedRequest inbound; + ArrayList> requests = new ArrayList<>(); + while ((inbound = embeddedChannel.readInbound()) != null) { + requests.add(inbound); + } + + ArrayList promises = new ArrayList<>(); + for (int i = 1; i < requests.size(); ++i) { + final FullHttpResponse httpResponse = new DefaultFullHttpResponse(HTTP_1_1, OK); + ChannelPromise promise = embeddedChannel.newPromise(); + promises.add(promise); + int sequence = requests.get(i).getSequence(); + NioHttpResponse resp = new NioHttpResponse(sequence, httpResponse); + embeddedChannel.writeAndFlush(resp, promise); + } + + for (ChannelPromise promise : promises) { + assertFalse(promise.isDone()); + } + embeddedChannel.close().syncUninterruptibly(); + for (ChannelPromise promise : promises) { + assertTrue(promise.isDone()); + assertTrue(promise.cause() instanceof ClosedChannelException); + } + } + + private void assertReadHttpMessageHasContent(EmbeddedChannel embeddedChannel, String expectedContent) { + FullHttpResponse response = (FullHttpResponse) embeddedChannel.outboundMessages().poll(); + assertNotNull("Expected response to exist, maybe you did not wait long enough?", response); + assertNotNull("Expected response to have content " + expectedContent, response.content()); + String data = new String(ByteBufUtil.getBytes(response.content()), StandardCharsets.UTF_8); + assertThat(data, is(expectedContent)); + } + + private FullHttpRequest createHttpRequest(String uri) { + return new DefaultFullHttpRequest(HTTP_1_1, HttpMethod.GET, uri); + } + + private static class AggregateUrisAndHeadersHandler extends SimpleChannelInboundHandler { + + static final Queue QUEUE_URI = new LinkedTransferQueue<>(); + + @Override + protected void channelRead0(ChannelHandlerContext ctx, HttpRequest request) throws Exception { + QUEUE_URI.add(request.uri()); + } + + } + + private class WorkEmulatorHandler extends SimpleChannelInboundHandler> { + + @Override + protected void channelRead0(final ChannelHandlerContext ctx, HttpPipelinedRequest pipelinedRequest) { + LastHttpContent request = pipelinedRequest.getRequest(); + final QueryStringDecoder decoder; + if (request instanceof FullHttpRequest) { + decoder = new QueryStringDecoder(((FullHttpRequest)request).uri()); + } else { + decoder = new QueryStringDecoder(AggregateUrisAndHeadersHandler.QUEUE_URI.poll()); + } + + final String uri = decoder.path().replace("/", ""); + final ByteBuf content = Unpooled.copiedBuffer(uri, StandardCharsets.UTF_8); + final DefaultFullHttpResponse httpResponse = new DefaultFullHttpResponse(HTTP_1_1, OK, content); + httpResponse.headers().add(CONTENT_LENGTH, content.readableBytes()); + + final CountDownLatch waitingLatch = new CountDownLatch(1); + waitingRequests.put(uri, waitingLatch); + final CountDownLatch finishingLatch = new CountDownLatch(1); + finishingRequests.put(uri, finishingLatch); + + handlerService.submit(() -> { + try { + waitingLatch.await(1000, TimeUnit.SECONDS); + final ChannelPromise promise = ctx.newPromise(); + eventLoopService.submit(() -> { + ctx.write(new NioHttpResponse(pipelinedRequest.getSequence(), httpResponse), promise); + finishingLatch.countDown(); + }); + } catch (InterruptedException e) { + fail(e.toString()); + } + }); + } + } +} diff --git a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningDisabledIT.java b/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/NioPipeliningIT.java similarity index 53% rename from modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningDisabledIT.java rename to plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/NioPipeliningIT.java index af0e7c85a8f..074aafd6eab 100644 --- a/modules/transport-netty4/src/test/java/org/elasticsearch/http/netty4/Netty4PipeliningDisabledIT.java +++ b/plugins/transport-nio/src/test/java/org/elasticsearch/http/nio/NioPipeliningIT.java @@ -16,65 +16,53 @@ * specific language governing permissions and limitations * under the License. */ -package org.elasticsearch.http.netty4; + +package org.elasticsearch.http.nio; import io.netty.handler.codec.http.FullHttpResponse; -import org.elasticsearch.ESNetty4IntegTestCase; -import org.elasticsearch.common.network.NetworkModule; -import org.elasticsearch.common.settings.Settings; +import org.elasticsearch.NioIntegTestCase; import org.elasticsearch.common.transport.TransportAddress; import org.elasticsearch.http.HttpServerTransport; import org.elasticsearch.test.ESIntegTestCase.ClusterScope; import org.elasticsearch.test.ESIntegTestCase.Scope; -import java.util.ArrayList; import java.util.Collection; -import java.util.List; import java.util.Locale; -import static org.hamcrest.Matchers.containsInAnyOrder; import static org.hamcrest.Matchers.hasSize; +import static org.hamcrest.Matchers.is; @ClusterScope(scope = Scope.TEST, supportsDedicatedMasters = false, numDataNodes = 1) -public class Netty4PipeliningDisabledIT extends ESNetty4IntegTestCase { +public class NioPipeliningIT extends NioIntegTestCase { @Override protected boolean addMockHttpTransport() { return false; // enable http } - @Override - protected Settings nodeSettings(int nodeOrdinal) { - return Settings.builder() - .put(super.nodeSettings(nodeOrdinal)) - .put("http.pipelining", false) - .build(); - } - - public void testThatNettyHttpServerDoesNotSupportPipelining() throws Exception { - ensureGreen(); - String[] requests = new String[] {"/", "/_nodes/stats", "/", "/_cluster/state", "/", "/_nodes", "/"}; + public void testThatNioHttpServerSupportsPipelining() throws Exception { + String[] requests = new String[]{"/", "/_nodes/stats", "/", "/_cluster/state", "/"}; HttpServerTransport httpServerTransport = internalCluster().getInstance(HttpServerTransport.class); TransportAddress[] boundAddresses = httpServerTransport.boundAddress().boundAddresses(); - TransportAddress transportAddress = (TransportAddress) randomFrom(boundAddresses); + TransportAddress transportAddress = randomFrom(boundAddresses); try (Netty4HttpClient nettyHttpClient = new Netty4HttpClient()) { Collection responses = nettyHttpClient.get(transportAddress.address(), requests); - assertThat(responses, hasSize(requests.length)); + assertThat(responses, hasSize(5)); - List opaqueIds = new ArrayList<>(Netty4HttpClient.returnOpaqueIds(responses)); - - assertResponsesOutOfOrder(opaqueIds); + Collection opaqueIds = Netty4HttpClient.returnOpaqueIds(responses); + assertOpaqueIdsInOrder(opaqueIds); } } - /** - * checks if all responses are there, but also tests that they are out of order because pipelining is disabled - */ - private void assertResponsesOutOfOrder(List opaqueIds) { - String message = String.format(Locale.ROOT, "Expected returned http message ids to be in any order of: %s", opaqueIds); - assertThat(message, opaqueIds, containsInAnyOrder("0", "1", "2", "3", "4", "5", "6")); + private void assertOpaqueIdsInOrder(Collection opaqueIds) { + // check if opaque ids are monotonically increasing + int i = 0; + String msg = String.format(Locale.ROOT, "Expected list of opaque ids to be monotonically increasing, got [%s]", opaqueIds); + for (String opaqueId : opaqueIds) { + assertThat(msg, opaqueId, is(String.valueOf(i++))); + } } } diff --git a/server/src/main/java/org/elasticsearch/common/settings/ClusterSettings.java b/server/src/main/java/org/elasticsearch/common/settings/ClusterSettings.java index c19cbe4687c..d9cf0f630c0 100644 --- a/server/src/main/java/org/elasticsearch/common/settings/ClusterSettings.java +++ b/server/src/main/java/org/elasticsearch/common/settings/ClusterSettings.java @@ -227,7 +227,6 @@ public final class ClusterSettings extends AbstractScopedSettings { HttpTransportSettings.SETTING_CORS_ENABLED, HttpTransportSettings.SETTING_CORS_MAX_AGE, HttpTransportSettings.SETTING_HTTP_DETAILED_ERRORS_ENABLED, - HttpTransportSettings.SETTING_PIPELINING, HttpTransportSettings.SETTING_CORS_ALLOW_ORIGIN, HttpTransportSettings.SETTING_HTTP_HOST, HttpTransportSettings.SETTING_HTTP_PUBLISH_HOST, diff --git a/server/src/main/java/org/elasticsearch/http/HttpHandlingSettings.java b/server/src/main/java/org/elasticsearch/http/HttpHandlingSettings.java index f86049292f3..df038e8303e 100644 --- a/server/src/main/java/org/elasticsearch/http/HttpHandlingSettings.java +++ b/server/src/main/java/org/elasticsearch/http/HttpHandlingSettings.java @@ -29,9 +29,11 @@ public class HttpHandlingSettings { private final boolean compression; private final int compressionLevel; private final boolean detailedErrorsEnabled; + private final int pipeliningMaxEvents; public HttpHandlingSettings(int maxContentLength, int maxChunkSize, int maxHeaderSize, int maxInitialLineLength, - boolean resetCookies, boolean compression, int compressionLevel, boolean detailedErrorsEnabled) { + boolean resetCookies, boolean compression, int compressionLevel, boolean detailedErrorsEnabled, + int pipeliningMaxEvents) { this.maxContentLength = maxContentLength; this.maxChunkSize = maxChunkSize; this.maxHeaderSize = maxHeaderSize; @@ -40,6 +42,7 @@ public class HttpHandlingSettings { this.compression = compression; this.compressionLevel = compressionLevel; this.detailedErrorsEnabled = detailedErrorsEnabled; + this.pipeliningMaxEvents = pipeliningMaxEvents; } public int getMaxContentLength() { @@ -73,4 +76,8 @@ public class HttpHandlingSettings { public boolean getDetailedErrorsEnabled() { return detailedErrorsEnabled; } + + public int getPipeliningMaxEvents() { + return pipeliningMaxEvents; + } } diff --git a/server/src/main/java/org/elasticsearch/http/HttpPipelinedMessage.java b/server/src/main/java/org/elasticsearch/http/HttpPipelinedMessage.java new file mode 100644 index 00000000000..7db8666e73a --- /dev/null +++ b/server/src/main/java/org/elasticsearch/http/HttpPipelinedMessage.java @@ -0,0 +1,37 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ +package org.elasticsearch.http; + +public class HttpPipelinedMessage implements Comparable { + + private final int sequence; + + public HttpPipelinedMessage(int sequence) { + this.sequence = sequence; + } + + public int getSequence() { + return sequence; + } + + @Override + public int compareTo(HttpPipelinedMessage o) { + return Integer.compare(sequence, o.sequence); + } +} diff --git a/server/src/main/java/org/elasticsearch/http/HttpPipelinedRequest.java b/server/src/main/java/org/elasticsearch/http/HttpPipelinedRequest.java new file mode 100644 index 00000000000..df8bd7ee1eb --- /dev/null +++ b/server/src/main/java/org/elasticsearch/http/HttpPipelinedRequest.java @@ -0,0 +1,33 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ +package org.elasticsearch.http; + +public class HttpPipelinedRequest extends HttpPipelinedMessage { + + private final R request; + + HttpPipelinedRequest(int sequence, R request) { + super(sequence); + this.request = request; + } + + public R getRequest() { + return request; + } +} diff --git a/server/src/main/java/org/elasticsearch/http/HttpPipeliningAggregator.java b/server/src/main/java/org/elasticsearch/http/HttpPipeliningAggregator.java new file mode 100644 index 00000000000..f38e9677979 --- /dev/null +++ b/server/src/main/java/org/elasticsearch/http/HttpPipeliningAggregator.java @@ -0,0 +1,81 @@ +/* + * Licensed to Elasticsearch under one or more contributor + * license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright + * ownership. Elasticsearch 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 + * + * 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. + */ +package org.elasticsearch.http; + +import org.elasticsearch.common.collect.Tuple; + +import java.util.ArrayList; +import java.util.Comparator; +import java.util.List; +import java.util.PriorityQueue; + +public class HttpPipeliningAggregator { + + private final int maxEventsHeld; + private final PriorityQueue> outboundHoldingQueue; + /* + * The current read and write sequence numbers. Read sequence numbers are attached to requests in the order they are read from the + * channel, and then transferred to responses. A response is not written to the channel context until its sequence number matches the + * current write sequence, implying that all preceding messages have been written. + */ + private int readSequence; + private int writeSequence; + + public HttpPipeliningAggregator(int maxEventsHeld) { + this.maxEventsHeld = maxEventsHeld; + this.outboundHoldingQueue = new PriorityQueue<>(1, Comparator.comparing(Tuple::v1)); + } + + public HttpPipelinedRequest read(final Request request) { + return new HttpPipelinedRequest<>(readSequence++, request); + } + + public List> write(final Response response, Listener listener) { + if (outboundHoldingQueue.size() < maxEventsHeld) { + ArrayList> readyResponses = new ArrayList<>(); + outboundHoldingQueue.add(new Tuple<>(response, listener)); + while (!outboundHoldingQueue.isEmpty()) { + /* + * Since the response with the lowest sequence number is the top of the priority queue, we know if its sequence + * number does not match the current write sequence number then we have not processed all preceding responses yet. + */ + final Tuple top = outboundHoldingQueue.peek(); + + if (top.v1().getSequence() != writeSequence) { + break; + } + outboundHoldingQueue.poll(); + readyResponses.add(top); + writeSequence++; + } + + return readyResponses; + } else { + int eventCount = outboundHoldingQueue.size() + 1; + throw new IllegalStateException("Too many pipelined events [" + eventCount + "]. Max events allowed [" + + maxEventsHeld + "]."); + } + } + + public List> removeAllInflightResponses() { + ArrayList> responses = new ArrayList<>(outboundHoldingQueue); + outboundHoldingQueue.clear(); + return responses; + } +} diff --git a/server/src/main/java/org/elasticsearch/http/HttpTransportSettings.java b/server/src/main/java/org/elasticsearch/http/HttpTransportSettings.java index 98451e0c304..4670137d09a 100644 --- a/server/src/main/java/org/elasticsearch/http/HttpTransportSettings.java +++ b/server/src/main/java/org/elasticsearch/http/HttpTransportSettings.java @@ -49,8 +49,6 @@ public final class HttpTransportSettings { new Setting<>("http.cors.allow-headers", "X-Requested-With,Content-Type,Content-Length", (value) -> value, Property.NodeScope); public static final Setting SETTING_CORS_ALLOW_CREDENTIALS = Setting.boolSetting("http.cors.allow-credentials", false, Property.NodeScope); - public static final Setting SETTING_PIPELINING = - Setting.boolSetting("http.pipelining", true, Property.NodeScope); public static final Setting SETTING_PIPELINING_MAX_EVENTS = Setting.intSetting("http.pipelining.max_events", 10000, Property.NodeScope); public static final Setting SETTING_HTTP_COMPRESSION = diff --git a/server/src/test/java/org/elasticsearch/discovery/single/SingleNodeDiscoveryIT.java b/server/src/test/java/org/elasticsearch/discovery/single/SingleNodeDiscoveryIT.java index fc284b9f5e8..fdc36152cc8 100644 --- a/server/src/test/java/org/elasticsearch/discovery/single/SingleNodeDiscoveryIT.java +++ b/server/src/test/java/org/elasticsearch/discovery/single/SingleNodeDiscoveryIT.java @@ -150,7 +150,6 @@ public class SingleNodeDiscoveryIT extends ESIntegTestCase { internalCluster().getClusterName(), configurationSource, 0, - false, "other", Arrays.asList(getTestTransportPlugin(), MockHttpTransport.TestPlugin.class), Function.identity())) { diff --git a/test/framework/src/main/java/org/elasticsearch/test/ESIntegTestCase.java b/test/framework/src/main/java/org/elasticsearch/test/ESIntegTestCase.java index 7210fadd7ea..505a5937d29 100644 --- a/test/framework/src/main/java/org/elasticsearch/test/ESIntegTestCase.java +++ b/test/framework/src/main/java/org/elasticsearch/test/ESIntegTestCase.java @@ -1829,7 +1829,7 @@ public abstract class ESIntegTestCase extends ESTestCase { return new InternalTestCluster(seed, createTempDir(), supportsDedicatedMasters, getAutoMinMasterNodes(), minNumDataNodes, maxNumDataNodes, InternalTestCluster.clusterName(scope.name(), seed) + "-cluster", nodeConfigurationSource, getNumClientNodes(), - InternalTestCluster.DEFAULT_ENABLE_HTTP_PIPELINING, nodePrefix, mockPlugins, getClientWrapper()); + nodePrefix, mockPlugins, getClientWrapper()); } protected NodeConfigurationSource getNodeConfigSource() { diff --git a/test/framework/src/main/java/org/elasticsearch/test/InternalTestCluster.java b/test/framework/src/main/java/org/elasticsearch/test/InternalTestCluster.java index de4b5a5ae05..efe775f7415 100644 --- a/test/framework/src/main/java/org/elasticsearch/test/InternalTestCluster.java +++ b/test/framework/src/main/java/org/elasticsearch/test/InternalTestCluster.java @@ -171,8 +171,6 @@ public final class InternalTestCluster extends TestCluster { static final int DEFAULT_MIN_NUM_CLIENT_NODES = 0; static final int DEFAULT_MAX_NUM_CLIENT_NODES = 1; - static final boolean DEFAULT_ENABLE_HTTP_PIPELINING = true; - /* sorted map to make traverse order reproducible, concurrent since we do checks on it not within a sync block */ private final NavigableMap nodes = new TreeMap<>(); @@ -219,7 +217,7 @@ public final class InternalTestCluster extends TestCluster { public InternalTestCluster(long clusterSeed, Path baseDir, boolean randomlyAddDedicatedMasters, boolean autoManageMinMasterNodes, int minNumDataNodes, int maxNumDataNodes, String clusterName, NodeConfigurationSource nodeConfigurationSource, int numClientNodes, - boolean enableHttpPipelining, String nodePrefix, Collection> mockPlugins, Function clientWrapper) { + String nodePrefix, Collection> mockPlugins, Function clientWrapper) { super(clusterSeed); this.autoManageMinMasterNodes = autoManageMinMasterNodes; this.clientWrapper = clientWrapper; @@ -300,7 +298,6 @@ public final class InternalTestCluster extends TestCluster { builder.put(Environment.PATH_REPO_SETTING.getKey(), baseDir.resolve("repos")); builder.put(TcpTransport.PORT.getKey(), 0); builder.put("http.port", 0); - builder.put("http.pipelining", enableHttpPipelining); if (Strings.hasLength(System.getProperty("tests.es.logger.level"))) { builder.put("logger.level", System.getProperty("tests.es.logger.level")); } diff --git a/test/framework/src/test/java/org/elasticsearch/test/test/InternalTestClusterTests.java b/test/framework/src/test/java/org/elasticsearch/test/test/InternalTestClusterTests.java index ca674e61651..23f44c560ba 100644 --- a/test/framework/src/test/java/org/elasticsearch/test/test/InternalTestClusterTests.java +++ b/test/framework/src/test/java/org/elasticsearch/test/test/InternalTestClusterTests.java @@ -19,8 +19,6 @@ */ package org.elasticsearch.test.test; -import org.elasticsearch.common.settings.Setting; -import org.elasticsearch.core.internal.io.IOUtils; import org.apache.lucene.util.LuceneTestCase; import org.elasticsearch.client.Client; import org.elasticsearch.cluster.ClusterName; @@ -28,6 +26,7 @@ import org.elasticsearch.cluster.node.DiscoveryNode; import org.elasticsearch.cluster.service.ClusterService; import org.elasticsearch.common.network.NetworkModule; import org.elasticsearch.common.settings.Settings; +import org.elasticsearch.core.internal.io.IOUtils; import org.elasticsearch.discovery.DiscoverySettings; import org.elasticsearch.discovery.zen.ZenDiscovery; import org.elasticsearch.env.NodeEnvironment; @@ -63,8 +62,6 @@ import static org.elasticsearch.discovery.zen.ElectMasterService.DISCOVERY_ZEN_M import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertFileExists; import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertFileNotExists; import static org.hamcrest.Matchers.equalTo; -import static org.hamcrest.Matchers.greaterThan; -import static org.hamcrest.Matchers.hasEntry; import static org.hamcrest.Matchers.not; /** @@ -86,16 +83,15 @@ public class InternalTestClusterTests extends ESTestCase { String clusterName = randomRealisticUnicodeOfCodepointLengthBetween(1, 10); NodeConfigurationSource nodeConfigurationSource = NodeConfigurationSource.EMPTY; int numClientNodes = randomIntBetween(0, 10); - boolean enableHttpPipelining = randomBoolean(); String nodePrefix = randomRealisticUnicodeOfCodepointLengthBetween(1, 10); Path baseDir = createTempDir(); InternalTestCluster cluster0 = new InternalTestCluster(clusterSeed, baseDir, masterNodes, randomBoolean(), minNumDataNodes, maxNumDataNodes, clusterName, nodeConfigurationSource, numClientNodes, - enableHttpPipelining, nodePrefix, Collections.emptyList(), Function.identity()); + nodePrefix, Collections.emptyList(), Function.identity()); InternalTestCluster cluster1 = new InternalTestCluster(clusterSeed, baseDir, masterNodes, randomBoolean(), minNumDataNodes, maxNumDataNodes, clusterName, nodeConfigurationSource, numClientNodes, - enableHttpPipelining, nodePrefix, Collections.emptyList(), Function.identity()); + nodePrefix, Collections.emptyList(), Function.identity()); // TODO: this is not ideal - we should have a way to make sure ports are initialized in the same way assertClusters(cluster0, cluster1, false); @@ -211,16 +207,15 @@ public class InternalTestClusterTests extends ESTestCase { } }; - boolean enableHttpPipelining = randomBoolean(); String nodePrefix = "foobar"; Path baseDir = createTempDir(); InternalTestCluster cluster0 = new InternalTestCluster(clusterSeed, baseDir, masterNodes, autoManageMinMasterNodes, minNumDataNodes, maxNumDataNodes, clusterName1, nodeConfigurationSource, numClientNodes, - enableHttpPipelining, nodePrefix, mockPlugins(), Function.identity()); + nodePrefix, mockPlugins(), Function.identity()); InternalTestCluster cluster1 = new InternalTestCluster(clusterSeed, baseDir, masterNodes, autoManageMinMasterNodes, minNumDataNodes, maxNumDataNodes, clusterName2, nodeConfigurationSource, numClientNodes, - enableHttpPipelining, nodePrefix, mockPlugins(), Function.identity()); + nodePrefix, mockPlugins(), Function.identity()); assertClusters(cluster0, cluster1, false); long seed = randomLong(); @@ -280,12 +275,11 @@ public class InternalTestClusterTests extends ESTestCase { .put(NetworkModule.TRANSPORT_TYPE_KEY, transportClient).build(); } }; - boolean enableHttpPipelining = randomBoolean(); String nodePrefix = "test"; Path baseDir = createTempDir(); InternalTestCluster cluster = new InternalTestCluster(clusterSeed, baseDir, masterNodes, true, minNumDataNodes, maxNumDataNodes, clusterName1, nodeConfigurationSource, numClientNodes, - enableHttpPipelining, nodePrefix, mockPlugins(), Function.identity()); + nodePrefix, mockPlugins(), Function.identity()); try { cluster.beforeTest(random(), 0.0); final int originalMasterCount = cluster.numMasterNodes(); @@ -390,7 +384,7 @@ public class InternalTestClusterTests extends ESTestCase { return Settings.builder() .put(NetworkModule.TRANSPORT_TYPE_KEY, transportClient).build(); } - }, 0, randomBoolean(), "", mockPlugins(), Function.identity()); + }, 0, "", mockPlugins(), Function.identity()); cluster.beforeTest(random(), 0.0); List roles = new ArrayList<>(); for (int i = 0; i < numNodes; i++) { @@ -473,13 +467,12 @@ public class InternalTestClusterTests extends ESTestCase { .put(NetworkModule.TRANSPORT_TYPE_KEY, transportClient).build(); } }; - boolean enableHttpPipelining = randomBoolean(); String nodePrefix = "test"; Path baseDir = createTempDir(); List> plugins = new ArrayList<>(mockPlugins()); plugins.add(NodeAttrCheckPlugin.class); InternalTestCluster cluster = new InternalTestCluster(randomLong(), baseDir, false, true, 2, 2, - "test", nodeConfigurationSource, 0, enableHttpPipelining, nodePrefix, + "test", nodeConfigurationSource, 0, nodePrefix, plugins, Function.identity()); try { cluster.beforeTest(random(), 0.0); diff --git a/x-pack/plugin/security/src/test/java/org/elasticsearch/xpack/security/audit/index/IndexAuditTrailTests.java b/x-pack/plugin/security/src/test/java/org/elasticsearch/xpack/security/audit/index/IndexAuditTrailTests.java index ec448f14e91..dab3d023f65 100644 --- a/x-pack/plugin/security/src/test/java/org/elasticsearch/xpack/security/audit/index/IndexAuditTrailTests.java +++ b/x-pack/plugin/security/src/test/java/org/elasticsearch/xpack/security/audit/index/IndexAuditTrailTests.java @@ -214,7 +214,7 @@ public class IndexAuditTrailTests extends SecurityIntegTestCase { mockPlugins.add(getTestTransportPlugin()); } remoteCluster = new InternalTestCluster(randomLong(), createTempDir(), false, true, numNodes, numNodes, cluster2Name, - cluster2SettingsSource, 0, false, SECOND_CLUSTER_NODE_PREFIX, mockPlugins, + cluster2SettingsSource, 0, SECOND_CLUSTER_NODE_PREFIX, mockPlugins, useSecurity ? getClientWrapper() : Function.identity()); remoteCluster.beforeTest(random(), 0.5); diff --git a/x-pack/plugin/security/src/test/java/org/elasticsearch/xpack/security/audit/index/RemoteIndexAuditTrailStartingTests.java b/x-pack/plugin/security/src/test/java/org/elasticsearch/xpack/security/audit/index/RemoteIndexAuditTrailStartingTests.java index 7002803a3d4..96bba962237 100644 --- a/x-pack/plugin/security/src/test/java/org/elasticsearch/xpack/security/audit/index/RemoteIndexAuditTrailStartingTests.java +++ b/x-pack/plugin/security/src/test/java/org/elasticsearch/xpack/security/audit/index/RemoteIndexAuditTrailStartingTests.java @@ -117,7 +117,7 @@ public class RemoteIndexAuditTrailStartingTests extends SecurityIntegTestCase { } }; remoteCluster = new InternalTestCluster(randomLong(), createTempDir(), false, true, numNodes, numNodes, - cluster2Name, cluster2SettingsSource, 0, false, SECOND_CLUSTER_NODE_PREFIX, getMockPlugins(), getClientWrapper()); + cluster2Name, cluster2SettingsSource, 0, SECOND_CLUSTER_NODE_PREFIX, getMockPlugins(), getClientWrapper()); remoteCluster.beforeTest(random(), 0.0); assertNoTimeout(remoteCluster.client().admin().cluster().prepareHealth().setWaitForGreenStatus().get()); } From 94ba78e09a7564fcecee710e07a229d7779f4688 Mon Sep 17 00:00:00 2001 From: lcawl Date: Wed, 23 May 2018 15:42:51 -0700 Subject: [PATCH 05/12] [DOCS] Splits auditing.asciidoc into smaller files --- .../event-types.asciidoc} | 236 +----------------- .../auditing/forwarding-logs.asciidoc | 24 ++ .../docs/en/security/auditing/index.asciidoc | 15 ++ .../security/auditing/output-index.asciidoc | 38 +++ .../security/auditing/output-logfile.asciidoc | 130 ++++++++++ .../en/security/auditing/overview.asciidoc | 40 +++ x-pack/docs/en/security/index.asciidoc | 2 +- 7 files changed, 250 insertions(+), 235 deletions(-) rename x-pack/docs/en/security/{auditing.asciidoc => auditing/event-types.asciidoc} (52%) create mode 100644 x-pack/docs/en/security/auditing/forwarding-logs.asciidoc create mode 100644 x-pack/docs/en/security/auditing/index.asciidoc create mode 100644 x-pack/docs/en/security/auditing/output-index.asciidoc create mode 100644 x-pack/docs/en/security/auditing/output-logfile.asciidoc create mode 100644 x-pack/docs/en/security/auditing/overview.asciidoc diff --git a/x-pack/docs/en/security/auditing.asciidoc b/x-pack/docs/en/security/auditing/event-types.asciidoc similarity index 52% rename from x-pack/docs/en/security/auditing.asciidoc rename to x-pack/docs/en/security/auditing/event-types.asciidoc index ee508a5ac8d..1a6d4b02b0c 100644 --- a/x-pack/docs/en/security/auditing.asciidoc +++ b/x-pack/docs/en/security/auditing/event-types.asciidoc @@ -1,50 +1,10 @@ [role="xpack"] -[[auditing]] -== Auditing security events - -You can enable auditing to keep track of security-related events such as -authentication failures and refused connections. Logging these events enables you -to monitor your cluster for suspicious activity and provides evidence in the -event of an attack. - -[IMPORTANT] -============================================================================ -Audit logs are **disabled** by default. To enable this functionality, you -must set `xpack.security.audit.enabled` to `true` in `elasticsearch.yml`. -============================================================================ - -{Security} provides two ways to persist audit logs: - -* The <> output, which persists events to - a dedicated `_access.log` file on the host's file system. -* The <> output, which persists events to an Elasticsearch index. -The audit index can reside on the same cluster, or a separate cluster. - -By default, only the `logfile` output is used when enabling auditing. -To facilitate browsing and analyzing the events, you can also enable -indexing by setting `xpack.security.audit.outputs` in `elasticsearch.yml`: - -[source,yaml] ----------------------------- -xpack.security.audit.outputs: [ index, logfile ] ----------------------------- - -The `index` output type should be used in conjunction with the `logfile` -output type Because it is possible for the `index` output type to lose -messages if the target index is unavailable, the `access.log` should be -used as the official record of events. - -NOTE: Audit events are batched for indexing so there is a lag before -events appear in the index. You can control how frequently batches of -events are pushed to the index by setting -`xpack.security.audit.index.flush_interval` in `elasticsearch.yml`. - [float] [[audit-event-types]] === Audit event types -Each request may generate multiple audit events. -The following is a list of the events that can be generated: +When you are <>, each request can generate +multiple audit events. The following is a list of the events that can be generated: |====== | `anonymous_access_denied` | | | Logged when a request is denied due to a missing @@ -281,195 +241,3 @@ The log level determines which attributes are included in a log entry. | `rule` | The <> rule that denied the request. |====== - -[float] -[[audit-log-output]] -=== Logfile audit output - -The `logfile` audit output is the default output for auditing. It writes data to -the `_access.log` file in the logs directory. - -[float] -[[audit-log-entry-format]] -=== Log entry format - -The format of a log entry is: - -[source,txt] ----------------------------------------------------------------------------- -[] [] [] [] ----------------------------------------------------------------------------- - -`` :: When the event occurred. You can configure the - timestamp format in `log4j2.properties`. -`` :: Information about the local node that generated - the log entry. You can control what node information - is included by configuring the - {ref}/auditing-settings.html#node-audit-settings[local node info settings]. -`` :: The layer from which this event originated: - `rest`, `transport` or `ip_filter`. -`` :: The type of event that occurred: `anonymous_access_denied`, - `authentication_failed`, `access_denied`, `access_granted`, - `connection_granted`, `connection_denied`. -`` :: A comma-separated list of key-value pairs that contain - data pertaining to the event. Formatted as - `attr1=[val1], attr2=[val2]`. See <> for the attributes that can be included - for each type of event. - -[float] -[[audit-log-settings]] -=== Logfile output settings - -The events and some other information about what gets logged can be -controlled using settings in the `elasticsearch.yml` file. See -{ref}/auditing-settings.html#event-audit-settings[Audited Event Settings] and -{ref}/auditing-settings.html#node-audit-settings[Local Node Info Settings]. - -IMPORTANT: No filtering is performed when auditing, so sensitive data may be -audited in plain text when including the request body in audit events. - -[[logging-file]] -You can also configure how the logfile is written in the `log4j2.properties` -file located in `CONFIG_DIR`. By default, audit information is appended to the -`_access.log` file located in the standard Elasticsearch `logs` directory -(typically located at `$ES_HOME/logs`). The file rolls over on a daily basis. - -[float] -[[audit-log-ignore-policy]] -=== Logfile audit events ignore policies - -The comprehensive audit trail is necessary to ensure accountability. It offers tremendous -value during incident response and can even be required for demonstrating compliance. - -The drawback of an audited system is represented by the inevitable performance penalty incurred. -In all truth, the audit trail spends _I/O ops_ that are not available anymore for the user's queries. -Sometimes the verbosity of the audit trail may become a problem that the event type restrictions, -<>, will not alleviate. - -*Audit events ignore policies* are a finer way to tune the verbosity of the audit trail. -These policies define rules that match audit events which will be _ignored_ (read as: not printed). -Rules match on the values of attributes of audit events and complement the <> method. -Imagine the corpus of audit events and the policies chopping off unwanted events. - -IMPORTANT: When utilizing audit events ignore policies you are acknowledging potential -accountability gaps that could render illegitimate actions undetectable. -Please take time to review these policies whenever your system architecture changes. - -A policy is a named set of filter rules. Each filter rule applies to a single event attribute, -one of the `users`, `realms`, `roles` or `indices` attributes. The filter rule defines -a list of {ref}/query-dsl-regexp-query.html#regexp-syntax[Lucene regexp], *any* of which has to match the value of the audit -event attribute for the rule to match. -A policy matches an event if *all* the rules comprising it match the event. -An audit event is ignored, therefore not printed, if it matches *any* policy. All other -non-matching events are printed as usual. - -All policies are defined under the `xpack.security.audit.logfile.events.ignore_filters` -settings namespace. For example, the following policy named _example1_ matches -events from the _kibana_ or _admin_user_ principals **and** operating over indices of the -wildcard form _app-logs*_: - -[source,yaml] ----------------------------- -xpack.security.audit.logfile.events.ignore_filters: - example1: - users: ["kibana", "admin_user"] - indices: ["app-logs*"] ----------------------------- - -An audit event generated by the _kibana_ user and operating over multiple indices -, some of which do not match the indices wildcard, will not match. -As expected, operations generated by all other users (even operating only on indices that -match the _indices_ filter) will not match this policy either. - -Audit events of different types may have <>. -If an event does not contain an attribute for which some policy defines filters, the -event will not match the policy. -For example, the following policy named _example2_, will never match `authentication_success` or -`authentication_failed` events, irrespective of the user's roles, because these -event schemas do not contain the `role` attribute: - -[source,yaml] ----------------------------- -xpack.security.audit.logfile.events.ignore_filters: - example2: - roles: ["admin", "ops_admin_*"] ----------------------------- - -Likewise, any events of users with multiple roles, some of which do not match the -regexps will not match this policy. - -For completeness, although practical use cases should be sparse, a filter can match -a missing attribute of an event, using the empty string ("") or the empty list ([]). -For example, the following policy will match events that do not have the `indices` -attribute (`anonymous_access_denied`, `authentication_success` and other types) as well -as events over the _next_ index. - -[source,yaml] ----------------------------- -xpack.security.audit.logfile.events.ignore_filters: - example3: - indices: ["next", ""] ----------------------------- - - -[float] -[[audit-index]] -=== Index audit output - -In addition to logging to a file, you can store audit logs in Elasticsearch -rolling indices. These indices can be either on the same cluster, or on a -remote cluster. You configure the following settings in -`elasticsearch.yml` to control how audit entries are indexed. To enable -this output, you need to configure the setting `xpack.security.audit.outputs` -in the `elasticsearch.yml` file: - -[source,yaml] ----------------------------- -xpack.security.audit.outputs: [ index, logfile ] ----------------------------- - -For more configuration options, see -{ref}/auditing-settings.html#index-audit-settings[Audit log indexing configuration settings]. - -IMPORTANT: No filtering is performed when auditing, so sensitive data may be -audited in plain text when including the request body in audit events. - -[float] -==== Audit index settings - -You can also configure settings for the indices that the events are stored in. -These settings are configured in the `xpack.security.audit.index.settings` namespace -in `elasticsearch.yml`. For example, the following configuration sets the -number of shards and replicas to 1 for the audit indices: - -[source,yaml] ----------------------------- -xpack.security.audit.index.settings: - index: - number_of_shards: 1 - number_of_replicas: 1 ----------------------------- - -[float] -==== Forwarding audit logs to a remote cluster - -To index audit events to a remote Elasticsearch cluster, you configure -the following `xpack.security.audit.index.client` settings: - -* `xpack.security.audit.index.client.hosts` -* `xpack.security.audit.index.client.cluster.name` -* `xpack.security.audit.index.client.xpack.security.user` - -For more information about these settings, see -{ref}/auditing-settings.html#remote-audit-settings[Remote Audit Log Indexing Configuration Settings]. - -You can pass additional settings to the remote client by specifying them in the -`xpack.security.audit.index.client` namespace. For example, to allow the remote -client to discover all of the nodes in the remote cluster you can specify the -`client.transport.sniff` setting: - -[source,yaml] ----------------------------- -xpack.security.audit.index.client.transport.sniff: true ----------------------------- diff --git a/x-pack/docs/en/security/auditing/forwarding-logs.asciidoc b/x-pack/docs/en/security/auditing/forwarding-logs.asciidoc new file mode 100644 index 00000000000..01ed0f72e74 --- /dev/null +++ b/x-pack/docs/en/security/auditing/forwarding-logs.asciidoc @@ -0,0 +1,24 @@ +[role="xpack"] +[float] +[[forwarding-audit-logfiles]] +==== Forwarding audit logs to a remote cluster + +To index audit events to a remote Elasticsearch cluster, you configure +the following `xpack.security.audit.index.client` settings: + +* `xpack.security.audit.index.client.hosts` +* `xpack.security.audit.index.client.cluster.name` +* `xpack.security.audit.index.client.xpack.security.user` + +For more information about these settings, see +{ref}/auditing-settings.html#remote-audit-settings[Remote Audit Log Indexing Configuration Settings]. + +You can pass additional settings to the remote client by specifying them in the +`xpack.security.audit.index.client` namespace. For example, to allow the remote +client to discover all of the nodes in the remote cluster you can specify the +`client.transport.sniff` setting: + +[source,yaml] +---------------------------- +xpack.security.audit.index.client.transport.sniff: true +---------------------------- diff --git a/x-pack/docs/en/security/auditing/index.asciidoc b/x-pack/docs/en/security/auditing/index.asciidoc new file mode 100644 index 00000000000..e82fd4397fb --- /dev/null +++ b/x-pack/docs/en/security/auditing/index.asciidoc @@ -0,0 +1,15 @@ + +:edit_url: https://github.com/elastic/elasticsearch/edit/{branch}/x-pack/docs/en/security/auditing/overview.asciidoc +include::overview.asciidoc[] + +:edit_url: https://github.com/elastic/elasticsearch/edit/{branch}/x-pack/docs/en/security/auditing/event-types.asciidoc +include::event-types.asciidoc[] + +:edit_url: https://github.com/elastic/elasticsearch/edit/{branch}/x-pack/docs/en/security/auditing/output-logfile.asciidoc +include::output-logfile.asciidoc[] + +:edit_url: https://github.com/elastic/elasticsearch/edit/{branch}/x-pack/docs/en/security/auditing/output-index.asciidoc +include::output-index.asciidoc[] + +:edit_url: https://github.com/elastic/elasticsearch/edit/{branch}/x-pack/docs/en/security/auditing/forwarding-logs.asciidoc +include::forwarding-logs.asciidoc[] \ No newline at end of file diff --git a/x-pack/docs/en/security/auditing/output-index.asciidoc b/x-pack/docs/en/security/auditing/output-index.asciidoc new file mode 100644 index 00000000000..e3ba805d715 --- /dev/null +++ b/x-pack/docs/en/security/auditing/output-index.asciidoc @@ -0,0 +1,38 @@ +[role="xpack"] +[float] +[[audit-index]] +=== Index audit output + +In addition to logging to a file, you can store audit logs in Elasticsearch +rolling indices. These indices can be either on the same cluster, or on a +remote cluster. You configure the following settings in +`elasticsearch.yml` to control how audit entries are indexed. To enable +this output, you need to configure the setting `xpack.security.audit.outputs` +in the `elasticsearch.yml` file: + +[source,yaml] +---------------------------- +xpack.security.audit.outputs: [ index, logfile ] +---------------------------- + +For more configuration options, see +{ref}/auditing-settings.html#index-audit-settings[Audit log indexing configuration settings]. + +IMPORTANT: No filtering is performed when auditing, so sensitive data may be +audited in plain text when including the request body in audit events. + +[float] +==== Audit index settings + +You can also configure settings for the indices that the events are stored in. +These settings are configured in the `xpack.security.audit.index.settings` namespace +in `elasticsearch.yml`. For example, the following configuration sets the +number of shards and replicas to 1 for the audit indices: + +[source,yaml] +---------------------------- +xpack.security.audit.index.settings: + index: + number_of_shards: 1 + number_of_replicas: 1 +---------------------------- diff --git a/x-pack/docs/en/security/auditing/output-logfile.asciidoc b/x-pack/docs/en/security/auditing/output-logfile.asciidoc new file mode 100644 index 00000000000..095f57cf61e --- /dev/null +++ b/x-pack/docs/en/security/auditing/output-logfile.asciidoc @@ -0,0 +1,130 @@ +[role="xpack"] +[float] +[[audit-log-output]] +=== Logfile audit output + +The `logfile` audit output is the default output for auditing. It writes data to +the `_access.log` file in the logs directory. + +[float] +[[audit-log-entry-format]] +=== Log entry format + +The format of a log entry is: + +[source,txt] +---------------------------------------------------------------------------- +[] [] [] [] +---------------------------------------------------------------------------- + +`` :: When the event occurred. You can configure the + timestamp format in `log4j2.properties`. +`` :: Information about the local node that generated + the log entry. You can control what node information + is included by configuring the + {ref}/auditing-settings.html#node-audit-settings[local node info settings]. +`` :: The layer from which this event originated: + `rest`, `transport` or `ip_filter`. +`` :: The type of event that occurred: `anonymous_access_denied`, + `authentication_failed`, `access_denied`, `access_granted`, + `connection_granted`, `connection_denied`. +`` :: A comma-separated list of key-value pairs that contain + data pertaining to the event. Formatted as + `attr1=[val1], attr2=[val2]`. See <> for the attributes that can be included + for each type of event. + +[float] +[[audit-log-settings]] +=== Logfile output settings + +The events and some other information about what gets logged can be +controlled using settings in the `elasticsearch.yml` file. See +{ref}/auditing-settings.html#event-audit-settings[Audited Event Settings] and +{ref}/auditing-settings.html#node-audit-settings[Local Node Info Settings]. + +IMPORTANT: No filtering is performed when auditing, so sensitive data may be +audited in plain text when including the request body in audit events. + +[[logging-file]] +You can also configure how the logfile is written in the `log4j2.properties` +file located in `CONFIG_DIR`. By default, audit information is appended to the +`_access.log` file located in the standard Elasticsearch `logs` directory +(typically located at `$ES_HOME/logs`). The file rolls over on a daily basis. + +[float] +[[audit-log-ignore-policy]] +=== Logfile audit events ignore policies + +The comprehensive audit trail is necessary to ensure accountability. It offers tremendous +value during incident response and can even be required for demonstrating compliance. + +The drawback of an audited system is represented by the inevitable performance penalty incurred. +In all truth, the audit trail spends _I/O ops_ that are not available anymore for the user's queries. +Sometimes the verbosity of the audit trail may become a problem that the event type restrictions, +<>, will not alleviate. + +*Audit events ignore policies* are a finer way to tune the verbosity of the audit trail. +These policies define rules that match audit events which will be _ignored_ (read as: not printed). +Rules match on the values of attributes of audit events and complement the <> method. +Imagine the corpus of audit events and the policies chopping off unwanted events. + +IMPORTANT: When utilizing audit events ignore policies you are acknowledging potential +accountability gaps that could render illegitimate actions undetectable. +Please take time to review these policies whenever your system architecture changes. + +A policy is a named set of filter rules. Each filter rule applies to a single event attribute, +one of the `users`, `realms`, `roles` or `indices` attributes. The filter rule defines +a list of {ref}/query-dsl-regexp-query.html#regexp-syntax[Lucene regexp], *any* of which has to match the value of the audit +event attribute for the rule to match. +A policy matches an event if *all* the rules comprising it match the event. +An audit event is ignored, therefore not printed, if it matches *any* policy. All other +non-matching events are printed as usual. + +All policies are defined under the `xpack.security.audit.logfile.events.ignore_filters` +settings namespace. For example, the following policy named _example1_ matches +events from the _kibana_ or _admin_user_ principals **and** operating over indices of the +wildcard form _app-logs*_: + +[source,yaml] +---------------------------- +xpack.security.audit.logfile.events.ignore_filters: + example1: + users: ["kibana", "admin_user"] + indices: ["app-logs*"] +---------------------------- + +An audit event generated by the _kibana_ user and operating over multiple indices +, some of which do not match the indices wildcard, will not match. +As expected, operations generated by all other users (even operating only on indices that +match the _indices_ filter) will not match this policy either. + +Audit events of different types may have <>. +If an event does not contain an attribute for which some policy defines filters, the +event will not match the policy. +For example, the following policy named _example2_, will never match `authentication_success` or +`authentication_failed` events, irrespective of the user's roles, because these +event schemas do not contain the `role` attribute: + +[source,yaml] +---------------------------- +xpack.security.audit.logfile.events.ignore_filters: + example2: + roles: ["admin", "ops_admin_*"] +---------------------------- + +Likewise, any events of users with multiple roles, some of which do not match the +regexps will not match this policy. + +For completeness, although practical use cases should be sparse, a filter can match +a missing attribute of an event, using the empty string ("") or the empty list ([]). +For example, the following policy will match events that do not have the `indices` +attribute (`anonymous_access_denied`, `authentication_success` and other types) as well +as events over the _next_ index. + +[source,yaml] +---------------------------- +xpack.security.audit.logfile.events.ignore_filters: + example3: + indices: ["next", ""] +---------------------------- diff --git a/x-pack/docs/en/security/auditing/overview.asciidoc b/x-pack/docs/en/security/auditing/overview.asciidoc new file mode 100644 index 00000000000..b60122612a0 --- /dev/null +++ b/x-pack/docs/en/security/auditing/overview.asciidoc @@ -0,0 +1,40 @@ +[role="xpack"] +[[auditing]] +== Auditing security events + +You can enable auditing to keep track of security-related events such as +authentication failures and refused connections. Logging these events enables you +to monitor your cluster for suspicious activity and provides evidence in the +event of an attack. + +[IMPORTANT] +============================================================================ +Audit logs are **disabled** by default. To enable this functionality, you +must set `xpack.security.audit.enabled` to `true` in `elasticsearch.yml`. +============================================================================ + +{Security} provides two ways to persist audit logs: + +* The <> output, which persists events to + a dedicated `_access.log` file on the host's file system. +* The <> output, which persists events to an Elasticsearch index. +The audit index can reside on the same cluster, or a separate cluster. + +By default, only the `logfile` output is used when enabling auditing. +To facilitate browsing and analyzing the events, you can also enable +indexing by setting `xpack.security.audit.outputs` in `elasticsearch.yml`: + +[source,yaml] +---------------------------- +xpack.security.audit.outputs: [ index, logfile ] +---------------------------- + +The `index` output type should be used in conjunction with the `logfile` +output type Because it is possible for the `index` output type to lose +messages if the target index is unavailable, the `access.log` should be +used as the official record of events. + +NOTE: Audit events are batched for indexing so there is a lag before +events appear in the index. You can control how frequently batches of +events are pushed to the index by setting +`xpack.security.audit.index.flush_interval` in `elasticsearch.yml`. diff --git a/x-pack/docs/en/security/index.asciidoc b/x-pack/docs/en/security/index.asciidoc index 96e9287ca01..ed891b9a1ba 100644 --- a/x-pack/docs/en/security/index.asciidoc +++ b/x-pack/docs/en/security/index.asciidoc @@ -108,7 +108,7 @@ include::authentication/overview.asciidoc[] include::authorization/overview.asciidoc[] :edit_url: https://github.com/elastic/elasticsearch/edit/{branch}/x-pack/docs/en/security/auditing.asciidoc -include::auditing.asciidoc[] +include::auditing/index.asciidoc[] :edit_url: https://github.com/elastic/elasticsearch/edit/{branch}/x-pack/docs/en/security/securing-communications.asciidoc include::securing-communications.asciidoc[] From 699153edc7789890bc4f96890dbfd72aba5d2d38 Mon Sep 17 00:00:00 2001 From: Igor Motov Date: Wed, 23 May 2018 17:35:09 -0400 Subject: [PATCH 06/12] Fix GeoShapeQueryBuilder serialization after backport Aligns the routing value serialization version after backport of #30760 --- .../org/elasticsearch/index/query/GeoShapeQueryBuilder.java | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/server/src/main/java/org/elasticsearch/index/query/GeoShapeQueryBuilder.java b/server/src/main/java/org/elasticsearch/index/query/GeoShapeQueryBuilder.java index 03fb2098180..490f8ee72c4 100644 --- a/server/src/main/java/org/elasticsearch/index/query/GeoShapeQueryBuilder.java +++ b/server/src/main/java/org/elasticsearch/index/query/GeoShapeQueryBuilder.java @@ -170,7 +170,7 @@ public class GeoShapeQueryBuilder extends AbstractQueryBuilder Date: Wed, 23 May 2018 16:41:04 -0700 Subject: [PATCH 07/12] [DOCS] Fixes typos in security settings --- x-pack/docs/en/settings/security-settings.asciidoc | 2 +- x-pack/docs/en/settings/ssl-settings.asciidoc | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/x-pack/docs/en/settings/security-settings.asciidoc b/x-pack/docs/en/settings/security-settings.asciidoc index 2c4df292857..4e9d85f1900 100644 --- a/x-pack/docs/en/settings/security-settings.asciidoc +++ b/x-pack/docs/en/settings/security-settings.asciidoc @@ -1145,7 +1145,7 @@ Path to the PEM encoded file containing the private key. The passphrase that is used to decrypt the private key. This value is optional as the key might not be encrypted. -`xpack.ssl.secure_key_passphrase` ({<>):: +`xpack.ssl.secure_key_passphrase` (<>):: The passphrase that is used to decrypt the private key. This value is optional as the key might not be encrypted. diff --git a/x-pack/docs/en/settings/ssl-settings.asciidoc b/x-pack/docs/en/settings/ssl-settings.asciidoc index 722aeeea155..655dfb74a64 100644 --- a/x-pack/docs/en/settings/ssl-settings.asciidoc +++ b/x-pack/docs/en/settings/ssl-settings.asciidoc @@ -90,7 +90,7 @@ Path to the keystore that holds the private key and certificate. +{ssl-prefix}.ssl.keystore.password+:: Password to the keystore. -+{ssl-prefix}.ssl.keystore.secure_password` (<>):: ++{ssl-prefix}.ssl.keystore.secure_password+ (<>):: Password to the keystore. +{ssl-prefix}.ssl.keystore.key_password+:: From e8b543b8cd6e19e29e16df19524356ab014bce03 Mon Sep 17 00:00:00 2001 From: Jason Tedor Date: Wed, 23 May 2018 23:15:19 -0400 Subject: [PATCH 08/12] Force stable file modes for built packages (#30823) If you have an unusual umask (e.g., 0002) and clone the GitHub repository then files that we stick into our packages like the README.textile and the license will have a file mode of 0664 on disk yet we expect them to be 0644. Additionally, the same thing happens with compiled artifacts like JARs. We try to set a default file mode yet it does not seem to take everywhere. This commit adds explicit file modes in some places that we were relying on the defaults to ensure that the built artifacts have a consistent file mode regardless of the underlying build host. --- distribution/build.gradle | 2 ++ distribution/packages/build.gradle | 10 ++++++++++ 2 files changed, 12 insertions(+) diff --git a/distribution/build.gradle b/distribution/build.gradle index 940a4152bfd..5f6f0b1579c 100644 --- a/distribution/build.gradle +++ b/distribution/build.gradle @@ -242,6 +242,8 @@ configure(subprojects.findAll { ['archives', 'packages'].contains(it.name) }) { if (it.relativePath.segments[-2] == 'bin') { // bin files, wherever they are within modules (eg platform specific) should be executable it.mode = 0755 + } else { + it.mode = 0644 } } if (oss) { diff --git a/distribution/packages/build.gradle b/distribution/packages/build.gradle index a6759a2e4f1..b15e5668e3d 100644 --- a/distribution/packages/build.gradle +++ b/distribution/packages/build.gradle @@ -122,6 +122,7 @@ Closure commonPackageConfig(String type, boolean oss) { } from(rootProject.projectDir) { include 'README.textile' + fileMode 0644 } into('modules') { with copySpec { @@ -135,6 +136,11 @@ Closure commonPackageConfig(String type, boolean oss) { for (int i = segments.length - 2; i > 0 && segments[i] != 'modules'; --i) { directory('/' + segments[0..i].join('/'), 0755) } + if (segments[-2] == 'bin') { + fcp.mode = 0755 + } else { + fcp.mode = 0644 + } } } } @@ -153,6 +159,7 @@ Closure commonPackageConfig(String type, boolean oss) { include oss ? 'APACHE-LICENSE-2.0.txt' : 'ELASTIC-LICENSE.txt' rename { 'LICENSE.txt' } } + fileMode 0644 } } @@ -180,14 +187,17 @@ Closure commonPackageConfig(String type, boolean oss) { // ========= systemd ========= into('/usr/lib/tmpfiles.d') { from "${packagingFiles}/systemd/elasticsearch.conf" + fileMode 0644 } into('/usr/lib/systemd/system') { fileType CONFIG | NOREPLACE from "${packagingFiles}/systemd/elasticsearch.service" + fileMode 0644 } into('/usr/lib/sysctl.d') { fileType CONFIG | NOREPLACE from "${packagingFiles}/systemd/sysctl/elasticsearch.conf" + fileMode 0644 } // ========= sysV init ========= From 2c7559c575c5654a2f7842746efea3bf4429fc0f Mon Sep 17 00:00:00 2001 From: Ryan Ernst Date: Wed, 23 May 2018 23:20:13 -0700 Subject: [PATCH 09/12] Packaging: Ensure upgrade_is_oss flag file is always deleted (#30732) This commit ensures the delete of the upgrade_is_oss indicator for the packaging tests is always deleted before each run. It works by moving the check on version which skips the task into the doFirst block, replacing the onlyIf. closes #30682 --- .../elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/buildSrc/src/main/groovy/org/elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy b/buildSrc/src/main/groovy/org/elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy index 6e8a5fd15ed..455d30f95db 100644 --- a/buildSrc/src/main/groovy/org/elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy +++ b/buildSrc/src/main/groovy/org/elasticsearch/gradle/vagrant/VagrantTestPlugin.groovy @@ -11,6 +11,7 @@ import org.gradle.api.internal.artifacts.dependencies.DefaultProjectDependency import org.gradle.api.tasks.Copy import org.gradle.api.tasks.Delete import org.gradle.api.tasks.Exec +import org.gradle.api.tasks.StopExecutionException import org.gradle.api.tasks.TaskState import static java.util.Collections.unmodifiableList @@ -285,8 +286,10 @@ class VagrantTestPlugin implements Plugin { dependsOn copyPackagingArchives doFirst { project.delete("${archivesDir}/upgrade_is_oss") + if (project.extensions.esvagrant.upgradeFromVersion.before('6.3.0')) { + throw new StopExecutionException("upgrade version is before 6.3.0") + } } - onlyIf { project.extensions.esvagrant.upgradeFromVersion.onOrAfter('6.3.0') } file "${archivesDir}/upgrade_is_oss" contents '' } From 0bdfb5c5b5b79f4c9d99e112420b6d0d9d95537d Mon Sep 17 00:00:00 2001 From: Simon Willnauer Date: Thu, 24 May 2018 09:46:48 +0200 Subject: [PATCH 10/12] Send client headers from TransportClient (#30803) This change adds a simple header to the transport client that is present on the servers thread context that ensures we can detect if a transport client talks to the server in a specific request. This change also adds a header for xpack to detect if the client has xpack installed. --- .../client/transport/TransportClient.java | 4 +++- .../elasticsearch/transport/TcpTransport.java | 1 + .../client/AbstractClientHeadersTestCase.java | 3 ++- .../client/transport/TransportClientTests.java | 17 +++++++++++++++++ .../xpack/core/XPackClientPlugin.java | 2 ++ 5 files changed, 25 insertions(+), 2 deletions(-) diff --git a/server/src/main/java/org/elasticsearch/client/transport/TransportClient.java b/server/src/main/java/org/elasticsearch/client/transport/TransportClient.java index fd56186fd9d..2aeea08d1a5 100644 --- a/server/src/main/java/org/elasticsearch/client/transport/TransportClient.java +++ b/server/src/main/java/org/elasticsearch/client/transport/TransportClient.java @@ -19,6 +19,7 @@ package org.elasticsearch.client.transport; +import org.elasticsearch.common.util.concurrent.ThreadContext; import org.elasticsearch.core.internal.io.IOUtils; import org.elasticsearch.action.Action; import org.elasticsearch.action.ActionListener; @@ -129,7 +130,8 @@ public abstract class TransportClient extends AbstractClient { providedSettings = Settings.builder().put(providedSettings).put(Node.NODE_NAME_SETTING.getKey(), "_client_").build(); } final PluginsService pluginsService = newPluginService(providedSettings, plugins); - final Settings settings = Settings.builder().put(defaultSettings).put(pluginsService.updatedSettings()).build(); + final Settings settings = Settings.builder().put(defaultSettings).put(pluginsService.updatedSettings()).put(ThreadContext.PREFIX + + "." + "transport_client", true).build(); final List resourcesToClose = new ArrayList<>(); final ThreadPool threadPool = new ThreadPool(settings); resourcesToClose.add(() -> ThreadPool.terminate(threadPool, 10, TimeUnit.SECONDS)); diff --git a/server/src/main/java/org/elasticsearch/transport/TcpTransport.java b/server/src/main/java/org/elasticsearch/transport/TcpTransport.java index 04a882f3e8b..0b3d4e1b0a1 100644 --- a/server/src/main/java/org/elasticsearch/transport/TcpTransport.java +++ b/server/src/main/java/org/elasticsearch/transport/TcpTransport.java @@ -1438,6 +1438,7 @@ public abstract class TcpTransport extends AbstractLifecycleComponent implements streamIn = new NamedWriteableAwareStreamInput(streamIn, namedWriteableRegistry); streamIn.setVersion(version); threadPool.getThreadContext().readHeaders(streamIn); + threadPool.getThreadContext().putTransient("_remote_address", remoteAddress); if (TransportStatus.isRequest(status)) { handleRequest(channel, profileName, streamIn, requestId, messageLengthBytes, version, remoteAddress, status); } else { diff --git a/server/src/test/java/org/elasticsearch/client/AbstractClientHeadersTestCase.java b/server/src/test/java/org/elasticsearch/client/AbstractClientHeadersTestCase.java index 8c1b22f7fb1..db9f9d83c81 100644 --- a/server/src/test/java/org/elasticsearch/client/AbstractClientHeadersTestCase.java +++ b/server/src/test/java/org/elasticsearch/client/AbstractClientHeadersTestCase.java @@ -139,6 +139,8 @@ public abstract class AbstractClientHeadersTestCase extends ESTestCase { protected static void assertHeaders(Map headers, Map expected) { assertNotNull(headers); + headers = new HashMap<>(headers); + headers.remove("transport_client"); // default header on TPC assertEquals(expected.size(), headers.size()); for (Map.Entry expectedEntry : expected.entrySet()) { assertEquals(headers.get(expectedEntry.getKey()), expectedEntry.getValue()); @@ -146,7 +148,6 @@ public abstract class AbstractClientHeadersTestCase extends ESTestCase { } protected static void assertHeaders(ThreadPool pool) { - Map headers = new HashMap<>(); Settings asSettings = HEADER_SETTINGS.getAsSettings(ThreadContext.PREFIX); assertHeaders(pool.getThreadContext().getHeaders(), asSettings.keySet().stream().collect(Collectors.toMap(Function.identity(), k -> asSettings.get(k)))); diff --git a/server/src/test/java/org/elasticsearch/client/transport/TransportClientTests.java b/server/src/test/java/org/elasticsearch/client/transport/TransportClientTests.java index c97418bae37..1830698d90c 100644 --- a/server/src/test/java/org/elasticsearch/client/transport/TransportClientTests.java +++ b/server/src/test/java/org/elasticsearch/client/transport/TransportClientTests.java @@ -26,6 +26,7 @@ import org.elasticsearch.common.io.stream.NamedWriteableRegistry.Entry; import org.elasticsearch.common.io.stream.StreamInput; import org.elasticsearch.common.io.stream.StreamOutput; import org.elasticsearch.common.settings.Settings; +import org.elasticsearch.common.util.concurrent.ThreadContext; import org.elasticsearch.env.Environment; import org.elasticsearch.plugins.Plugin; import org.elasticsearch.test.ESTestCase; @@ -63,6 +64,17 @@ public class TransportClientTests extends ESTestCase { } } + public void testDefaultHeaderContainsPlugins() { + Settings baseSettings = Settings.builder() + .put(Environment.PATH_HOME_SETTING.getKey(), createTempDir()) + .build(); + try (TransportClient client = new MockTransportClient(baseSettings, Arrays.asList(MockPlugin.class))) { + ThreadContext threadContext = client.threadPool().getThreadContext(); + assertEquals("true", threadContext.getHeader("transport_client")); + assertEquals("true", threadContext.getHeader("test")); + } + } + public static class MockPlugin extends Plugin { @Override @@ -70,6 +82,11 @@ public class TransportClientTests extends ESTestCase { return Arrays.asList(new Entry[]{ new Entry(MockNamedWriteable.class, MockNamedWriteable.NAME, MockNamedWriteable::new)}); } + @Override + public Settings additionalSettings() { + return Settings.builder().put(ThreadContext.PREFIX + "." + "test", true).build(); + } + public class MockNamedWriteable implements NamedWriteable { static final String NAME = "mockNamedWritable"; diff --git a/x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/XPackClientPlugin.java b/x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/XPackClientPlugin.java index 4853588bd3e..c719cdcbd02 100644 --- a/x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/XPackClientPlugin.java +++ b/x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/XPackClientPlugin.java @@ -16,6 +16,7 @@ import org.elasticsearch.common.settings.Setting; import org.elasticsearch.common.settings.Settings; import org.elasticsearch.common.util.BigArrays; import org.elasticsearch.common.util.PageCacheRecycler; +import org.elasticsearch.common.util.concurrent.ThreadContext; import org.elasticsearch.common.xcontent.NamedXContentRegistry; import org.elasticsearch.indices.breaker.CircuitBreakerService; import org.elasticsearch.license.DeleteLicenseAction; @@ -193,6 +194,7 @@ public class XPackClientPlugin extends Plugin implements ActionPlugin, NetworkPl final Settings.Builder builder = Settings.builder(); builder.put(SecuritySettings.addTransportSettings(settings)); builder.put(SecuritySettings.addUserSettings(settings)); + builder.put(ThreadContext.PREFIX + "." + "has_xpack", true); return builder.build(); } else { return Settings.EMPTY; From ff0b6c795ae12d8eca7274fe9712993cdeb7a820 Mon Sep 17 00:00:00 2001 From: David Turner Date: Thu, 24 May 2018 09:05:09 +0100 Subject: [PATCH 11/12] Decouple ClusterStateTaskListener & ClusterApplier (#30809) Today, the `ClusterApplier` and `MasterService` both use the `ClusterStateTaskListener` interface to notify their callers when asynchronous activities have completed. However, this is not wholly appropriate: none of the callers into the `ClusterApplier` care about the `ClusterState` arguments that they receive. This change introduces a dedicated ClusterApplyListener interface for callers into the `ClusterApplier`, to distinguish these listeners from the real `ClusterStateTaskListener`s that are waiting for responses from the `MasterService`. --- .../cluster/service/ClusterApplier.java | 21 ++++++++++- .../service/ClusterApplierService.java | 33 ++++++++--------- .../discovery/single/SingleNodeDiscovery.java | 8 ++-- .../discovery/zen/ZenDiscovery.java | 8 ++-- .../service/ClusterApplierServiceTests.java | 37 ++++++++++--------- .../single/SingleNodeDiscoveryTests.java | 5 +-- .../discovery/zen/ZenDiscoveryUnitTests.java | 5 +-- .../store/IndicesStoreIntegrationIT.java | 5 ++- .../test/ClusterServiceUtils.java | 10 ++--- 9 files changed, 71 insertions(+), 61 deletions(-) diff --git a/server/src/main/java/org/elasticsearch/cluster/service/ClusterApplier.java b/server/src/main/java/org/elasticsearch/cluster/service/ClusterApplier.java index 0a2ef347d06..c587ab272e9 100644 --- a/server/src/main/java/org/elasticsearch/cluster/service/ClusterApplier.java +++ b/server/src/main/java/org/elasticsearch/cluster/service/ClusterApplier.java @@ -20,7 +20,6 @@ package org.elasticsearch.cluster.service; import org.elasticsearch.cluster.ClusterState; -import org.elasticsearch.cluster.ClusterStateTaskListener; import java.util.function.Supplier; @@ -38,11 +37,29 @@ public interface ClusterApplier { * @param clusterStateSupplier the cluster state supplier which provides the latest cluster state to apply * @param listener callback that is invoked after cluster state is applied */ - void onNewClusterState(String source, Supplier clusterStateSupplier, ClusterStateTaskListener listener); + void onNewClusterState(String source, Supplier clusterStateSupplier, ClusterApplyListener listener); /** * Creates a new cluster state builder that is initialized with the cluster name and all initial cluster state customs. */ ClusterState.Builder newClusterStateBuilder(); + /** + * Listener for results of cluster state application + */ + interface ClusterApplyListener { + /** + * Called on successful cluster state application + * @param source information where the cluster state came from + */ + default void onSuccess(String source) { + } + + /** + * Called on failure during cluster state application + * @param source information where the cluster state came from + * @param e exception that occurred + */ + void onFailure(String source, Exception e); + } } diff --git a/server/src/main/java/org/elasticsearch/cluster/service/ClusterApplierService.java b/server/src/main/java/org/elasticsearch/cluster/service/ClusterApplierService.java index 01fa5837387..2fb7c25671c 100644 --- a/server/src/main/java/org/elasticsearch/cluster/service/ClusterApplierService.java +++ b/server/src/main/java/org/elasticsearch/cluster/service/ClusterApplierService.java @@ -27,7 +27,6 @@ import org.elasticsearch.cluster.ClusterStateApplier; import org.elasticsearch.cluster.ClusterStateListener; import org.elasticsearch.cluster.ClusterStateObserver; import org.elasticsearch.cluster.ClusterStateTaskConfig; -import org.elasticsearch.cluster.ClusterStateTaskListener; import org.elasticsearch.cluster.LocalNodeMasterListener; import org.elasticsearch.cluster.NodeConnectionsService; import org.elasticsearch.cluster.TimeoutClusterStateListener; @@ -141,10 +140,10 @@ public class ClusterApplierService extends AbstractLifecycleComponent implements } class UpdateTask extends SourcePrioritizedRunnable implements Function { - final ClusterStateTaskListener listener; + final ClusterApplyListener listener; final Function updateFunction; - UpdateTask(Priority priority, String source, ClusterStateTaskListener listener, + UpdateTask(Priority priority, String source, ClusterApplyListener listener, Function updateFunction) { super(priority, source); this.listener = listener; @@ -301,7 +300,7 @@ public class ClusterApplierService extends AbstractLifecycleComponent implements } public void runOnApplierThread(final String source, Consumer clusterStateConsumer, - final ClusterStateTaskListener listener, Priority priority) { + final ClusterApplyListener listener, Priority priority) { submitStateUpdateTask(source, ClusterStateTaskConfig.build(priority), (clusterState) -> { clusterStateConsumer.accept(clusterState); @@ -311,13 +310,13 @@ public class ClusterApplierService extends AbstractLifecycleComponent implements } public void runOnApplierThread(final String source, Consumer clusterStateConsumer, - final ClusterStateTaskListener listener) { + final ClusterApplyListener listener) { runOnApplierThread(source, clusterStateConsumer, listener, Priority.HIGH); } @Override public void onNewClusterState(final String source, final Supplier clusterStateSupplier, - final ClusterStateTaskListener listener) { + final ClusterApplyListener listener) { Function applyFunction = currentState -> { ClusterState nextState = clusterStateSupplier.get(); if (nextState != null) { @@ -331,12 +330,12 @@ public class ClusterApplierService extends AbstractLifecycleComponent implements private void submitStateUpdateTask(final String source, final ClusterStateTaskConfig config, final Function executor, - final ClusterStateTaskListener listener) { + final ClusterApplyListener listener) { if (!lifecycle.started()) { return; } try { - UpdateTask updateTask = new UpdateTask(config.priority(), source, new SafeClusterStateTaskListener(listener, logger), executor); + UpdateTask updateTask = new UpdateTask(config.priority(), source, new SafeClusterApplyListener(listener, logger), executor); if (config.timeout() != null) { threadPoolExecutor.execute(updateTask, config.timeout(), () -> threadPool.generic().execute( @@ -417,7 +416,7 @@ public class ClusterApplierService extends AbstractLifecycleComponent implements } if (previousClusterState == newClusterState) { - task.listener.clusterStateProcessed(task.source, newClusterState, newClusterState); + task.listener.onSuccess(task.source); TimeValue executionTime = TimeValue.timeValueMillis(Math.max(0, TimeValue.nsecToMSec(currentTimeInNanos() - startTimeNS))); logger.debug("processing [{}]: took [{}] no change in cluster state", task.source, executionTime); warnAboutSlowTaskIfNeeded(executionTime, task.source); @@ -486,7 +485,7 @@ public class ClusterApplierService extends AbstractLifecycleComponent implements callClusterStateListeners(clusterChangedEvent); - task.listener.clusterStateProcessed(task.source, previousClusterState, newClusterState); + task.listener.onSuccess(task.source); } private void callClusterStateAppliers(ClusterChangedEvent clusterChangedEvent) { @@ -511,11 +510,11 @@ public class ClusterApplierService extends AbstractLifecycleComponent implements }); } - private static class SafeClusterStateTaskListener implements ClusterStateTaskListener { - private final ClusterStateTaskListener listener; + private static class SafeClusterApplyListener implements ClusterApplyListener { + private final ClusterApplyListener listener; private final Logger logger; - SafeClusterStateTaskListener(ClusterStateTaskListener listener, Logger logger) { + SafeClusterApplyListener(ClusterApplyListener listener, Logger logger) { this.listener = listener; this.logger = logger; } @@ -532,14 +531,12 @@ public class ClusterApplierService extends AbstractLifecycleComponent implements } @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { try { - listener.clusterStateProcessed(source, oldState, newState); + listener.onSuccess(source); } catch (Exception e) { logger.error(new ParameterizedMessage( - "exception thrown by listener while notifying of cluster state processed from [{}], old cluster state:\n" + - "{}\nnew cluster state:\n{}", - source, oldState, newState), e); + "exception thrown by listener while notifying of cluster state processed from [{}]", source), e); } } } diff --git a/server/src/main/java/org/elasticsearch/discovery/single/SingleNodeDiscovery.java b/server/src/main/java/org/elasticsearch/discovery/single/SingleNodeDiscovery.java index 94ea33d1a16..cd775e29f5a 100644 --- a/server/src/main/java/org/elasticsearch/discovery/single/SingleNodeDiscovery.java +++ b/server/src/main/java/org/elasticsearch/discovery/single/SingleNodeDiscovery.java @@ -22,18 +22,16 @@ package org.elasticsearch.discovery.single; import org.apache.logging.log4j.message.ParameterizedMessage; import org.elasticsearch.cluster.ClusterChangedEvent; import org.elasticsearch.cluster.ClusterState; -import org.elasticsearch.cluster.ClusterStateTaskListener; import org.elasticsearch.cluster.block.ClusterBlocks; import org.elasticsearch.cluster.node.DiscoveryNode; import org.elasticsearch.cluster.node.DiscoveryNodes; import org.elasticsearch.cluster.service.ClusterApplier; +import org.elasticsearch.cluster.service.ClusterApplier.ClusterApplyListener; import org.elasticsearch.cluster.service.MasterService; import org.elasticsearch.common.component.AbstractLifecycleComponent; import org.elasticsearch.common.settings.Settings; import org.elasticsearch.discovery.Discovery; import org.elasticsearch.discovery.DiscoveryStats; -import org.elasticsearch.discovery.zen.PendingClusterStateStats; -import org.elasticsearch.discovery.zen.PublishClusterStateStats; import org.elasticsearch.transport.TransportService; import java.io.IOException; @@ -65,9 +63,9 @@ public class SingleNodeDiscovery extends AbstractLifecycleComponent implements D clusterState = event.state(); CountDownLatch latch = new CountDownLatch(1); - ClusterStateTaskListener listener = new ClusterStateTaskListener() { + ClusterApplyListener listener = new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); ackListener.onNodeAck(transportService.getLocalNode(), null); } diff --git a/server/src/main/java/org/elasticsearch/discovery/zen/ZenDiscovery.java b/server/src/main/java/org/elasticsearch/discovery/zen/ZenDiscovery.java index 02b2822fcf4..55ecf7ca25f 100644 --- a/server/src/main/java/org/elasticsearch/discovery/zen/ZenDiscovery.java +++ b/server/src/main/java/org/elasticsearch/discovery/zen/ZenDiscovery.java @@ -21,7 +21,6 @@ package org.elasticsearch.discovery.zen; import org.apache.logging.log4j.Logger; import org.apache.logging.log4j.message.ParameterizedMessage; -import org.apache.logging.log4j.util.Supplier; import org.elasticsearch.core.internal.io.IOUtils; import org.elasticsearch.ElasticsearchException; import org.elasticsearch.ExceptionsHelper; @@ -34,12 +33,11 @@ import org.elasticsearch.cluster.ClusterStateTaskExecutor; import org.elasticsearch.cluster.ClusterStateTaskListener; import org.elasticsearch.cluster.NotMasterException; import org.elasticsearch.cluster.block.ClusterBlocks; -import org.elasticsearch.cluster.metadata.IndexMetaData; -import org.elasticsearch.cluster.metadata.MetaData; import org.elasticsearch.cluster.node.DiscoveryNode; import org.elasticsearch.cluster.node.DiscoveryNodes; import org.elasticsearch.cluster.routing.allocation.AllocationService; import org.elasticsearch.cluster.service.ClusterApplier; +import org.elasticsearch.cluster.service.ClusterApplier.ClusterApplyListener; import org.elasticsearch.cluster.service.MasterService; import org.elasticsearch.common.Priority; import org.elasticsearch.common.component.AbstractLifecycleComponent; @@ -789,9 +787,9 @@ public class ZenDiscovery extends AbstractLifecycleComponent implements Discover clusterApplier.onNewClusterState("apply cluster state (from master [" + reason + "])", this::clusterState, - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { try { pendingStatesQueue.markAsProcessed(newClusterState); } catch (Exception e) { diff --git a/server/src/test/java/org/elasticsearch/cluster/service/ClusterApplierServiceTests.java b/server/src/test/java/org/elasticsearch/cluster/service/ClusterApplierServiceTests.java index 7a8261776bd..3e7c415db7b 100644 --- a/server/src/test/java/org/elasticsearch/cluster/service/ClusterApplierServiceTests.java +++ b/server/src/test/java/org/elasticsearch/cluster/service/ClusterApplierServiceTests.java @@ -30,6 +30,7 @@ import org.elasticsearch.cluster.NodeConnectionsService; import org.elasticsearch.cluster.block.ClusterBlocks; import org.elasticsearch.cluster.node.DiscoveryNode; import org.elasticsearch.cluster.node.DiscoveryNodes; +import org.elasticsearch.cluster.service.ClusterApplier.ClusterApplyListener; import org.elasticsearch.common.logging.Loggers; import org.elasticsearch.common.settings.ClusterSettings; import org.elasticsearch.common.settings.Settings; @@ -135,9 +136,9 @@ public class ClusterApplierServiceTests extends ESTestCase { clusterApplierService.currentTimeOverride = System.nanoTime(); clusterApplierService.runOnApplierThread("test1", currentState -> clusterApplierService.currentTimeOverride += TimeValue.timeValueSeconds(1).nanos(), - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } @@ -151,9 +152,9 @@ public class ClusterApplierServiceTests extends ESTestCase { clusterApplierService.currentTimeOverride += TimeValue.timeValueSeconds(2).nanos(); throw new IllegalArgumentException("Testing handling of exceptions in the cluster state task"); }, - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { fail(); } @@ -166,9 +167,9 @@ public class ClusterApplierServiceTests extends ESTestCase { // We don't check logging for this on since there is no guarantee that it will occur before our check clusterApplierService.runOnApplierThread("test3", currentState -> {}, - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } @@ -216,9 +217,9 @@ public class ClusterApplierServiceTests extends ESTestCase { clusterApplierService.currentTimeOverride = System.nanoTime(); clusterApplierService.runOnApplierThread("test1", currentState -> clusterApplierService.currentTimeOverride += TimeValue.timeValueSeconds(1).nanos(), - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); processedFirstTask.countDown(); } @@ -234,9 +235,9 @@ public class ClusterApplierServiceTests extends ESTestCase { clusterApplierService.currentTimeOverride += TimeValue.timeValueSeconds(32).nanos(); throw new IllegalArgumentException("Testing handling of exceptions in the cluster state task"); }, - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { fail(); } @@ -247,9 +248,9 @@ public class ClusterApplierServiceTests extends ESTestCase { }); clusterApplierService.runOnApplierThread("test3", currentState -> clusterApplierService.currentTimeOverride += TimeValue.timeValueSeconds(34).nanos(), - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } @@ -262,9 +263,9 @@ public class ClusterApplierServiceTests extends ESTestCase { // We don't check logging for this on since there is no guarantee that it will occur before our check clusterApplierService.runOnApplierThread("test4", currentState -> {}, - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } @@ -340,10 +341,10 @@ public class ClusterApplierServiceTests extends ESTestCase { CountDownLatch latch = new CountDownLatch(1); clusterApplierService.onNewClusterState("test", () -> ClusterState.builder(clusterApplierService.state()).build(), - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } @@ -390,9 +391,9 @@ public class ClusterApplierServiceTests extends ESTestCase { CountDownLatch latch = new CountDownLatch(1); clusterApplierService.onNewClusterState("test", () -> ClusterState.builder(clusterApplierService.state()).build(), - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } diff --git a/server/src/test/java/org/elasticsearch/discovery/single/SingleNodeDiscoveryTests.java b/server/src/test/java/org/elasticsearch/discovery/single/SingleNodeDiscoveryTests.java index 23a510a257f..d045adcaead 100644 --- a/server/src/test/java/org/elasticsearch/discovery/single/SingleNodeDiscoveryTests.java +++ b/server/src/test/java/org/elasticsearch/discovery/single/SingleNodeDiscoveryTests.java @@ -23,7 +23,6 @@ import org.elasticsearch.core.internal.io.IOUtils; import org.elasticsearch.Version; import org.elasticsearch.cluster.ClusterName; import org.elasticsearch.cluster.ClusterState; -import org.elasticsearch.cluster.ClusterStateTaskListener; import org.elasticsearch.cluster.node.DiscoveryNode; import org.elasticsearch.cluster.node.DiscoveryNodes; import org.elasticsearch.cluster.service.ClusterApplier; @@ -72,9 +71,9 @@ public class SingleNodeDiscoveryTests extends ESTestCase { @Override public void onNewClusterState(String source, Supplier clusterStateSupplier, - ClusterStateTaskListener listener) { + ClusterApplyListener listener) { clusterState.set(clusterStateSupplier.get()); - listener.clusterStateProcessed(source, clusterState.get(), clusterState.get()); + listener.onSuccess(source); } }); discovery.start(); diff --git a/server/src/test/java/org/elasticsearch/discovery/zen/ZenDiscoveryUnitTests.java b/server/src/test/java/org/elasticsearch/discovery/zen/ZenDiscoveryUnitTests.java index 0ecb5a296f5..a2121d3ca4e 100644 --- a/server/src/test/java/org/elasticsearch/discovery/zen/ZenDiscoveryUnitTests.java +++ b/server/src/test/java/org/elasticsearch/discovery/zen/ZenDiscoveryUnitTests.java @@ -26,7 +26,6 @@ import org.elasticsearch.cluster.ClusterChangedEvent; import org.elasticsearch.cluster.ClusterModule; import org.elasticsearch.cluster.ClusterName; import org.elasticsearch.cluster.ClusterState; -import org.elasticsearch.cluster.ClusterStateTaskListener; import org.elasticsearch.cluster.ESAllocationTestCase; import org.elasticsearch.cluster.metadata.IndexMetaData; import org.elasticsearch.cluster.metadata.MetaData; @@ -314,8 +313,8 @@ public class ZenDiscoveryUnitTests extends ESTestCase { } @Override - public void onNewClusterState(String source, Supplier clusterStateSupplier, ClusterStateTaskListener listener) { - listener.clusterStateProcessed(source, clusterStateSupplier.get(), clusterStateSupplier.get()); + public void onNewClusterState(String source, Supplier clusterStateSupplier, ClusterApplyListener listener) { + listener.onSuccess(source); } }; ZenDiscovery zenDiscovery = new ZenDiscovery(settings, threadPool, service, new NamedWriteableRegistry(ClusterModule.getNamedWriteables()), diff --git a/server/src/test/java/org/elasticsearch/indices/store/IndicesStoreIntegrationIT.java b/server/src/test/java/org/elasticsearch/indices/store/IndicesStoreIntegrationIT.java index 7f6155979c9..6b57c975736 100644 --- a/server/src/test/java/org/elasticsearch/indices/store/IndicesStoreIntegrationIT.java +++ b/server/src/test/java/org/elasticsearch/indices/store/IndicesStoreIntegrationIT.java @@ -36,6 +36,7 @@ import org.elasticsearch.cluster.routing.ShardRoutingState; import org.elasticsearch.cluster.routing.TestShardRouting; import org.elasticsearch.cluster.routing.allocation.command.MoveAllocationCommand; import org.elasticsearch.cluster.routing.allocation.decider.EnableAllocationDecider; +import org.elasticsearch.cluster.service.ClusterApplier.ClusterApplyListener; import org.elasticsearch.cluster.service.ClusterApplierService; import org.elasticsearch.cluster.service.ClusterService; import org.elasticsearch.common.settings.Settings; @@ -446,9 +447,9 @@ public class IndicesStoreIntegrationIT extends ESIntegTestCase { .routingTable(RoutingTable.builder().add(indexRoutingTableBuilder).build()) .build(); CountDownLatch latch = new CountDownLatch(1); - clusterApplierService.onNewClusterState("test", () -> newState, new ClusterStateTaskListener() { + clusterApplierService.onNewClusterState("test", () -> newState, new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } diff --git a/test/framework/src/main/java/org/elasticsearch/test/ClusterServiceUtils.java b/test/framework/src/main/java/org/elasticsearch/test/ClusterServiceUtils.java index df1e216f4bb..8c4076e327d 100644 --- a/test/framework/src/main/java/org/elasticsearch/test/ClusterServiceUtils.java +++ b/test/framework/src/main/java/org/elasticsearch/test/ClusterServiceUtils.java @@ -24,13 +24,13 @@ import org.elasticsearch.Version; import org.elasticsearch.cluster.ClusterChangedEvent; import org.elasticsearch.cluster.ClusterName; import org.elasticsearch.cluster.ClusterState; -import org.elasticsearch.cluster.ClusterStateTaskListener; import org.elasticsearch.cluster.ClusterStateUpdateTask; import org.elasticsearch.cluster.NodeConnectionsService; import org.elasticsearch.cluster.block.ClusterBlocks; import org.elasticsearch.cluster.node.DiscoveryNode; import org.elasticsearch.cluster.node.DiscoveryNodes; import org.elasticsearch.cluster.service.ClusterApplier; +import org.elasticsearch.cluster.service.ClusterApplier.ClusterApplyListener; import org.elasticsearch.cluster.service.ClusterApplierService; import org.elasticsearch.cluster.service.ClusterService; import org.elasticsearch.cluster.service.MasterService; @@ -72,9 +72,9 @@ public class ClusterServiceUtils { CountDownLatch latch = new CountDownLatch(1); AtomicReference exception = new AtomicReference<>(); executor.onNewClusterState("test setting state", - () -> ClusterState.builder(clusterState).version(clusterState.version() + 1).build(), new ClusterStateTaskListener() { + () -> ClusterState.builder(clusterState).version(clusterState.version() + 1).build(), new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } @@ -163,9 +163,9 @@ public class ClusterServiceUtils { CountDownLatch latch = new CountDownLatch(1); AtomicReference ex = new AtomicReference<>(); clusterApplier.onNewClusterState("mock_publish_to_self[" + event.source() + "]", () -> event.state(), - new ClusterStateTaskListener() { + new ClusterApplyListener() { @Override - public void clusterStateProcessed(String source, ClusterState oldState, ClusterState newState) { + public void onSuccess(String source) { latch.countDown(); } From aafcd85f50e9ae0d996da2a8fc43503f1ba0bcd2 Mon Sep 17 00:00:00 2001 From: David Roberts Date: Thu, 24 May 2018 09:17:17 +0100 Subject: [PATCH 12/12] Move persistent task registrations to core (#30755) Persistent tasks was moved from X-Pack to core in #28455. However, registration of the named writables and named X-content was left in X-Pack. This change moves the registration of the named writables and named X-content into core. Additionally, the persistent task actions are no longer registered in the X-Pack client plugin, as they are already registered in ActionModule. --- .../elasticsearch/cluster/ClusterModule.java | 9 ++++++++ .../PersistentTasksCustomMetaData.java | 1 - .../persistent/TestPersistentTasksPlugin.java | 10 --------- .../xpack/core/XPackClientPlugin.java | 22 +------------------ .../MlNativeAutodetectIntegTestCase.java | 6 ----- 5 files changed, 10 insertions(+), 38 deletions(-) diff --git a/server/src/main/java/org/elasticsearch/cluster/ClusterModule.java b/server/src/main/java/org/elasticsearch/cluster/ClusterModule.java index 9baa47fbc26..7f16c3f85ff 100644 --- a/server/src/main/java/org/elasticsearch/cluster/ClusterModule.java +++ b/server/src/main/java/org/elasticsearch/cluster/ClusterModule.java @@ -69,8 +69,11 @@ import org.elasticsearch.common.settings.Settings; import org.elasticsearch.common.xcontent.NamedXContentRegistry; import org.elasticsearch.gateway.GatewayAllocator; import org.elasticsearch.ingest.IngestMetadata; +import org.elasticsearch.persistent.PersistentTasksCustomMetaData; +import org.elasticsearch.persistent.PersistentTasksNodeService; import org.elasticsearch.plugins.ClusterPlugin; import org.elasticsearch.script.ScriptMetaData; +import org.elasticsearch.tasks.Task; import org.elasticsearch.tasks.TaskResultsService; import java.util.ArrayList; @@ -140,6 +143,10 @@ public class ClusterModule extends AbstractModule { registerMetaDataCustom(entries, IngestMetadata.TYPE, IngestMetadata::new, IngestMetadata::readDiffFrom); registerMetaDataCustom(entries, ScriptMetaData.TYPE, ScriptMetaData::new, ScriptMetaData::readDiffFrom); registerMetaDataCustom(entries, IndexGraveyard.TYPE, IndexGraveyard::new, IndexGraveyard::readDiffFrom); + registerMetaDataCustom(entries, PersistentTasksCustomMetaData.TYPE, PersistentTasksCustomMetaData::new, + PersistentTasksCustomMetaData::readDiffFrom); + // Task Status (not Diffable) + entries.add(new Entry(Task.Status.class, PersistentTasksNodeService.Status.NAME, PersistentTasksNodeService.Status::new)); return entries; } @@ -154,6 +161,8 @@ public class ClusterModule extends AbstractModule { ScriptMetaData::fromXContent)); entries.add(new NamedXContentRegistry.Entry(MetaData.Custom.class, new ParseField(IndexGraveyard.TYPE), IndexGraveyard::fromXContent)); + entries.add(new NamedXContentRegistry.Entry(MetaData.Custom.class, new ParseField(PersistentTasksCustomMetaData.TYPE), + PersistentTasksCustomMetaData::fromXContent)); return entries; } diff --git a/server/src/main/java/org/elasticsearch/persistent/PersistentTasksCustomMetaData.java b/server/src/main/java/org/elasticsearch/persistent/PersistentTasksCustomMetaData.java index 6611ff7f2a3..6d2c21a764a 100644 --- a/server/src/main/java/org/elasticsearch/persistent/PersistentTasksCustomMetaData.java +++ b/server/src/main/java/org/elasticsearch/persistent/PersistentTasksCustomMetaData.java @@ -25,7 +25,6 @@ import org.elasticsearch.cluster.AbstractNamedDiffable; import org.elasticsearch.cluster.ClusterState; import org.elasticsearch.cluster.NamedDiff; import org.elasticsearch.cluster.metadata.MetaData; -import org.elasticsearch.cluster.node.DiscoveryNodes; import org.elasticsearch.common.Nullable; import org.elasticsearch.common.ParseField; import org.elasticsearch.common.Strings; diff --git a/server/src/test/java/org/elasticsearch/persistent/TestPersistentTasksPlugin.java b/server/src/test/java/org/elasticsearch/persistent/TestPersistentTasksPlugin.java index 35c26105672..556d6d1983e 100644 --- a/server/src/test/java/org/elasticsearch/persistent/TestPersistentTasksPlugin.java +++ b/server/src/test/java/org/elasticsearch/persistent/TestPersistentTasksPlugin.java @@ -33,9 +33,7 @@ import org.elasticsearch.action.support.tasks.TransportTasksAction; import org.elasticsearch.client.Client; import org.elasticsearch.client.ElasticsearchClient; import org.elasticsearch.cluster.ClusterState; -import org.elasticsearch.cluster.NamedDiff; import org.elasticsearch.cluster.metadata.IndexNameExpressionResolver; -import org.elasticsearch.cluster.metadata.MetaData; import org.elasticsearch.cluster.node.DiscoveryNode; import org.elasticsearch.cluster.service.ClusterService; import org.elasticsearch.common.ParseField; @@ -100,12 +98,6 @@ public class TestPersistentTasksPlugin extends Plugin implements ActionPlugin, P public List getNamedWriteables() { return Arrays.asList( new NamedWriteableRegistry.Entry(PersistentTaskParams.class, TestPersistentTasksExecutor.NAME, TestParams::new), - new NamedWriteableRegistry.Entry(Task.Status.class, - PersistentTasksNodeService.Status.NAME, PersistentTasksNodeService.Status::new), - new NamedWriteableRegistry.Entry(MetaData.Custom.class, PersistentTasksCustomMetaData.TYPE, - PersistentTasksCustomMetaData::new), - new NamedWriteableRegistry.Entry(NamedDiff.class, PersistentTasksCustomMetaData.TYPE, - PersistentTasksCustomMetaData::readDiffFrom), new NamedWriteableRegistry.Entry(Task.Status.class, TestPersistentTasksExecutor.NAME, Status::new) ); } @@ -113,8 +105,6 @@ public class TestPersistentTasksPlugin extends Plugin implements ActionPlugin, P @Override public List getNamedXContent() { return Arrays.asList( - new NamedXContentRegistry.Entry(MetaData.Custom.class, new ParseField(PersistentTasksCustomMetaData.TYPE), - PersistentTasksCustomMetaData::fromXContent), new NamedXContentRegistry.Entry(PersistentTaskParams.class, new ParseField(TestPersistentTasksExecutor.NAME), TestParams::fromXContent), new NamedXContentRegistry.Entry(Task.Status.class, new ParseField(TestPersistentTasksExecutor.NAME), Status::fromXContent) diff --git a/x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/XPackClientPlugin.java b/x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/XPackClientPlugin.java index c719cdcbd02..0b22cd86fe6 100644 --- a/x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/XPackClientPlugin.java +++ b/x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/XPackClientPlugin.java @@ -91,13 +91,7 @@ import org.elasticsearch.xpack.core.ml.action.ValidateJobConfigAction; import org.elasticsearch.xpack.core.ml.datafeed.DatafeedState; import org.elasticsearch.xpack.core.ml.job.config.JobTaskStatus; import org.elasticsearch.xpack.core.monitoring.MonitoringFeatureSetUsage; -import org.elasticsearch.persistent.CompletionPersistentTaskAction; import org.elasticsearch.persistent.PersistentTaskParams; -import org.elasticsearch.persistent.PersistentTasksCustomMetaData; -import org.elasticsearch.persistent.PersistentTasksNodeService; -import org.elasticsearch.persistent.RemovePersistentTaskAction; -import org.elasticsearch.persistent.StartPersistentTaskAction; -import org.elasticsearch.persistent.UpdatePersistentTaskStatusAction; import org.elasticsearch.xpack.core.rollup.RollupFeatureSetUsage; import org.elasticsearch.xpack.core.rollup.RollupField; import org.elasticsearch.xpack.core.rollup.action.DeleteRollupJobAction; @@ -255,11 +249,6 @@ public class XPackClientPlugin extends Plugin implements ActionPlugin, NetworkPl GetCalendarEventsAction.INSTANCE, PostCalendarEventsAction.INSTANCE, PersistJobAction.INSTANCE, - // licensing - StartPersistentTaskAction.INSTANCE, - UpdatePersistentTaskStatusAction.INSTANCE, - RemovePersistentTaskAction.INSTANCE, - CompletionPersistentTaskAction.INSTANCE, // security ClearRealmCacheAction.INSTANCE, ClearRolesCacheAction.INSTANCE, @@ -324,18 +313,12 @@ public class XPackClientPlugin extends Plugin implements ActionPlugin, NetworkPl // ML - Custom metadata new NamedWriteableRegistry.Entry(MetaData.Custom.class, "ml", MlMetadata::new), new NamedWriteableRegistry.Entry(NamedDiff.class, "ml", MlMetadata.MlMetadataDiff::new), - new NamedWriteableRegistry.Entry(MetaData.Custom.class, PersistentTasksCustomMetaData.TYPE, - PersistentTasksCustomMetaData::new), - new NamedWriteableRegistry.Entry(NamedDiff.class, PersistentTasksCustomMetaData.TYPE, - PersistentTasksCustomMetaData::readDiffFrom), // ML - Persistent action requests new NamedWriteableRegistry.Entry(PersistentTaskParams.class, StartDatafeedAction.TASK_NAME, StartDatafeedAction.DatafeedParams::new), new NamedWriteableRegistry.Entry(PersistentTaskParams.class, OpenJobAction.TASK_NAME, OpenJobAction.JobParams::new), // ML - Task statuses - new NamedWriteableRegistry.Entry(Task.Status.class, PersistentTasksNodeService.Status.NAME, - PersistentTasksNodeService.Status::new), new NamedWriteableRegistry.Entry(Task.Status.class, JobTaskStatus.NAME, JobTaskStatus::new), new NamedWriteableRegistry.Entry(Task.Status.class, DatafeedState.NAME, DatafeedState::fromStream), new NamedWriteableRegistry.Entry(XPackFeatureSet.Usage.class, XPackField.MACHINE_LEARNING, @@ -370,8 +353,6 @@ public class XPackClientPlugin extends Plugin implements ActionPlugin, NetworkPl // ML - Custom metadata new NamedXContentRegistry.Entry(MetaData.Custom.class, new ParseField("ml"), parser -> MlMetadata.METADATA_PARSER.parse(parser, null).build()), - new NamedXContentRegistry.Entry(MetaData.Custom.class, new ParseField(PersistentTasksCustomMetaData.TYPE), - PersistentTasksCustomMetaData::fromXContent), // ML - Persistent action requests new NamedXContentRegistry.Entry(PersistentTaskParams.class, new ParseField(StartDatafeedAction.TASK_NAME), StartDatafeedAction.DatafeedParams::fromXContent), @@ -387,8 +368,7 @@ public class XPackClientPlugin extends Plugin implements ActionPlugin, NetworkPl new NamedXContentRegistry.Entry(MetaData.Custom.class, new ParseField(LicensesMetaData.TYPE), LicensesMetaData::fromXContent), //rollup - new NamedXContentRegistry.Entry(PersistentTaskParams.class, new ParseField(RollupField.TASK_NAME), - parser -> RollupJob.fromXContent(parser)), + new NamedXContentRegistry.Entry(PersistentTaskParams.class, new ParseField(RollupField.TASK_NAME), RollupJob::fromXContent), new NamedXContentRegistry.Entry(Task.Status.class, new ParseField(RollupJobStatus.NAME), RollupJobStatus::fromXContent) ); } diff --git a/x-pack/qa/ml-native-tests/src/test/java/org/elasticsearch/xpack/ml/integration/MlNativeAutodetectIntegTestCase.java b/x-pack/qa/ml-native-tests/src/test/java/org/elasticsearch/xpack/ml/integration/MlNativeAutodetectIntegTestCase.java index f3b29a48aa9..f70efc72506 100644 --- a/x-pack/qa/ml-native-tests/src/test/java/org/elasticsearch/xpack/ml/integration/MlNativeAutodetectIntegTestCase.java +++ b/x-pack/qa/ml-native-tests/src/test/java/org/elasticsearch/xpack/ml/integration/MlNativeAutodetectIntegTestCase.java @@ -26,8 +26,6 @@ import org.elasticsearch.common.xcontent.XContentType; import org.elasticsearch.common.xcontent.json.JsonXContent; import org.elasticsearch.index.query.QueryBuilders; import org.elasticsearch.persistent.PersistentTaskParams; -import org.elasticsearch.persistent.PersistentTasksCustomMetaData; -import org.elasticsearch.persistent.PersistentTasksNodeService; import org.elasticsearch.plugins.Plugin; import org.elasticsearch.search.SearchHit; import org.elasticsearch.search.SearchHits; @@ -447,14 +445,10 @@ abstract class MlNativeAutodetectIntegTestCase extends ESIntegTestCase { List entries = new ArrayList<>(ClusterModule.getNamedWriteables()); entries.addAll(new SearchModule(Settings.EMPTY, true, Collections.emptyList()).getNamedWriteables()); entries.add(new NamedWriteableRegistry.Entry(MetaData.Custom.class, "ml", MlMetadata::new)); - entries.add(new NamedWriteableRegistry.Entry(MetaData.Custom.class, PersistentTasksCustomMetaData.TYPE, - PersistentTasksCustomMetaData::new)); entries.add(new NamedWriteableRegistry.Entry(PersistentTaskParams.class, StartDatafeedAction.TASK_NAME, StartDatafeedAction.DatafeedParams::new)); entries.add(new NamedWriteableRegistry.Entry(PersistentTaskParams.class, OpenJobAction.TASK_NAME, OpenJobAction.JobParams::new)); - entries.add(new NamedWriteableRegistry.Entry(Task.Status.class, PersistentTasksNodeService.Status.NAME, - PersistentTasksNodeService.Status::new)); entries.add(new NamedWriteableRegistry.Entry(Task.Status.class, JobTaskStatus.NAME, JobTaskStatus::new)); entries.add(new NamedWriteableRegistry.Entry(Task.Status.class, DatafeedState.NAME, DatafeedState::fromStream)); entries.add(new NamedWriteableRegistry.Entry(ClusterState.Custom.class, TokenMetaData.TYPE, TokenMetaData::new));