Getting Started

Initial Contributions

Writing a new feature is just one way to contribute to the Cassandra project. In fact, making sure that supporting tasks, such as quality testing, documentation, and helping users are completed is just as important. Tracking the development of new features is an ongoing challenge for this project, like most open source projects. We suggest learning how this project gets things done before tackling a new feature. Here are some suggestions for ways to contribute:

  • Update the documentation

  • Answer questions on the user list

  • Review and test a submitted patch

  • Investigate and fix a reported bug

  • Create unit tests and d-tests

Updating documentation

The Cassandra documentation is maintained in the Cassandra source repository along with the Cassandra code base. To submit changes to the documentation, follow the standard process for submitting a patch.

Answering questions on the user list

Subscribe to the user list, look for some questions you can answer and write a reply. Simple as that! See the community page for details on how to subscribe to the mailing list.

Reviewing and testing a submitted patch

Reviewing patches is not the sole domain of committers. If others review a patch, it can reduce the load on the committers. Less time spent reviewing patches means committers can more great features or review more complex patches. Follow the instructions in How to review or alternatively, create a build with the patch and test it with your own workload. Add a comment to the JIRA ticket to let others know you’ve reviewed and tested, along with the results of your work. For example:

“I tested this performance enhancement on our application’s standard production load test and found a 3% improvement.”

Investigate and/or fix a reported bug

Often, the hardest work in fixing a bug is reproducing it. Even if youdon’t have the knowledge to produce a fix, figuring out a way to reliably reproduce an issue can be a massive contribution. Document your method of reproduction in a JIRA comment or, better yet, produce an automated test that reproduces the issue and attach it to the ticket. If you go as far as producing a fix, follow the process for submitting a patch.

To create a JIRA account, please request it on the #cassandra or #cassandra-dev channels on ASF Slack, or on the user or dev mailing list.

Create unit tests and Dtests

Test coverage for Cassandra will always benefit from more automated test coverage, as with most code bases. Before starting work on a particular area of code, consider reviewing and enhancing the existing test coverage. You’ll both improve your knowledge of the code before you start on an enhancement, and reduce the chance introducing issues with your change. See testing and patches for more detail.

Building and IDE Integration

Prerequisites

Building Cassandra from source is the first important step in contributing to the Apache Cassandra project. You will need to install Java (JDK 8 or 11, depending on which version you want to build Cassandra against, best is to install both and then switch between them when needed), Git, and Ant (including ant-optional).

Getting the source code

The source code for Cassandra is shared on the central Apache Git repository and organized by branch, one branch for each major version. You can access the code for the current development branch at gitbox.apache.org/repos/asf/cassandra.git (the primary location) or at github.com/apache/cassandra (a mirror location).

However, to develop a patch or a new feature, you should fork the Cassandra project first and then clone it from your own fork:

  1. git clone https://github.com/<your_git_name>/cassandra.git cassandra

Building Cassandra

Cassandra is a Java project which is built using Ant. The build file, build.xml located in the root of the project content, has various tasks defined (you can list all of them with some short descriptions by running ant -p).

The build uses the Java compiler which is currently set up in your shell. By default, the build uses Java 8. If you want to build with Java 11, you need to either add a build property -Duse.jdk11=true to your Ant build command or export the environment variable CASSANDRA_USE_JDK11=true. Otherwise, if you run the build using Java 11, the build script complains.

Now you can build Cassandra with the default task - just execute ant or ant jar. This may take a significant amount of time depending on artifacts that have to be downloaded or the number of classes that need to be compiled. When the build completes, you can find a JAR file in the build directory, and the database service can be started with the bin/cassandra script.

Building a distribution

Some tasks you might be interested in are:

  • build - compile the production code without producing any JARs

  • build-test - compile the test code without producing any JARs

  • artifacts - generate Cassandra distribution in build/dist directory and package it to tar.gz (with and without sources)

  • mvn-install - generate cassandra-all JAR artifact along with sources and JavaDoc, and install it in the local Maven repository

  • realclean - clean the project from any build products, as well as from any dependencies (in fact that means removing build and lib directories)

Hint

Remember to clean the project after switching branches as build artifacts are versioned.

There are other tasks related to testing, and they are covered in Testing section.

Setting up Cassandra in IntelliJ IDEA

IntelliJ IDEA by JetBrains is one of the most popular IDEs for Cassandra and Java development in general. The Community Edition can be freely downloaded with all features needed to get started developing Cassandra.

Use the following procedure for Cassandra 2.1.5+:

  1. Generate project files by executing the following target from Ant build:
  1. ant generate-idea-files
  1. Start IDEA.

  2. Open the IDEA project from the checked-out Cassandra directory using File > Open in IDEA’s menu.

The project generated by ant generate-idea-files contains nearly everything you need to debug Cassandra and execute unit tests.

Hint

Although you do not need to build the project with ant jar to import it some tests run tools as external processes and need the regular artifacts to be created. Such tests may require you to rebuild the project using Ant before each run even when you run the test in IDEA.

You should be able to:

  • Run/debug defaults for JUnit

  • Run/debug configuration for Cassandra daemon

  • Read/modify the license header for Java source files

  • Study Cassandra code style

  • Inspections

Hint

If you wish to work with older Cassandra versions, see our wiki for instructions.

Opening Cassandra in Apache NetBeans

Apache NetBeans is the elder of the open sourced java IDEs, and can be used for Cassandra development. There is no project setup or generation required to open Cassandra in NetBeans. Use the following procedure for Cassandra 4.0+.

First, clone and build Cassandra. Then execute the following steps to use NetBeans.

  1. Start Apache NetBeans

  2. Open the NetBeans project from the ide/ folder of the checked-out Cassandra directory using File > Open Project in NetBeans’ menu.

You should be able to:

  • Build code

  • Run code

  • Debug code

  • Profile code

These capabilities use the build.xml script. Build/Run/Debug Project are available via the Run/Debug menus, or the project context menu. Profile Project is available via the Profile menu. In the opened Profiler tab, click the green “Profile” button. Cassandra’s code style is honored in ide/nbproject/project.properties. The JAVA8_HOME system environment variable must be set for NetBeans to execute the Run/Debug/Profile ant targets to execute.

Setting up Cassandra in Eclipse

Eclipse is a popular open source IDE that can be used for Cassandra development. Various Eclipse environments are available from the download page. The following guide was created with “Eclipse IDE for Java Developers”.

These instructions were tested on Ubuntu 16.04 with Eclipse Neon (4.6) using Cassandra versions 2.1 through 3.x.

First, clone and build Cassandra. Then execute the following steps to use Eclipse.

  1. Generate the IDEA files using ant:
  1. ant generate-eclipse-files
  1. Start Eclipse.

  2. Open the Eclipse project from the checked-out Cassandra directory using File > Import > Existing Projects and Workspace > Select git directory. Select the correct branch, such as cassandra-trunk.

  3. Confirm and select Finish to import your project.

Find the project in Package Explorer or Project Explorer. You should not get errors if you build the project automatically using these instructions. Don’t set up the project before generating the files with ant.

You should be able to:

  • Run/debug defaults for JUnit

  • Run/debug Cassandra

  • Study Cassandra code style

Unit tests can be run from Eclipse by simply right-clicking the class file or method and selecting Run As > JUnit Test. Tests can be debugged by defining breakpoints (double-click line number) and selecting Debug As > JUnit Test.

Alternatively all unit tests can be run from the command line as described in testing.

Debugging Cassandra Using Eclipse

There are two ways to start a local Cassandra instance with Eclipse for debugging. You can either start Cassandra from the command line or from within Eclipse.

Debugging Cassandra started at command line
  1. Set environment variable to define remote debugging options for the JVM: export JVM_EXTRA_OPTS="-agentlib:jdwp=transport=dt_socket,server=y,suspend=n,address=1414"

  2. Start Cassandra by executing the ./bin/cassandra

Next, connect to the running Cassandra process by:

  1. In Eclipse, select Run > Debug Configurations.

image

  1. Create new remote application.

image

  1. Configure connection settings by specifying a name and port 1414. Confirm Debug and start debugging.

image

Debugging Cassandra started from Eclipse

Cassandra can also be started directly from Eclipse if you don’t want to use the command line.

  1. In Eclipse, select Run > Run Configurations.

image

  1. Create new application.

image

  1. Specify name, project and main class org.apache.cassandra.service.CassandraDaemon

image

  1. Configure additional JVM specific parameters that will start Cassandra with some of the settings created by the regular startup script. Change heap related values as needed.
  1. -Xms1024M -Xmx1024M -Xmn220M -Xss256k -ea -XX:+UseThreadPriorities -XX:ThreadPriorityPolicy=42 -XX:+UseParNewGC -XX:+UseConcMarkSweepGC -XX:+CMSParallelRemarkEnabled -XX:+UseCondCardMark -javaagent:./lib/jamm-0.3.0.jar -Djava.net.preferIPv4Stack=true

image

  1. Confirm Debug and you should see the output of Cassandra start up in the Eclipse console.

You can now set breakpoints and start debugging!

General notes

You may sometimes encounter some odd build failures when running the ant commands above. If you do, start ant with the realclean option:

  1. ant realclean

Remember that all the tasks mentioned above may depend on building source files. If there are actual compilation errors in the code, you may not be able to generate project files for IntelliJ Idea, Netbeans, or Eclipse. It is especially important that you have imported the project adequately into IDE before doing merges or rebases. Otherwise, if there are conflicts, the project cannot be opened in IDE, and you will be unable to use any fancy conflict resolution tools offered by those IDEs.

Testing

Creating tests is one of the most important and also most difficult parts of developing Cassandra. There are different ways to test your code depending on what you’re working on.

Cassandra tests can be divided into three main categories, based on the way how they are executed:

  • Java tests - tests implemented in Java and being a part of the Cassandra project. You can distinguish the following subcategories there:

    • JUnit tests - consists of unit tests, single-node integration tests and some tool tests; those tests may run a server with limited functionality in the same JVM as the test code

    • JVM distributed tests - integrated tests against one or multiple nodes, each running in their own classloader; also contains upgrade tests

    • Micro-benchmarks - micro-benchmarks implemented with JMH framework

  • CQLSH tests - CQLSH tests are Python tests written with the Nose test framework. They verify the CQLSH client that can be found in the bin directory. They aim at verifying CQLSH specific behavior like output formatting, autocompletion, parsing, etc).

  • Python distributed tests - Python distributed tests are implemented on top of the PyTest framework and located outside the main Cassandra project in the separate repository apache/cassandra-dtest. They test Cassandra via CCM verifying operation results, logs, and cluster state. Python Distributed tests are Cassandra version agnostic. They include upgrade tests.

In case you want to run DTests with your own version of CCM, please refer to requirements.txt in apache/cassandra-dtest how to do it.

The recipes for running those tests can be found in the cassandra-builds repository here.

Running full test suites locally takes hours, if not days. Beyond running specific tests you know are applicable, or are failing, to the work at hand, it is recommended to rely upon the project’s Continuous Integration systems. If you are not a committer, and don’t have access to a premium CircleCI plan, ask one of the committers to test your patch on the project’s ci-cassandra.apache.org.

Java tests

The simplest test to write for Cassandra code is a unit test. Cassandra uses JUnit as a testing framework and test cases can be found in the test/unit directory. Ideally, you’ll create a unit test for your implementation that exclusively covers the class you created (the unit under test).

Unfortunately, this is not always possible, because Cassandra doesn’t have a very mock friendly code base. Often you’ll find yourself in a situation where you have to use the embedded Cassandra instance to interact with your test. If you want to use CQL in your test, you can extend CQLTester and use some convenient helper methods, as shown here:

  1. @Test
  2. public void testBatchAndList() throws Throwable
  3. {
  4. createTable("CREATE TABLE %s (k int PRIMARY KEY, l list<int>)");
  5. execute("BEGIN BATCH " +
  6. "UPDATE %1$s SET l = l +[ 1 ] WHERE k = 0; " +
  7. "UPDATE %1$s SET l = l + [ 2 ] WHERE k = 0; " +
  8. "UPDATE %1$s SET l = l + [ 3 ] WHERE k = 0; " +
  9. "APPLY BATCH");
  10. assertRows(execute("SELECT l FROM %s WHERE k = 0"),
  11. row(list(1, 2, 3)));
  12. }

JUnit tests

To run the unit tests:

  1. ant test

However, this is probably not what you want to do, since that command would run all the unit tests (those from test/unit). It would take about an hour or more to finish.

To run the specific test class or even a method, use the following command:

  1. ant testsome -Dtest.name=<TestClassName> -Dtest.methods=<testMethodName>
  • test.name property is for either a simple or fully qualified class name

  • test.methods property is optional; if not specified, all test cases from the specified class are executed. Though, you can also specify multiple methods separating them by comma

You can also use the IDE to run the tests - when you generate IDE files and properly import the Cassandra project, you can run the tests by right-clicking on the test class or package name. Remember that it is not enough to compile with IDE for some tests, and you need to call ant jar to build the distribution artifacts. When the test runs some tool as an external process, the tool expects Cassandra artifacts to be in the build directory.

Note that those commands apply to the tests in the test/unit directory. There are, however, some other test categories that have tests in individual directories:

  • test/burn - to run them, call ant test-burn or ant burn-testsome; ant burn-test-jar builds a self-contained jar for e.g. remote execution; not currently used for running burn tests in our scripts. ant burn-test-jar exists only on 4.0+ branches

  • test/long - to run them, call ant long-test or ant long-testsome

  • test/memory - to run them, call ant test-memory

  • test/microbench discussed in Micro-benchmarks

  • test/distributed discussed in JVM distributed tests

Hint

If you get the error similar to the one below, install the ant-optional package because you need the JUnitTask class (see prerequisites).

  1. Throws: cassandra-trunk/build.xml:1134: taskdef A class needed by class org.krummas.junit.JStackJUnitTask cannot be found:
  2. org/apache/tools/ant/taskdefs/optional/junit/JUnitTask using the classloader
  3. AntClassLoader[/…/cassandra-trunk/lib/jstackjunit-0.0.1.jar]

Stress and FQLTool tests

Stress and FQLTool are separate modules located under the tools directory in the Cassandra project. They have their own source code and unit tests. To run the tests for those tools, first, build jar artifacts for them but calling:

  1. ant fqltool-build fqltool-build-test
  2. ant stress-build stress-build-test

Then you can execute the tests with either one of the commands:

  1. ant fqltool-test
  2. ant stress-test
  3. and stress-test-some

or using your IDE.

JVM distributed tests

JVM distributed tests can run a cluster of nodes inside a single JVM - they utilize a particular framework (that can be found at apache/cassandra-in-jvm-dtest-api) for that purpose. Those tests are intended to test features that require more started nodes or verify specific behaviors when the nodes get restarted, including upgrading them from one version to another. The tests are located at the test/distributed directory of the Cassandra project; however, only org.apache.cassandra.distributed.test and org.apache.cassandra.upgrade packages contain the actual tests. The rest of the files are various utilities related to the distributed test framework.

The distributed tests can be run in few ways. ant test-jvm-dtest command runs all the distributed JVM tests. It is not very useful; thus, there is also ant test-jvm-dtest-some, which allows specifying test class and test name in the similar way as you could do that for the ant testsome command, for example:

  1. ant test-jvm-dtest-some -Dtest.name=org.apache.cassandra.distributed.test.SchemaTest
  2. ant test-jvm-dtest-some -Dtest.name=org.apache.cassandra.distributed.test.SchemaTest -Dtest.methods=readRepair
Hint

Unlike for JUnit tests, for JVM distributed tests you need to provide fully qualified class name

Distributed tests can also be run using IDE (in fact, you can even debug them).

Upgrade tests

JVM upgrade tests can be run precisely in the same way as any other JVM distributed tests. However, running them requires some preparation - for example, if a test verifies the upgrade from Cassandra 3.0 and Cassandra 3.11 to the current version (say Cassandra 4.0), you need to have prepared dtest uber JARs for all involved versions. To do this:

  1. Check out Cassandra 3.0 based branch you want to test the upgrade from into some other directory

  2. Build dtest uber JAR with ant dtest-jar command

  3. Copy the created build/dtest-3.0.x.jar to the build directory of your target Cassandra project

  4. Repeat the procedure for Cassandra 3.11

  5. Once you have dtest jars of all the involved versions for the upgrade test, you can finally execute the test using your favorite method, say:

  1. ant test-jvm-dtest-some -Dtest.name=org.apache.cassandra.distributed.upgrade.MixedModeReadTest
Hint

You may pre-generate dtest uber JARs for certain past Cassandra releases, store is somewhere and reuse in you future work - no need to rebuild them all the time.

Running multiple tests

It is possible to define a list of test classes to run with a single command. Define a text file, by default called testlist.txt, and put it into your project directory. Here is an example of that file:

  1. org/apache/cassandra/db/ReadCommandTest.java
  2. org/apache/cassandra/db/ReadCommandVerbHandlerTest.java

Essentially, you list the paths to the class files of the tests you want to run. Then you call ant testclasslist, which uses the text file to run the listed tests. Note that, by default, it applies to the tests under the test/unit directory and takes the testlist.txt file, but this behavior can be modified by providing additional parameters:

  1. ant testclasslist -Dtest.classlistprefix=<category> -Dtest.classlistfile=<class list file>

For example, if you want to run the distributed tests this way, and say our tests were listed in the distributed-tests-set.txt file (paths to test classes relative to test/distributed directory), you can do that by calling:

  1. ant testclasslist -Dtest.classlistprefix=distributed -Dtest.classlistfile=distributed-tests-set.txt

Running coverage analysis

Coverage reports from the executed JVM tests can be obtained in two ways - through IDE - for example, IntelliJ supports running tests with coverage analysis (another run button next to the one for running in debug mode).

The other way is to run Ant target codecoverage. Basically, it works for all the ways mentioned above of running JVM tests - the only difference is that instead of specifying the target directly, you pass it as a property called taskname. For example - given the original test command is:

  1. ant testsome -Dtest.name=org.apache.cassandra.utils.concurrent.AccumulatorTest

to run it with coverage analysis, do:

  1. ant codecoverage -Dtaskname=testsome -Dtest.name=org.apache.cassandra.utils.concurrent.AccumulatorTest

It applies to all the targets like test, testsome, test-long, etc., even testclasslist. You can find the coverage report in build/jacoco (index.html is the entry point for the HTML version, but there are also XML and CSV reports).

Note that if you run various tests that way, the coverage information is added to the previously collected runs. That is, you get the cumulative coverage from all runs unless you clean up the project or at least clean up the recorded coverage information by executing the command ant jacoco-cleanup.

Micro-benchmarks

To run micro-benchmarks, first build the uber jar for the JMH framework. Use the following ant command:

  1. ant build-jmh

Then, you can run either all benchmarks (from the test/microbench directory) or the tests matching the name specified by the benchmark.name property when executing the ant microbench command. Whether you run all benchmarks or just a selected one, only classes under the microbench package are selected. The class selection pattern is actually .*microbench.*${benchmark.name}. For example, in order to run org.apache.cassandra.test.microbench.ChecksumBench, execute:

  1. ant microbench -Dbenchmark.name=ChecksumBench

The ant microbench command runs the benchmarks with default parameters as defined in the build.xml file (see the microbench target definition). If you want to run JMH with custom parameters, consider using the test/bin/jmh script. In addition to allowing you to customize JMH options, it also sets up the environment and JVM options by running Cassandra init script (conf/cassandra-env.sh). Therefore, it lets the environment for running the tests to be more similar to the production environment. For example:

  1. test/bin/jmh -gc true org.apache.cassandra.test.microbench.CompactionBench.compactTest

You may also find it useful to run the command to list all the tests: test/bin/jmh -l or test/bin/jmh -lp (also showing the default parameters). The list of all options can be shown by running test/bin/jmh -h

Python tests

Docker

The Docker approach is recommended for running Python distributed tests. The behavior will be more repeatable, matching the same environment as the official testing on Cassandra CI.

Setup Docker

If you are on Linux, you need to install Docker using the system package manager.

If you are on MacOS, you can use either Docker Desktop or some other approach.

Pull the Docker image

The Docker image used on the official Cassandra CI can be found in this repository. You can use either docker/testing/ubuntu2004_j11.docker or docker/testing/ubuntu2004_j11_w_dependencies.docker The second choice has prefetched dependencies for building each main Cassandra branch. Those images can be either built locally (as per instructions in the GitHub repo) or pulled from the Docker Hub - see here.

First, pull the image from Docker Hub (it will either fetch or update the image you previously fetched):

  1. docker pull apache/cassandra-testing-ubuntu2004-java11-w-dependencies

Start the container

  1. docker run -di -m 8G --cpus 4 \
  2. --mount type=bind,source=/path/to/cassandra/project,target=/home/cassandra/cassandra \
  3. --mount type=bind,source=/path/to/cassandra-dtest,target=/home/cassandra/cassandra-dtest \
  4. --name test \
  5. apache/cassandra-testing-ubuntu2004-java11-w-dependencies \
  6. dumb-init bash
Hint

Many distributed tests are not that demanding in terms of resources - 4G / 2 cores should be enough to start one node. However, some tests really run multiple nodes, and some of them are automatically skipped if the machine has less than 32G (there is a way to force running them though). Usually 8G / 4 cores is a convenient choice which is enough for most of the tests.

To log into the container, use the following docker exec command:

  1. docker exec -it `docker container ls -f name=test -q` bash

Setup Python environment

The tests are implemented in Python, so a Python virtual environment (see here for details) with all the required dependencies is good to be set up. If you are familiar with the Python ecosystem, you know what it is all about. Otherwise, follow the instructions; it should be enough to run the tests.

For Python distributed tests do:

  1. cd /home/cassandra/cassandra-dtest
  2. virtualenv --python=python3 --clear --always-copy ../dtest-venv
  3. source ../dtest-venv/bin/activate
  4. CASS_DRIVER_NO_CYTHON=1 pip install -r requirements.txt

For CQLSH tests, replace some paths:

  1. cd /home/cassandra/cassandra/pylib
  2. virtualenv --python=python3 --clear --always-copy ../../cqlsh-venv
  3. source ../../cqlsh-venv/bin/activate
  4. CASS_DRIVER_NO_CYTHON=1 pip install -r requirements.txt
Hint

You may wonder why this weird environment variable CASS_DRIVER_NO_CYTHON=1 was added - it is not required at all. Still, it allows avoiding the compilation of Cassandra driver with Cython, which is not needed unless you want to test that Cython compiled driver. In the end, it speeds up the installation of the requirements significantly from the order of minutes to the order of seconds.

The above commands are also helpful for importing those test projects into your IDE. In that case, you need to run them on your host system rather than in Docker container. For example, when you open the project in IntelliJ, the Python plugin may ask you to select the runtime environment. In this case, choose the existing virtualenv based environment and point to bin/python under the created dtest-venv directory (or cqlsh-venv, or whichever name you have chosen).

Whether you want to play with Python distributed tests or CQLSH tests, you need to select the right virtual environment. Remember to switch to the one you want:

  1. deactivate
  2. source /home/cassandra/dtest-venv/bin/activate

or

  1. deactivate
  2. source /home/cassandra/cqlsh-venv/bin/activate

CQLSH tests

CQLSH tests are located in the pylib/cqlshlib/test directory. They are based on the Nose framework. They require a running Cassandra cluster (it can be one or more nodes cluster) as they start a CQL shell client which tries to connect to a live node. Each test case starts the CQLSH client as a subprocess, issues some commands, and verifies the outcome returned by CQLSH to the console.

You need to set up and activate the virtualenv for CQLSH tests (see Setup Python environment section for details).

So let’s start the cluster first - you can use CCM for that (note that CCM gets automatically installed with the virtualenv and is immediately available once the virtualenv is activated):

  1. ccm create test -n 1 --install-dir=/home/cassandra/cassandra
  2. ccm updateconf "enable_user_defined_functions: true"
  3. ccm updateconf "enable_scripted_user_defined_functions: true"
  4. ccm updateconf "cdc_enabled: true"
  5. ccm start --wait-for-binary-proto

When those commands complete successfully, there is a cluster up and running, and you can run the CQLSH tests. To do so, go to the pylib/cqlshlib directory (not to the test subdirectory) and call the nosetests command without any arguments. The tests take around 5 minutes to complete.

Finally, remember that since you manually started the cluster, you need to stop it manually - just call:

  1. ccm remove test

There is a helper script that does all of those things for you. In particular, it builds the Cassandra project, creates a virtual environment, runs the CCM cluster, executes the tests, and eventually removes the cluster. You find the script in the pylib directory. The only argument it requires is the Cassandra project directory:

  1. cassandra@b69a382da7cd:~/cassandra/pylib$ ./cassandra-cqlsh-tests.sh /home/cassandra/cassandra

As you noticed, if you have already built Cassandra, the previous method of running tests is much faster. Just remember that all the ccm updateconf calls must be aligned with the Cassandra version you are testing, with the supported features enabled. Otherwise, Cassandra won’t start.

Running selected tests

You may run all test tests from the selected file by passing that file as an argument:

  1. ~/cassandra/pylib/cqlshlib$ nosetests test/test_constants.py

To run a specific test case, you need to specify the module, class name, and the test name, for example:

  1. ~/cassandra/pylib/cqlshlib$ nosetests cqlshlib.test.test_cqlsh_output:TestCqlshOutput.test_boolean_output

For more information on selecting tests with the Nose framework, see this page.

Python distributed tests

One way of doing integration or system testing at larger scale is using dtest (Cassandra distributed test). These dtests automatically setup Cassandra clusters with certain configurations and simulate use cases you want to test.

The best way to learn how to write dtests is probably by reading the introduction “http://www.datastax.com/dev/blog/how-to-write-a-dtest\[How to Write a Dtest]“. Looking at existing, recently updated tests in the project is another good activity. New tests must follow certain style conventions that are checked before contributions are accepted. In contrast to Cassandra, dtest issues and pull requests are managed on github, therefore you should make sure to link any created dtests in your Cassandra ticket and also refer to the ticket number in your dtest PR.

Creating a good dtest can be tough, but it should not prevent you from submitting patches! Please ask in the corresponding JIRA ticket how to write a good dtest for the patch. In most cases a reviewer or committer will able to support you, and in some cases they may offer to write a dtest for you.

Run the tests - quick examples

Note that you need to set up and activate the virtualenv for DTests (see Setup Python environment section for details). Tests are implemented with the PyTest framework, so you use the pytest command to run them. Let’s run some tests:

  1. pytest --cassandra-dir=/home/cassandra/cassandra schema_metadata_test.py::TestSchemaMetadata::test_clustering_order

That command runs the test_clustering_order test case from TestSchemaMetadata class, located in the schema_metadata_test.py file. You may also provide the file and class to run all test cases from that class:

  1. pytest --cassandra-dir=/home/cassandra/cassandra schema_metadata_test.py::TestSchemaMetadata

or just the file name to run all test cases from all classes defined in that file.

  1. pytest --cassandra-dir=/home/cassandra/cassandra schema_metadata_test.py

You may also specify more individual targets:

  1. pytest --cassandra-dir=/home/cassandra/cassandra schema_metadata_test.py::TestSchemaMetadata::test_basic_table_datatype schema_metadata_test.py::TestSchemaMetadata::test_udf

If you run pytest without specifying any test, it considers running all the tests it can find. More on the test selection here You probably noticed that --cassandra-dir=/home/cassandra/cassandra is constantly added to the command line. It is one of the cassandra-dtest custom arguments - the mandatory one - unless it is defined, you cannot run any Cassandra dtest.

Setting up PyTest

All the possible options can be listed by invoking pytest --help. You see tons of possible parameters - some of them are native PyTest options, and some come from Cassandra DTest. When you look carefully at the help note, you notice that some commonly used options, usually fixed for all the invocations, can be put into the pytest.ini file. In particular, it is quite practical to define the following:

  1. cassandra_dir = /home/cassandra/cassandra
  2. log_cli = True
  3. log_cli_level = DEBUG

so that you do not have to provide --cassandra-dir param each time you run a test. The other two options set up console logging - remove them if you want logs stored only in log files.

Running tests with specific configuration

There are a couple of options to enforce exact test configuration (their names are quite self-explanatory):

  • --use-vnodes

  • --num-token=xxx - enables the support of virtual nodes with a certain number of tokens

  • --use-off-heap-memtables - use off-heap memtables instead of the default heap-based

  • `—data-dir-count-per-instance=xxx - the number of data directories configured per each instance

Note that the list can grow in the future as new predefined configurations can be added to dtests. It is also possible to pass extra Java properties to each Cassandra node started by the tests - define those options in the JVM_EXTRA_OPTS environment variable before running the test.

Listing the tests

You can do a dry run, so that the tests are only listed and not invoked. To do that, add --collect-only to the pytest command. That additional -q option will print the results in the same format as you would pass the test name to the pytest command:

  1. pytest --collect-only -q

lists all the tests pytest would run if no particular test is specified. Similarly, to list test cases in some class, do:

  1. $ pytest --collect-only -q schema_metadata_test.py::TestSchemaMetadata
  2. schema_metadata_test.py::TestSchemaMetadata::test_creating_and_dropping_keyspace
  3. schema_metadata_test.py::TestSchemaMetadata::test_creating_and_dropping_table
  4. schema_metadata_test.py::TestSchemaMetadata::test_creating_and_dropping_table_with_2ary_indexes
  5. schema_metadata_test.py::TestSchemaMetadata::test_creating_and_dropping_user_types
  6. schema_metadata_test.py::TestSchemaMetadata::test_creating_and_dropping_udf
  7. schema_metadata_test.py::TestSchemaMetadata::test_creating_and_dropping_uda
  8. schema_metadata_test.py::TestSchemaMetadata::test_basic_table_datatype
  9. schema_metadata_test.py::TestSchemaMetadata::test_collection_table_datatype
  10. schema_metadata_test.py::TestSchemaMetadata::test_clustering_order
  11. schema_metadata_test.py::TestSchemaMetadata::test_compact_storage
  12. schema_metadata_test.py::TestSchemaMetadata::test_compact_storage_composite
  13. schema_metadata_test.py::TestSchemaMetadata::test_nondefault_table_settings
  14. schema_metadata_test.py::TestSchemaMetadata::test_indexes
  15. schema_metadata_test.py::TestSchemaMetadata::test_durable_writes
  16. schema_metadata_test.py::TestSchemaMetadata::test_static_column
  17. schema_metadata_test.py::TestSchemaMetadata::test_udt_table
  18. schema_metadata_test.py::TestSchemaMetadata::test_udf
  19. schema_metadata_test.py::TestSchemaMetadata::test_uda

You can copy/paste the selected test case to the pytest command to run it.

Filtering tests

Based on configuration

Most tests run with any configuration, but a subset of tests (test cases) only run if a specific configuration is used. In particular, there are tests annotated with:

  • @pytest.mark.vnodes - the test is only invoked when the support of virtual nodes is enabled

  • @pytest.mark.no_vnodes - the test is only invoked when the support of virtual nodes is disabled

  • @pytest.mark.no_offheap_memtables - the test is only invoked if off-heap memtables are not used

Note that enabling or disabling vnodes is obviously mutually exclusive. If a test is marked to run only with vnodes, it does not run when vnodes is disabled; similarly, when a test is marked to run only without vnodes, it does not run when vnodes is enabled - therefore, there are always some tests which would not run with a single configuration.

Based on resource usage

There are also tests marked with:

@pytest.mark.resource_intensive

which means that the test requires more resources than a regular test because it usually starts a cluster of several nodes. The meaning of resource-intensive is hardcoded to 32GB of available memory, and unless your machine or docker container has at least that amount of RAM, such test is skipped. There are a couple of arguments that allow for some control of that automatic exclusion:

  • --force-resource-intensive-tests - forces the execution of tests marked as resource_intensive, regardless of whether there is enough memory available or not

  • --only-resource-intensive-tests - only run tests marked as resource_intensive - it makes all the tests without resource_intensive annotation to be filtered out; technically, it is equivalent to passing native PyTest argument: -m resource_intensive

  • --skip-resource-intensive-tests - skip all tests marked as resource_intensive - it is the opposite argument to the previous one, and it is equivalent to the PyTest native argument: -m 'not resource_intensive'

Based on the test type

Upgrade tests are marked with:

@pytest.mark.upgrade_test

Those tests are not invoked by default at all (just like running PyTest with -m 'not upgrade_test'), and you have to add some extra options to run them: * --execute-upgrade-tests - enables execution of upgrade tests along with other tests - when this option is added, the upgrade tests are not filtered out * --execute-upgrade-tests-only - execute only upgrade tests and filter out all other tests which do not have @pytest.mark.upgrade_test annotation (just like running PyTest with -m 'upgrade_test')

Filtering examples

It does not matter whether you want to invoke individual tests or all tests or whether you only want to list them; the above filtering rules apply. So by using --collect-only option, you can learn which tests would be invoked.

To list all the applicable tests for the current configuration, use the following command:

  1. pytest --collect-only -q --execute-upgrade-tests --force-resource-intensive-tests

List tests specific to vnodes (which would only run if vnodes are enabled):

  1. pytest --collect-only -q --execute-upgrade-tests --force-resource-intensive-tests --use-vnodes -m vnodes

List tests that are not resource-intensive

  1. pytest --collect-only -q --execute-upgrade-tests --skip-resource-intensive-tests

Upgrade tests

Upgrade tests always involve more than one product version. There are two kinds of upgrade tests regarding the product versions they span - let’s call them fixed and generated.

In case of fixed tests, the origin and target versions are hardcoded. They look pretty usual, for example:

  1. pytest --collect-only -q --execute-upgrade-tests --execute-upgrade-tests-only upgrade_tests/upgrade_supercolumns_test.py

prints:

  1. upgrade_tests/upgrade_supercolumns_test.py::TestSCUpgrade::test_upgrade_super_columns_through_all_versions
  2. upgrade_tests/upgrade_supercolumns_test.py::TestSCUpgrade::test_upgrade_super_columns_through_limited_versions

When you look into the code, you will see the fixed upgrade path:

  1. def test_upgrade_super_columns_through_all_versions(self):
  2. self._upgrade_super_columns_through_versions_test(upgrade_path=[indev_2_2_x, indev_3_0_x, indev_3_11_x, indev_trunk])

The generated upgrade tests are listed several times - the first occurrence of the test case is a generic test definition, and then it is repeated many times in generated test classes. For example:

  1. pytest --cassandra-dir=/home/cassandra/cassandra --collect-only -q --execute-upgrade-tests --execute-upgrade-tests-only upgrade_tests/cql_tests.py -k test_set

prints:

  1. upgrade_tests/cql_tests.py::cls::test_set
  2. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_current_2_2_x_To_indev_2_2_x::test_set
  3. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_current_3_0_x_To_indev_3_0_x::test_set
  4. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_current_3_11_x_To_indev_3_11_x::test_set
  5. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_current_4_0_x_To_indev_4_0_x::test_set
  6. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_indev_2_2_x_To_indev_3_0_x::test_set
  7. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_indev_2_2_x_To_indev_3_11_x::test_set
  8. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_indev_3_0_x_To_indev_3_11_x::test_set
  9. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_indev_3_0_x_To_indev_4_0_x::test_set
  10. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_indev_3_11_x_To_indev_4_0_x::test_set
  11. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_indev_4_0_x_To_indev_trunk::test_set
  12. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_current_2_2_x_To_indev_2_2_x::test_set
  13. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_current_3_0_x_To_indev_3_0_x::test_set
  14. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_current_3_11_x_To_indev_3_11_x::test_set
  15. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_current_4_0_x_To_indev_4_0_x::test_set
  16. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_indev_2_2_x_To_indev_3_0_x::test_set
  17. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_indev_2_2_x_To_indev_3_11_x::test_set
  18. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_indev_3_0_x_To_indev_3_11_x::test_set
  19. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_indev_3_0_x_To_indev_4_0_x::test_set
  20. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_indev_3_11_x_To_indev_4_0_x::test_set
  21. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_indev_4_0_x_To_indev_trunk::test_set

In this example, the test case name is just test_set, and the class name is TestCQL - the suffix of the class name is automatically generated from the provided specification. The first component is the cluster specification - there are two variants: Nodes2RF1 and Nodes3RF3 - they denote that the upgrade is tested on 2 nodes cluster with a keyspace using replication factor = 1. Analogously the second variant uses 3 nodes cluster with RF = 3.

Then, there is the upgrade specification - for example, Upgrade_indev_3_11_x_To_indev_4_0_x - which means that this test upgrades from the development version of Cassandra 3.11 to the development version of Cassandra 4.0 - the meaning of indev/current and where they are defined is explained later.

When you look into the implementation, you notice that such upgrade test classes inherit from UpgradeTester class, and they have the specifications defined at the end of the file. In this particular case, it is something like:

  1. topology_specs = [
  2. {'NODES': 3,
  3. 'RF': 3,
  4. 'CL': ConsistencyLevel.ALL},
  5. {'NODES': 2,
  6. 'RF': 1},
  7. ]
  8. specs = [dict(s, UPGRADE_PATH=p, __test__=True)
  9. for s, p in itertools.product(topology_specs, build_upgrade_pairs())]

As you can see, there is a list of the cluster specifications and the cross product is calculated with upgrade paths returned by the build_upgrade_pairs() function. That list of specifications is used to dynamically generate upgrade tests.

Suppose you need to test something specifically for your scenario. In that case, you can add more cluster specifications, like a test with 1 node or a test with 5 nodes with some different replication factor or consistency level. The build_upgrade_pairs() returns the list of upgrade paths (actually just the origin and target version). That list is generated according to the upgrade manifest.

Upgrade manifest

The upgrade manifest is a file where all the upgrade paths are defined. It is a regular Python file located at upgrade_tests/upgrade_manifest.py. As you noticed, Cassandra origin and target version descriptions mentioned in the upgrade test consist of indev or current prefix followed by version string. The definitions of each such version description can be found in the manifest, for example:

  1. indev_3_11_x = VersionMeta(name='indev_3_11_x', family=CASSANDRA_3_11, variant='indev', version='github:apache/cassandra-3.11', min_proto_v=3, max_proto_v=4, java_versions=(8,))
  2. current_3_11_x = VersionMeta(name='current_3_11_x', family=CASSANDRA_3_11, variant='current', version='3.11.10', min_proto_v=3, max_proto_v=4, java_versions=(8,))

There are a couple of different properties which describe those two versions:

  • name - is a name as you can see in the names of the generated test classes

  • family - families is an enumeration defined in the beginning of the upgrade manifest - say family CASSANDRA_3_11 is just a string "3.11". Some major features were introduced or removed with new version families, and therefore some checks can be done or some features can be enabled/disabled according to that, for example:

  1. if self.cluster.version() < CASSANDRA_4_0:
  2. node1.nodetool("enablethrift")

But it is also used to determine whether our checked-out version matches the target version in the upgrade pair (more on that later)

  • variant and version - there are indev or current variants:

    • indev variant means that the development version of Cassandra will be used. That is, that version is checked out from the Git repository and built before running the upgrade (CCM does it). In this case, the version string is specified as github:apache/cassandra-3.11, which means that it will checkout the cassandra-3.11 branch from the GitHub repository whose alias is apache. Aliases are defined in CCM configuration file, usually located at ~/.ccm/config - in this particular case, it could be something like:
  1. [aliases]
  2. apache:git@github.com:apache/cassandra.git
  • current variant means that a released version of Cassandra will be used. It means that Cassandra distribution denoted by the specified version (3.11.10 in this case) is downloaded from the Apache repository/mirror - again, the repository can be defined in CCM configuration file, under repositories section, something like:
  1. [repositories]
  2. cassandra=https://archive.apache.org/dist/cassandra
  • min_proto_v, max_proto_v - the range of usable Cassandra driver protocol versions

  • java_versions - supported Java versions

The possible upgrade paths are defined later in the upgrade manifest - when you scroll the file, you will find the MANIFEST map which may look similar to:

  1. MANIFEST = {
  2. current_2_1_x: [indev_2_2_x, indev_3_0_x, indev_3_11_x],
  3. current_2_2_x: [indev_2_2_x, indev_3_0_x, indev_3_11_x],
  4. current_3_0_x: [indev_3_0_x, indev_3_11_x, indev_4_0_x],
  5. current_3_11_x: [indev_3_11_x, indev_4_0_x],
  6. current_4_0_x: [indev_4_0_x, indev_trunk],
  7. indev_2_2_x: [indev_3_0_x, indev_3_11_x],
  8. indev_3_0_x: [indev_3_11_x, indev_4_0_x],
  9. indev_3_11_x: [indev_4_0_x],
  10. indev_4_0_x: [indev_trunk]
  11. }

It is a simple map where for the origin version (as a key), there is a list of possible target versions (as a value). Say:

  1. current_4_0_x: [indev_4_0_x, indev_trunk]

means that upgrades from current_4_0_x to indev_4_0_x and from current_4_0_x to indev_trunk will be considered. You may make changes to that upgrade scenario in your development branch according to your needs. There is a command-line option that allows filtering across upgrade scenarios: --upgrade-version-selection=xxx. The possible values for that options are as follows:

  • indev - which is the default, only selects those upgrade scenarios where the target version is in indev variant

  • both - selects upgrade paths where either both origin and target versions are in the same variant or have the same version family

  • releases - selects upgrade paths between versions in current variant or from the current to indev variant if both have the same version family

  • all - no filtering at all - all variants are tested

Running upgrades with local distribution

The upgrade test can use your local Cassandra distribution, the one specified by the cassandra_dir property, as the target version if the following preconditions are satisfied:

  • the target version is in the indev variant,

  • the version family set in the version description matches the version family of your local distribution

For example, your local distribution is branched off from the cassandra-4.0 branch, likely matching indev_4_0_x. It means that the upgrade path with target version indev_4_0_x uses your local distribution. There is a handy command line option which will filter out all the upgrade tests which do not match the local distribution: --upgrade-target-version-only. Given you are on cassandra-4.0 branch, when applied to the previous example, it will be something similar to:

  1. pytest --cassandra-dir=/home/cassandra/cassandra --collect-only -q --execute-upgrade-tests --execute-upgrade-tests-only upgrade_tests/cql_tests.py -k test_set --upgrade-target-version-only

prints:

  1. upgrade_tests/cql_tests.py::cls::test_set
  2. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_current_4_0_x_To_indev_4_0_x::test_set
  3. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_indev_3_0_x_To_indev_4_0_x::test_set
  4. upgrade_tests/cql_tests.py::TestCQLNodes3RF3_Upgrade_indev_3_11_x_To_indev_4_0_x::test_set
  5. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_current_4_0_x_To_indev_4_0_x::test_set
  6. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_indev_3_0_x_To_indev_4_0_x::test_set
  7. upgrade_tests/cql_tests.py::TestCQLNodes2RF1_Upgrade_indev_3_11_x_To_indev_4_0_x::test_set

You can see that the upgrade tests were limited to the ones whose target version is indev and family matches 4.0.

Logging

A couple of common PyTest arguments control what is logged to the file and the console from the Python test code. Those arguments which start from --log-xxx are pretty well described in the help message (pytest --help) and in PyTest documentation, so it will not be discussed further. However, most of the tests start with the cluster of Cassandra nodes, and each node generates its own logging information and has its own data directories.

By default the logs from the nodes are copied to the unique directory created under logs subdirectory under root of dtest project. For example:

  1. (venv) cassandra@b69a382da7cd:~/cassandra-dtest$ ls logs/ -1
  2. 1627455923457_test_set
  3. 1627456019264_test_set
  4. 1627456474949_test_set
  5. 1627456527540_test_list
  6. last

The last item is a symbolic link to the directory containing the logs from the last executed test. Each such directory includes logs from each started node - system, debug, GC as well as standard streams registered upon each time the node was started:

  1. (venv) cassandra@b69a382da7cd:~/cassandra-dtest$ ls logs/last -1
  2. node1.log
  3. node1_debug.log
  4. node1_gc.log
  5. node1_startup-1627456480.3398306-stderr.log
  6. node1_startup-1627456480.3398306-stdout.log
  7. node1_startup-1627456507.2186499-stderr.log
  8. node1_startup-1627456507.2186499-stdout.log
  9. node2.log
  10. node2_debug.log
  11. node2_gc.log
  12. node2_startup-1627456481.10463-stderr.log
  13. node2_startup-1627456481.10463-stdout.log

Those log files are not collected if --delete-logs command-line option is added to PyTest. The nodes also produce data files which may be sometimes useful to examine to resolve some failures. Those files are usually deleted when the test is completed, but there are some options to control that behavior:

  • --keep-test-dir - keep the whole CCM directory with data files and logs when the test completes

  • --keep-failed-test-dir – only keep that directory when the test has failed

Now, how to find where is that directory for the certain test - you need to grab that information from the test logs - for example, you may add -s option to the command line and then look for "dtest_setup INFO" messages. For example:

  1. 05:56:06,383 dtest_setup INFO cluster ccm directory: /tmp/dtest-0onwvgkr

says that the cluster work directory is /tmp/dtest-0onwvgkr, and all node directories can be found under the test subdirectory:

  1. (venv) cassandra@b69a382da7cd:~/cassandra-dtest$ ls /tmp/dtest-0onwvgkr/test -1
  2. cluster.conf
  3. node1
  4. node2

Performance Testing

Performance tests for Cassandra are a special breed of tests that are not part of the usual patch contribution process. In fact, many people contribute a lot of patches to Cassandra without ever running performance tests. However, they are important when working on performance improvements; such improvements must be measurable.

Several tools exist for running performance tests. Here are a few to investigate:

Code Style

The Cassandra project follows Sun’s Java coding conventions for anything not expressly outlined in this document.

Note that the project has a variety of styles that have accumulated in different subsystems. Where possible a balance should be struck between these guidelines and the style of the code that is being modified as part of a patch. Patches should also limit their scope to the minimum necessary for safely addressing the concerns of the patch.

Checkstyle

Cassandra uses Checkstyle project for enforcing various checkstyle policies the project follows. Checkstyle is part of the build from Cassandra 4.1 included. You can consult the checkstyle configuration file called checkstyle.xml for the source code in src directory and checkstyle_test.xml for all code in test directory. The configuration files are located in the root of the Cassandra repository. Checkstyle can be executed independently for the main source code as well as for the tests by executing ant checkstyle and ant checkstyle-test respectively.

The checkstyle target is executed by default when e.g. build or jar targets are executed. There is a flag you can use for not enforcing checkstyle. This is particularly handy upon development. For example, by default, the checkstyle target checks that your changes in Java code do not include imports which are not used. However, while you develop, you do not want this check to be enforced because you are not interested in it while you develop as your code tends to be in the in-progress state. You can turn whole checkstyle off by specifying -Dno-checkstyle=true on the command line, for example like this: ant build -Dno-checkstyle=true.

Naming and Clear Semantics

Class, Method and Variable Naming

Avoid extraneous words, for example prefer x() over getX() or setX() where it makes semantic sense. At the same time, do not avoid using words that are necessary, for example if a descriptive word provides semantic context such as liveReplicas over replicas. This is essential when there are many conceptual instantiations for a variable that are not enforced by the type system, but be sure to be consistent in the word choice and order across all instantiations of the variable.

e.g. allReplicas, naturalReplicas, pendingReplicas, allLiveReplicas, etc.

Method and Variable Naming Consistency

Ensure consistency of naming within a method, and between methods. It may be that multiple names are appropriate for a concept, but these should not be mixed and matched within the project. If you modify a concept, or improve the naming of a concept, make all relevant - including existing - code consistent with the new terminology. If possible, correspond with a prior author before modifying their semantics.

Standard word meanings in method or property names

calculateX,computeX

Perform some potentially expensive work to produce x

refreshX,updateX

Recompute a memoized x

lookupX

Find x in a map, or other structure, that is efficient but not free

x

Return x, relatively cheaply

toX

Return a potentially expensive translation to x

asX

Return a cheap translation to x

asXView

Return a cheap translation to x, that will reflect changes in the source

isX, hasX, canX

Boolean property or method indicating a capability or logical state

For boolean variables, fields and methods, choose names that sound like predicates and cannot be confused with nouns.

Semantic Distinctions via the Type System

If possible, enforce semantic distinctions at compile time with the type system.

e.g. RangesAtEndpoint, EndpointsForRange and EndpointsForToken are all semantically different variants on a collection of replicas.

This makes the intent of the code clearer, and helps the compiler indicate where we may have unintentionally conflated concepts. They also provide opportunities to insert stronger runtime checks that our assumptions hold, and these constraints can provide further clarity when reading the code.

In the case of EndpointsForX, for instance, we enforce that we have no duplicate endpoints, and that all of the endpoints do fully cover X.

Enums for Boolean Properties

Prefer an enum to boolean properties and parameters, unless clarity will be harmed (e.g. helper methods that accept a computed boolean predicate result, of the same name as used in the method they assist). Try to balance name clashes that would affect static imports, against clear and simple names that represent the behavioural switch.

Semantic Distinctions via Member Variables

If a separate type for all concepts is too burdensome, a type that aggregates concepts together within member variables might be applicable.

The most obvious counter-example is not to use Pair, or a similar tuple. Unless it is extremely obvious, prefer a dedicated type with well named member variables.

For example, FetchReplicas for source and target replicas, and ReplicaLayout for the distinction between natural and pending replicas.

This may help authors notice other semantics they had overlooked, that might have led to subtly incorrect parameter provision to methods. Conversely, methods may choose to accept one of these encapsulating types, so that callers do not need to consider which member they should provide.

e.g. ConsistencyLevel.assureSufficientLiveReplicas requires very specific replica collections, that are quite distinct, that might be easily incorrectly provided (though this is still inadequate, as it needs to distinguish between live and non-live semantics, which remains to be improved)

Public APIs

These considerations are especially important for public APIs, including CQL, virtual tables, JMX, yaml, system properties, etc. Any planned additions must be carefully considered in the context of any existing APIs. Where possible the approach of any existing API should be followed. Where the existing API is poorly suited, a strategy should be developed to modify or replace the existing API with one that is more coherent in light of the changes - which should also carefully consider any planned or expected future changes to minimise churn. Any strategy for modifying APIs should be brought to dev@cassandra.apache.org for discussion.

Code Structure

Necessity

If an interface has only one implementation, remove it. If a method isn’t used, delete it.

Don’t implement hashCode(), equals(), toString() or other methods unless they provide immediate utility.

Specificity

Don’t overgeneralise. Implement the most specific method or class that you can, that handles the present use cases.

Methods and classes should have a single clear purpose, and should avoid special-cases where practical.

Class Layout

Consider where your methods and inner classes live with respect to each other. Methods that are of a similar category should be adjacent, as should methods that are primarily dependent on each other. Try to use a consistent pattern, e.g. helper methods may occur either before or after the method that uses them, but not both; method signatures that cover different combinations of parameters should occur in a consistent order visiting the parameter space.

Class declaration order should, approximately, go: inner classes, static properties, instance properties, constructors (incl static factory methods), getters/setters, main functional/API methods, helper (incl static) methods and classes. Clarity should always come first, however.

Method Clarity

A method should be short. There is no hard size limit, but a filled screen is a good warning size. However, be careful not to over-minimise your methods; a page of tiny functions is also hard to read.

The body of a method should be limited to the main conceptual work being done. Substantive ancillary logic, such as computing an intermediate result, evaluating complex predicates, performing auditing, logging, etc, are prime candidates for helper methods.

Compiler Assistance

Always use @Override annotations when implementing abstract or interface methods or overriding a parent method.

@Nullable, @NonNull, @ThreadSafe, @NotThreadSafe and @Immutable should be used as appropriate to communicate to both the compiler and readers.

Boilerplate

Prefer public final fields to private fields with getters (but prefer encapsulating behavior in “real” methods to either).

Declare class properties final wherever possible, but never declare local variables and parameters final. Variables and parameters should still be treated as immutable wherever possible, with explicit code blocks introduced as necessary to minimize the scope of any mutable variables.

Prefer initialization in a constructor to setters, and builders where the constructor is complex with many optional parameters.

Avoid redundant this references to member fields or methods, except for consistency with other assignments e.g. in the constructor

Exception handling

Never ever write catch (…) {} or catch (…) { logger.error() } merely to satisfy Java’s compile-time exception checking.

Always catch the narrowest exception type possible for achieving your goal. If Throwable must be caught for handling exceptional termination, it must be rethrown. If an exception cannot be safely handled locally, propagate it - but use unchecked exceptions if no caller expects to handle the case. Rethrow as RuntimeException, IOError, or your own UncheckedXException, or IllegalStateException if it “can’t happen”

Only if an exception is an explicitly acceptable condition can it be ignored, but this must be explained carefully in a comment detailing how this is handled correctly.

Formatting

{ and } are placed on a new line except when empty or opening a multi-line lambda expression. Braces may be elided to a depth of one if the condition or loop guards a single expression.

Lambda expressions accepting a single parameter should elide the braces that encapsulate the parameter. E.g. x → doSomething() and (x, y) → doSomething()

Multiline statements

Where possible prefer keeping a logical action to a single line. Prefer introducing additional variables, or well-named methods encapsulating actions, to multi-line statements - unless this harms clarity (e.g. in an already short method).

Try to keep lines under 120 characters, but use good judgment. It is better to exceed this limit, than to split a line that has no natural splitting points, particularly when the remainder of the line is boilerplate or easily inferred by the reader.

If a line wraps inside a method call, first extract any long parameter expressions to local variables before trying to group natural parameters together on a single line, aligning the start of parameters on each line, e.g.

  1. Type newType = new Type(someValueWithLongName, someOtherRelatedValueWithLongName,
  2. someUnrelatedValueWithLongName,
  3. someDoublyUnrelatedValueWithLongName);

When splitting a ternary, use one line per clause, carry the operator, and where possible align the start of the ternary condition, e.g.

  1. var = bar == null
  2. ? doFoo()
  3. : doBar();

It is usually preferable to carry the operator for multiline expressions, with the exception of some multiline string literals.

Whitespace

Make sure to use 4 spaces instead of the tab character for all your indentation. Many lines in the current files have a bunch of trailing whitespace. If you encounter incorrect whitespace, clean up in a separate patch. Current and future reviewers won’t want to review whitespace diffs.

Static Imports

Consider using static imports for frequently used utility methods that are unambiguous. E.g. String.format, ByteBufferUtil.bytes, Iterables.filter/any/transform.

When naming static methods, select names that maintain semantic legibility when statically imported, and are unlikely to clash with other method names that may be mixed in the same context.

Imports

Observe the following order for your imports:

  1. java
  2. [blank line]
  3. com.google.common
  4. org.apache.commons
  5. org.junit
  6. org.slf4j
  7. [blank line]
  8. everything else alphabetically

Format files for IDEs

How-to Commit

If you are a committer, feel free to pick any process that works for you - so long as you are planning to commit the work yourself.

Patch based Contribution

Here is how committing and merging will usually look for merging and pushing for tickets that follow the convention (if patch-based):

Hypothetical CASSANDRA-12345 ticket is a cassandra-4.0 based bug fix that requires different code for cassandra-4.0, cassandra-4.1, and trunk. Contributor Jackie supplied a patch for the root branch (12345-4.0.patch), and patches for the remaining branches (12345-4.1.patch, 12345-5.0.patch, 12345-trunk.patch).

On cassandra-4.0

  1. git am -3 12345-4.0.patch (any problem b/c of CHANGES.txt not merging anymore, fix it in place)

  2. ant realclean && ant jar (rebuild to make sure code compiles)

  3. git commit --amend (Notice this will squash the 4.0 applied patch into the forward merge commit)

On cassandra-4.1

  1. git merge cassandra-4.0 -s ours --log

  2. git apply -3 12345-4.1.patch (any issue with CHANGES.txt : fix and git add CHANGES.txt)

  3. ant realclean && ant jar (rebuild to make sure code compiles)

  4. git commit --amend (Notice this will squash the 4.1 applied patch into the forward merge commit)

On cassandra-5.0

  1. git merge cassandra-4.1 -s ours --log

  2. git apply -3 12345-5.0.patch (any issue with CHANGES.txt : fix and git add CHANGES.txt)

  3. ant realclean && ant jar check (rebuild to make sure code compiles)

  4. git commit --amend (Notice this will squash the 4.1 applied patch into the forward merge commit)

On trunk

  1. git merge cassandra-5.0 -s ours --log

  2. git apply -3 12345-trunk.patch (any issue with CHANGES.txt : fix and git add CHANGES.txt)

  3. ant realclean && ant jar check (rebuild to make sure code compiles)

  4. git commit --amend (Notice this will squash the trunk applied patch into the forward merge commit)

On any branch

  1. git push origin cassandra-4.0 cassandra-4.1 cassandra-5.0 trunk --atomic -n (dryrun check)

  2. git push origin cassandra-4.0 cassandra-4.1 cassandra-5.0 trunk --atomic

Git branch based Contribution

Same scenario, but a branch-based contribution:

On cassandra-4.0

  1. git cherry-pick <sha-of-4.0-commit> (any problem b/c of CHANGES.txt not merging anymore, fix it in place)

  2. ant realclean && ant jar (rebuild to make sure code compiles)

On cassandra-4.1

  1. git merge cassandra-4.0 -s ours --log

  2. git format-patch -1 <sha-of-4.1-commit> (alternative to format-patch and apply is cherry-pick -n)

  3. git apply -3 <sha-of-4.1-commit>.patch (any issue with CHANGES.txt : fix and git add CHANGES.txt)

  4. ant realclean && ant jar (rebuild to make sure code compiles)

  5. git commit --amend (Notice this will squash the 4.1 applied patch into the forward merge commit)

On cassandra-5.0

  1. git merge cassandra-4.1 -s ours --log

  2. git format-patch -1 <sha-of-5.0-commit> (alternative to format-patch and apply is cherry-pick -n)

  3. git apply -3 <sha-of-5.0-commit>.patch (any issue with CHANGES.txt : fix and git add CHANGES.txt)

  4. ant realclean && ant jar check (rebuild to make sure code compiles)

  5. git commit --amend (Notice this will squash the 5.0 applied patch into the forward merge commit)

On trunk

  1. git merge cassandra-5.0 -s ours --log

  2. git format-patch -1 <sha-of-trunk-commit> (alternative to format-patch and apply is cherry-pick -n)

  3. git apply -3 <sha-of-trunk-commit>.patch (any issue with CHANGES.txt : fix and git add CHANGES.txt)

  4. ant realclean && ant jar check (rebuild to make sure code compiles)

  5. git commit --amend (Notice this will squash the trunk applied patch into the forward merge commit)

On any branch

  1. git push origin cassandra-4.0 cassandra-4.1 cassandra-5.0 trunk --atomic -n (dryrun check)

  2. git push origin cassandra-4.0 cassandra-4.1 cassandra-5.0 trunk --atomic

Contributions only for release branches

If the patch is for an older branch, and doesn’t impact later branches (such as trunk), we still need to merge up.

On cassandra-4.0

  1. git cherry-pick <sha-of-4.0-commit> (any problem b/c of CHANGES.txt not merging anymore, fix it in place)

  2. ant realclean && ant jar (rebuild to make sure code compiles)

On cassandra-4.1

  1. git merge cassandra-4.0 -s ours --log

  2. ant realclean && ant jar (rebuild to make sure code compiles)

On cassandra-5.0

  1. git merge cassandra-4.1 -s ours --log

  2. ant realclean && ant jar check (rebuild to make sure code compiles)

On trunk

  1. git merge cassandra-4.1 -s ours --log

  2. ant realclean && ant jar check (rebuild to make sure code compiles)

On any branch

  1. git push origin cassandra-4.0 cassandra-4.1 trunk --atomic -n (dryrun check)

  2. git push origin cassandra-4.0 cassandra-4.1 trunk --atomic

Tips

Tip

A template for commit messages:

  1. <One sentence description, usually Jira title or CHANGES.txt summary>
  2. <Optional lengthier description>
  3. patch by <Authors>; reviewed by <Reviewers> for CASSANDRA-#####
  4. Co-authored-by: Name1 <email1>
  5. Co-authored-by: Name2 <email2>
Tip

Notes on git flags: -3 flag to am and apply will instruct git to perform a 3-way merge for you. If a conflict is detected, you can either resolve it manually or invoke git mergetool - for both am and apply.

—atomic flag to git push does the obvious thing: pushes all or nothing. Without the flag, the command is equivalent to running git push once per each branch. This is nifty in case a race condition happens - you won’t push half the branches, blocking other committers’ progress while you are resolving the issue.

Tip

The fastest way to get a patch from someone’s commit in a branch on GH - if you don’t have their repo in remotes - is to append .patch to the commit url, e.g. curl -O github.com/apache/cassandra/commit/7374e9b5ab08c1f1e612bf72293ea14c959b0c3c.patch

Tip

git cherry-pick -n <sha-of-X.X-commit> can be used in place of the git format-patch -1 <sha-of-X.X-commit> ; git apply -3 <sha-of-X.X-commit>.patch steps.

Review Checklist

When reviewing tickets in Apache JIRA, the following items should be covered as part of the review process:

General

  • Does it conform to the code_style guidelines?

  • Is there any redundant or duplicate code?

  • Is the code as modular as possible?

  • Can any singletons be avoided?

  • Can any of the code be replaced with library functions?

  • Are units of measurement used in the code consistent, both internally and with the rest of the ecosystem?

Error-Handling

  • Are all data inputs and outputs checked (for the correct type, length, format, and range) and encoded?

  • Where third-party utilities are used, are returning errors being caught?

  • Are invalid parameter values handled?

  • Are any Throwable/Exceptions passed to the JVMStabilityInspector?

  • Are errors well-documented? Does the error message tell the user how to proceed?

  • Do exceptions propagate to the appropriate level in the code?

Documentation

  • Do comments exist and describe the intent of the code (the “why”, not the “how”)?

  • Are javadocs added where appropriate?

  • Is any unusual behavior or edge-case handling described?

  • Are data structures and units of measurement explained?

  • Is there any incomplete code? If so, should it be removed or flagged with a suitable marker like ‘TODO’?

  • Does the code self-document via clear naming, abstractions, and flow control?

  • Have NEWS.txt, the cql3 docs, and the native protocol spec been updated if needed?

  • Is the ticket tagged with “client-impacting” and “doc-impacting”, where appropriate?

  • Has lib/licences been updated for third-party libs? Are they Apache License compatible?

  • Is the Component on the JIRA ticket set appropriately?

Testing

  • Is the code testable? i.e. don’t add too many or hide dependencies, unable to initialize objects, test frameworks can use methods etc.

  • Do tests exist and are they comprehensive?

  • Do unit tests actually test that the code is performing the intended functionality?

  • Could any test code use common functionality (e.g. ccm, dtest, or CqlTester methods) or abstract it there for reuse?

  • If the code may be affected by multi-node clusters, are there dtests?

  • If the code may take a long time to test properly, are there CVH tests?

  • Is the test passing on CI for all affected branches (up to trunk, if applicable)? Are there any regressions?

  • If patch affects read/write path, did we test for performance regressions w/multiple workloads?

  • If adding a new feature, were tests added and performed confirming it meets the expected SLA/use-case requirements for the feature?

Logging

  • Are logging statements logged at the correct level?

  • Are there logs in the critical path that could affect performance?

  • Is there any log that could be added to communicate status or troubleshoot potential problems in this feature?

  • Can any unnecessary logging statement be removed?

Contributing Code Changes

Choosing What to Work on

Submitted patches can include bug fixes, changes to the Java code base, improvements for tooling (both Java or Python), documentation, testing or any other changes that requires changing the code base. Although the process of contributing code is always the same, the amount of work and time it takes to get a patch accepted also depends on the kind of issue you’re addressing.

As a general rule of thumb

  • Major new features and significant changes to the code base will likely not be accepted without deeper discussion within the developer community.

  • Bug fixes take higher priority compared to features.

  • The extent to which tests are required depends on how likely your changes will effect the stability of Cassandra in production. Tooling changes requires fewer tests than storage engine changes.

  • Less complex patches will be reviewed faster; consider breaking up an issue into individual tasks and contributions that can be reviewed separately.

Hint

Not sure what to work? Just pick an issue marked as Low Hanging Fruit Complexity in JIRA, which flags issues that often turn out to be good starter tasks for beginners.

To create a JIRA account, please request it on the #cassandra or #cassandra-dev channels on ASF Slack, or on the user or dev mailing list.

Before You Start Coding

Although contributions are highly appreciated, we do not guarantee that every contribution will become a part of Cassandra. Therefore, it’s generally a good idea to first get some feedback on the thing you plan to do, especially about any new features or major changes to the code base. You can reach out to other developers on the mailing list or Slack.

You should also

  • Avoid redundant work by searching for already reported issues in JIRA to work on.

  • Create a new issue early in the process describing what you’re working on - before finishing your patch.

  • Link related JIRA issues with your own ticket to provide a better context.

  • Update your ticket from time to time by giving feedback on your progress and link a GitHub WIP branch with your current code.

  • Ping people who you actively like to ask for advice on JIRA by mentioning users.

There are also some fixed rules that you need to be aware

  • Patches will only be applied to branches by following the release model

  • Code must be testable

  • Code must follow the code style convention

  • Changes must not break compatibility between different Cassandra versions

  • Contributions must be covered by the Apache License

Choosing the Right Branches to Work on

There are currently multiple Cassandra versions maintained in individual branches:

VersionPolicy

4.0

Code freeze (see below)

3.11

Critical bug fixes only

3.0

Critical bug fixes only

2.2

Critical bug fixes only

2.1

Critical bug fixes only

Corresponding branches in git are easy to recognize as they are named cassandra-<release> (e.g. cassandra-3.0). The trunk branch is an exception, as it contains the most recent commits from all other branches and is used for creating new branches for future tick-tock releases.

4.0 Code Freeze

Patches for new features are currently not accepted for 4.0 or any earlier versions. All efforts should focus on stabilizing the 4.0 branch before the first official release. During that time, only the following patches will be considered for acceptance:

  • Bug fixes

  • Measurable performance improvements

  • Changes not distributed as part of the release such as:

  • Testing related improvements and fixes

  • Build and infrastructure related changes

  • Documentation

Bug Fixes

Creating patches for bug fixes is a bit more complicated and will depend on how many different versions of Cassandra are affected. In each case, the order for merging such changes will be cassandra-2.1cassandra-2.2cassandra-3.0cassandra-3.xtrunk. But don’t worry, merging from 2.1 would be the worst case for bugs that affect all currently supported versions, an uncommon event. As a contributor, you’re also not expected to provide a single patch for each version. What you need to do however is:

  • Be clear about which versions you could verify to be affected by the bug

  • For 2.x: ask if a bug qualifies to be fixed in this release line, as this may be handled on case by case bases

  • If possible, create a patch against the lowest version in the branches listed above (e.g. if you found the bug in 3.9 you should try to fix it already in 3.0)

  • Test if the patch can be merged cleanly across branches in the direction listed above

  • Be clear which branches may need attention by the committer or even create custom patches for those if you can

Creating a Patch

So you’ve finished coding and the great moment arrives: it’s time to submit your patch!

  1. Create a branch for your changes if you haven’t done already. Many contributors name their branches based on ticket number and Cassandra version, e.g. git checkout -b 12345-3.0 or git checkout -b CASSANDRA-12345-3.0.
Hint

When working on multiple patches at the same time, instead of cloning the repository separately for each of them, set up multiple working trees for different Cassandra versions from the same repository using Git worktrees feature.

  1. Verify that you follow Cassandra’s code style

  2. Make sure all tests (including yours) pass using ant as described in testing. If you suspect a test failure is unrelated to your change, it may be useful to check the test’s status by searching the issue tracker or looking at CI results for the relevant upstream version. Note that the full test suites take many hours to complete, so it is common to only run specific relevant tests locally before uploading a patch. Once a patch has been uploaded, the reviewer or committer can help setup CI jobs to run the full test suites.

  3. Consider going through the how to review page for your code. This will help you to understand how others will consider your change for inclusion.

  4. Don’t make the committer squash commits for you in the root branch either. Multiple commits are fine - and often preferable - during review stage, especially for incremental review, but once +1d, do either:

  5. Attach a patch to JIRA with a single squashed commit in it (per branch), or

  6. Squash the commits in-place in your branches into one

  7. Include a CHANGES.txt entry (put it at the top of the list), and format the commit message appropriately in your patch as below. Please note that only user-impacting items should be listed in CHANGES.txt. If you fix a test that does not affect users and does not require changes in runtime code, then no CHANGES.txt entry is necessary.

    1. <One sentence description, usually Jira title and CHANGES.txt summary>
    2. <Optional lengthier description>
    3. patch by <Authors>; reviewed by <Reviewers> for CASSANDRA-#####
  8. When you’re happy with the result, create a patch. We suggest that you use a similar format (note blank lines) for the commit log message:

    1. <one sentence description>
    2. <optional lengthier description>
    3. Patch by <authors>; reviewed by <Reviewers> for CASSANDRA-#####

If you don’t know who is reviewing your change yet, you can use TBD and amend the commit later to note the people who helped you.

  1. git add <any new or modified file>
  2. git commit
  3. git format-patch HEAD~1
  4. mv <patch-file> <ticket-branchname.txt> (e.g. 12345-trunk.txt, 12345-3.0.txt)

Alternatively, many contributors prefer to make their branch available on GitHub. In this case, fork the Cassandra repository on GitHub and push your branch:

  1. git push --set-upstream origin 12345-3.0
  1. To make life easier for your reviewer/committer, you may want to make sure your patch applies cleanly to later branches and create additional patches/branches for later Cassandra versions to which your original patch does not apply cleanly. That said, this is not critical, and you will receive feedback on your patch regardless.

  2. Attach the newly generated patch to the ticket/add a link to your branch and click “Submit Patch” at the top of the ticket. This will move the ticket into “Patch Available” status, indicating that your submission is ready for review.

  3. Wait for other developers or committers to review it and hopefully +1 the ticket (see how to review). If your change does not receive a +1, do not be discouraged. If possible, the reviewer will give suggestions to improve your patch or explain why it is not suitable.

  4. If the reviewer has given feedback to improve the patch, make the necessary changes and move the ticket into “Patch Available” once again.

  5. Once you’ve had the patch reviewed you can amend the commit to update the commit message TBD with the reviewers who helped you.

Once the review process is complete, you will receive a +1. Wait for a committer to commit it. Do not delete your branches immediately after they’ve been committed - keep them on GitHub for a while. Alternatively, attach a patch to JIRA for historical record. It’s not that uncommon for a committer to mess up a merge. In case of that happening, access to the original code is required, or else you’ll have to redo some of the work.

CI Environments

About CI testing and Apache Cassandra

Cassandra can be automatically tested using various test suites, that are either implemented based on JUnit or the dtest scripts written in Python. As outlined in testing, each kind of test suite addresses a different way to test Cassandra. Eventually, all of the tests will be executed together on the CI platform at builds.apache.org, running Jenkins.

Setting up your own Jenkins server

Jenkins is an open source solution that can be installed on a large number of platforms. Setting up a custom Jenkins instance for Cassandra may be desirable for users who have hardware to spare, or organizations that want to run Cassandra tests for custom patches before contribution.

Please refer to the Jenkins download and documentation pages for details on how to get Jenkins running, possibly also including slave build executor instances. The rest of the document will focus on how to setup Cassandra jobs in your Jenkins environment.

Required plugins

In addition, the following plugins need to be installed along with the standard plugins (git, ant, ..).

You can install any missing plugins using the install manager.

Go to Manage Jenkins → Manage Plugins → Available and install the following plugins and respective dependencies:

  • Job DSL

  • Javadoc Plugin

  • description setter plugin

  • Throttle Concurrent Builds Plug-in

  • Test stability history

  • Post Build Script

Setup seed job

  1. Config New Item

    • Name it Cassandra-Job-DSL

    • Select Freestyle project

  2. Under Source Code Management select Git using the repository: [github.com/apache/cassandra-builds](https://github.com/apache/cassandra-builds)

  3. Under Build, confirm Add build stepProcess Job DSLs and enter at Look on Filesystem: jenkins-dsl/cassandra_job_dsl_seed.groovy

Generated jobs will be created based on the Groovy script’s default settings. You may want to override settings by checking This project is parameterized and add String Parameter for on the variables that can be found in the top of the script. This will allow you to setup jobs for your own repository and branches (e.g. working branches).

  1. When done, confirm “Save”.

You should now find a new entry with the given name in your project list. However, building the project will still fail and abort with an error message “Processing DSL script cassandra_job_dsl_seed.groovy ERROR: script not yet approved for use”. Go to Manage JenkinsIn-process Script Approval to fix this issue. Afterwards you should be able to run the script and have it generate numerous new jobs based on the found branches and configured templates.

Jobs are triggered by either changes in Git or are scheduled to execute periodically, e.g. on daily basis. Jenkins will use any available executor with the label “cassandra”, once the job is to be run. Please make sure to make any executors available by selecting Build Executor StatusConfigure → Add “cassandra” as label and save.

Executors need to have “JDK 1.8 (latest)” installed. This is done under Manage Jenkins → Global Tool Configuration → JDK Installations…. Executors also need to have the virtualenv package installed on their system.

CircleCI

Cassandra ships with a default CircleCI configuration to enable running tests on your branches. Go to the CircleCI website, click “Login” and log in with your github account. Then give CircleCI permission to watch your repositories.

Once you have done that, you can optionally configure CircleCI to run tests in parallel if you wish:

  1. Click Projects and select your github account, and then click the settings for your project.

  2. Set the parallelism setting. If you leave the default value of 1 for Cassandra, only ant eclipse-warnings and ant test will be run. If you change the value to 4, Circle CI also runs ant long-test, ant test-compression and ant stress-test.

Dependency Management

New dependencies should not be included without community consensus first being obtained via a [DISCUSS] thread on the dev@cassandra.apache.org mailing list.

As Cassandra is an ASF project, all included libraries must follow Apache’s software license requirements.

Cassandra uses the Ant build system and Maven POMs for dependency specification. In Cassandra 5.0 the format of POMs was moved from within the build.xml file to separate POM template files that are processed by Ant. In both pre-5.0 and post-5.0 Cassandra, there are several POMs that dependencies can be included in:

  • parent-pom

    • Contains all dependencies with the respective version. All other poms will refer to the artifacts with specified versions listed here.

    • Since Cassandra 5.0, the parent-pom template is .build/parent-pom-template.xml.

  • build-deps-pom(-sources) + coverage-deps-pom

    • used by the ant build target. Listed dependencies will be resolved and copied to build/lib/{jar,sources} by executing the maven-ant-tasks-retrieve-build target. This should contain libraries that are required for build tools (grammar, docs, instrumentation), but are not shipped as part of the Cassandra distribution.

    • Since Cassandra 4.0, coverage-deps-pom has been removed and the build-deps-pom template is .build/cassandra-build-deps-template.xml.

  • all-pom

  • test-deps-pom

    • Referenced by maven-ant-tasks-retrieve-test to retrieve and save dependencies to build/test/lib. Exclusively used during JUnit test execution.

    • Since Cassandra 3.0, test-deps-pom has been removed.

The ant write-poms target produces valid POM files in the build/ directory.

Dependencies added to the lib/ directory are built into the release artifacts by the ant artifacts target (see target resolver-dist-lib). Libraries distributed this way must meet the ASF distribution policy.

Dependency management before Cassandra 5.0

To update dependencies, update the parent and child POM definitions in build.xml. The parent POM should include the dependency tag with groupId, artifactId, version, and optional scope fields. The child POM(s) should include the dependency tag with groupId and artifactId. See the Maven docs for a complete reference on how to reference dependencies across parent and child POMs.

Here is an example of a commit that changes dependency versions pre-5.0.

Dependency management in Cassandra 5.0 and later

In Cassandra 5.0 and later, dependencies are managed in Maven POM templates in .build/*-template.xml. These templates are processed into valid Maven POMs and copied to build/\*.pom by the ant write-poms task.

For new dependencies, add to parent-pom-template and cassandra-deps-template, and optionally cassandra-build-deps-template if the dependency is required for build only. See the Maven docs on how to reference dependencies in the parent POM from the child POMs.

For dependency versions that need to be available in build.xml and parent-pom-template, specify the version as a property in build.xml, add it to the ant write-poms target, then add the property to parent-pom-template with the value of the template substitution.

Here is an example of a commit that changes dependency versions since 5.0.

Troubleshooting and conflict resolution

Here are some useful commands that may help you out resolving conflicts.

  • ant realclean - gets rid of the build directory, including build artifacts.

  • mvn dependency:tree -f build/apache-cassandra-\*-SNAPSHOT.pom -Dverbose -Dincludes=org.slf4j

    • shows transitive dependency tree for artifacts, e.g. org.slf4j. In case the command above fails due to a missing parent pom file, try running ant mvn-install.
  • rm ~/.m2/repository/org/apache/cassandra/apache-cassandra/ - removes cached local Cassandra maven artifacts

Working on Documentation

Working on Documentation

How Cassandra is documented

The official Cassandra documentation lives in the project’s git repository. We use a static site generator, Antora, to create pages hosted at cassandra.apache.org.

<!— You’ll also find developer-centric content about Cassandra internals in our retired wiki (not covered by this guide). -→

Using a static site generator often requires the use of a markup language instead of visual editors (which some people would call good news). Antora processes AsciiDoc, the markup language used to generate our documentation. Markup languages allow you to format text using certain syntax elements. Your document structure will also have to follow specific conventions. Feel free to take a look at existing documents to get a better idea how we structure our documents.

So how do you actually start making contributions?

GitHub based work flow

Recommended for shorter documents and minor changes on existing content (e.g. fixing typos or updating descriptions)

Follow these steps to contribute using GitHub. It’s assumed that you’re logged in with an existing account.

  1. Fork the GitHub mirror of the Cassandra repository

image

  1. Create a new branch that you can use to make your edits. It’s recommended to have a separate branch for each of your working projects. It will also make it easier to create a pull request later to when you decide you’re ready to contribute your work.

image

  1. Navigate to document sources doc/source/modules to find the .adoc file to edit. The URL of the document should correspond to the directory structure within the modules, where first the component name, such as cassandra is listed, and then the actual pages inside the pages directory. New files can be created using the “Create new file” button:

image

  1. At this point you should be able to edit the file using the GitHub web editor. Start by naming your file and add some content. Have a look at other existing .adoc files to get a better idea what format elements to use.

image

Make sure to preview added content before committing any changes.

image

  1. Commit your work when you’re done. Make sure to add a short description of all your edits since the last time you committed before.

image

  1. Finally if you decide that you’re done working on your branch, it’s time to create a pull request!

image

Afterwards the GitHub Cassandra mirror will list your pull request and you’re done. Congratulations! Please give us some time to look at your suggested changes before we get back to you.

Jira based work flow

Recommended for major changes

Significant changes to the documentation are best managed through our Jira issue tracker. Please follow the same contribution guides as for regular code contributions. Creating high quality content takes a lot of effort. It’s therefore always a good idea to create a ticket before you start and explain what you’re planning to do. This will create the opportunity for other contributors and committers to comment on your ideas and work so far. Eventually your patch gets a formal review before it is committed.

Working on documents locally using Antora

Recommended for advanced editing

Using the GitHub web interface should allow you to use most common layout elements including images. More advanced formatting options and navigation elements depend on Antora to render correctly. Therefore, it’s a good idea to setup Antora locally for any serious editing. Please follow the instructions in the Cassandra source directory at doc/README.md. Setup is very easy (at least on OSX and Linux).

Notes for committers

Please feel free to get involved and merge pull requests created on the GitHub mirror if you’re a committer. As this is a read-only repository, you won’t be able to merge a PR directly on GitHub. You’ll have to commit the changes against the Apache repository with a comment that will close the PR when the committ syncs with GitHub.

You may use a git work flow like this:

  1. git remote add github https://github.com/apache/cassandra.git
  2. git fetch github pull/<PR-ID>/head:<PR-ID>
  3. git checkout <PR-ID>

Now either rebase or squash the commit, e.g. for squashing:

  1. git reset --soft origin/trunk
  2. git commit --author <PR Author>

Make sure to add a proper commit message including a “Closes #<PR-ID>” text to automatically close the PR.

Publishing

Details for building and publishing of the site at cassandra.apache.org can be found here.

Release Process

The steps for Release Managers to create, vote, and publish releases for Apache Cassandra.

While a committer can perform the initial steps of creating and calling a vote on a proposed release, only a PMC member can complete the process of publishing and announcing the release.

Prerequisites

Background docs

A debian based linux OS is required to run the release steps from. Debian-based distros provide the required RPM, dpkg and repository management tools.

Create and publish your GPG key

To create a GPG key, follow the guidelines. The key must be 4096 bit RSA.

Publish your GPG key in a PGP key server, such as MIT Keyserver. Some gpg clients are publishing the keys here. You are welcome to set the server where the keys will be published by following this guide.

Once completed, you need to create a ticket like this and ask a PMC to add your key to KEYS file.

A PMC will include your public key to this file:

  1. https://dist.apache.org/repos/dist/release/cassandra/KEYS

Create Release Artifacts

Any committer can perform the following steps to create and call a vote on a proposed release.

Check that there are no open urgent Jira tickets currently being worked on. Also check with the PMC that there’s security vulnerabilities currently being worked on in private. Current project habit is to check the timing for a new release on the dev mailing lists.

For successful building process, install this tooling locally: svn, git, ant, devscripts, reprepro, rpmsign, docker, createrepo (the script is checking this tooling is present before proceeding any further). The names of these “packages” are Debian-centric, but equivalents should be discoverable in other systems too.

There is a package called createrepo-c in Debian Bullseye. Please beware that createrepo package is not located in Ubuntu 20.04 LTS. createrepo package is present in Ubuntu Bionic (18.04), createrepo-c is in Ubuntu Jammy (22.04 LTS) and more recent.

Perform the Release

Run the following commands to generate and upload release artifacts, to the ASF nexus staging repository and dev distribution location:

  1. cd ~/git
  2. git clone https://github.com/apache/cassandra-builds.git
  3. git clone https://github.com/apache/cassandra.git
  1. # Edit the variables at the top of the `prepare_release.sh` file
  2. edit cassandra-builds/cassandra-release/prepare_release.sh

You must specify your ASF username to asf_username variable. Next, gpg_key environment variable must be set to a fingerprint of your gpg key. Execute gpg --list-keys or a similar command to get the value. Finally, you must add ASF remote to your cloned repository and git_asf_remote variable needs to be set to point to that. For example, when this command is executed:

  1. git remote add asf https://gitbox.apache.org/repos/asf/cassandra.git

then git_asf_remote variable needs to be set to asf. NOTE: This is very important step as tags are pushed to ASF repository and they are synchronized to GitHub automatically.

  1. # Ensure your 4096 RSA key is the default secret key
  2. edit ~/.gnupg/gpg.conf # update the `default-key` line

A reference configuration should look like these examples:

  1. default-key <fingerprint of your key>
  2. personal-digest-preferences SHA512
  3. cert-digest-algo SHA512
  4. default-preference-list SHA512 SHA384 SHA256 SHA224 AES256 AES192 AES CAST5 ZLIB BZIP2 ZIP Uncompressed
  1. edit ~/.rpmmacros # update the `%_gpg_name <key_id>` line
  2. # Ensure DEBFULLNAME and DEBEMAIL is defined and exported, in the debian scripts configuration
  3. edit ~/.devscripts

The reference content of these files is:

  1. $ cat ~/.rpmmacros
  2. %_gpg_name email@youusedforyourkey.org
  3. $ cat ~/.devscripts
  4. DEBFULLNAME="Your Name"
  5. DEBEMAIL=email@youusedforyourkey.org

Empirical testing shows that you also must have the above DEB* environemnt variables exported before proceeding.

Additionally, you must configure $HOME/.m2/settings.xml to contain the credentials used to upload artifacts to staging repository. The credentials are your ASF credentials.

  1. $ cat ~/.m2/settings.xml
  2. <settings>
  3. <servers>
  4. <server>
  5. <id>apache.releases.https</id>
  6. <username>yourasflogin</username>
  7. <password>yourasfpassword</password>
  8. </server>
  9. <server>
  10. <id>apache.snapshots.https</id>
  11. <username>yourasflogin</username>
  12. <password>yourasfpassword</password>
  13. </server>
  14. </servers>
  15. </settings>

The script will eventually ask you for a username and password to push artifacts to SVN. The default prompt for user will be equal to a username of an account at your machine. If your ASF login is not same as your username locally, just error out the prompt (put there wrong password and confirm), and it will ask you for username again without providing any default value.

The prepare_release.sh is run from the actual cassandra git checkout, on the branch/commit that we wish to tag for the tentative release along with version number to tag.

  1. cd cassandra
  2. git switch cassandra-<version-branch>
  3. # The following cuts the release artifacts (including deb and rpm packages) and deploy to staging environments
  4. ../cassandra-builds/cassandra-release/prepare_release.sh -v <version>

Follow the prompts.

Once artifacts are built and pushed to the staging area, the script will pause and require you to go to the Staging repositories, where you will find the repository. Select the Cassandra repository and push the “Close” button. Please take a note of the number of that repository like orgapachecassandra-1283 - number is 1283. The script will require the repo number to proceed. It will use this number in the rendered e-mail template sent to the dev list, etc.

If building the deb or rpm packages fail, those steps can be repeated individually using the -d and -r flags, respectively.

Once DEBs and RPMs are also uploaded, do not forget to merge your commit to prepare the release to trunk and push after all artifacts are uploaded finish the process. You will be also reminded to do that by the script itself at the end.

Call for a Vote

Fill out the following email template you find in $HOME/Mail directory and send to the dev mailing list:

  1. I propose the following artifacts for release as <version>.
  2. sha1: <git-sha>
  3. Git: https://gitbox.apache.org/repos/asf?p=cassandra.git;a=shortlog;h=refs/tags/<version>-tentative
  4. Artifacts: https://repository.apache.org/content/repositories/orgapachecassandra-<nexus-id>/org/apache/cassandra/apache-cassandra/<version>/
  5. Staging repository: https://repository.apache.org/content/repositories/orgapachecassandra-<nexus-id>/
  6. The distribution packages are available here: https://dist.apache.org/repos/dist/dev/cassandra/${version}/
  7. The vote will be open for 72 hours (longer if needed).
  8. [1]: (CHANGES.txt) https://git1-us-west.apache.org/repos/asf?p=cassandra.git;a=blob_plain;f=CHANGES.txt;hb=<version>-tentative
  9. [2]: (NEWS.txt) https://git1-us-west.apache.org/repos/asf?p=cassandra.git;a=blob_plain;f=NEWS.txt;hb=<version>-tentative

Post-vote operations

Any PMC member can perform the following steps to formalize and publish a successfully voted release.

Publish Artifacts

Run the following commands to publish the voted release artifacts:

  1. cd ~/git
  2. # edit the variables at the top of the `finish_release.sh` file
  3. edit cassandra-builds/cassandra-release/finish_release.sh
  4. # After cloning cassandra-builds repo, `finish_release.sh` is run from the actual cassandra git checkout,
  5. # on the tentative release tag that we wish to tag for the final release version number tag.
  6. cd ~/git/cassandra/
  7. git checkout <version>-tentative
  8. ../cassandra-builds/cassandra-release/finish_release.sh -v <version>

If successful, take note of the email text output which can be used in the next section “Send Release Announcement”. The output will also list the next steps that are required.

Promote Nexus Repository

  • Login to Nexus repository again.

  • Click on “Staging Repositories” and then on the repository with id “cassandra-staging”.

  • Find your closed staging repository, select it and choose “Release”. This may take some time, but eventually the repository will no longer show in Staging Repositories.

  • Next click on “Repositories”, and select “Public Repositories” and validate that your artifacts exist as you expect them.

Update and Publish Website

See docs for building and publishing the website.

Also update the CQL doc if appropriate.

Release version in JIRA

Release the JIRA version.

  • In JIRA go to the version that you want to release and release it.

  • Create a new version, if it has not been done before.

Update to Next Development Version

Update the codebase to point to the next development version:

  1. cd ~/git/cassandra/
  2. git checkout cassandra-<version-branch>
  3. edit build.xml # update `<property name="base.version" value="…"/> `
  4. edit debian/changelog # add entry for new version
  5. edit CHANGES.txt # add entry for new version
  6. git commit -m "Increment version to <next-version>" build.xml debian/changelog CHANGES.txt
  7. # …and forward merge and push per normal procedure

Wait for Artifacts to Sync

Wait for the artifacts to sync at downloads.apache.org/cassandra/

Send Release Announcement

Fill out the following email template and send to both user and dev mailing lists:

  1. The Cassandra team is pleased to announce the release of Apache Cassandra version <version>.
  2. Apache Cassandra is a fully distributed database. It is the right choice
  3. when you need scalability and high availability without compromising
  4. performance.
  5. http://cassandra.apache.org/
  6. Downloads of source and binary distributions are listed in our download
  7. section:
  8. http://cassandra.apache.org/download/
  9. This version is <the first|a bug fix> release[1] on the <version-base> series. As always,
  10. please pay attention to the release notes[2] and let us know[3] if you
  11. were to encounter any problem.
  12. Enjoy!
  13. [1]: (CHANGES.txt) https://git1-us-west.apache.org/repos/asf?p=cassandra.git;a=blob_plain;f=CHANGES.txt;hb=<version>
  14. [2]: (NEWS.txt) https://git1-us-west.apache.org/repos/asf?p=cassandra.git;a=blob_plain;f=NEWS.txt;hb=<version>
  15. [3]: https://issues.apache.org/jira/browse/CASSANDRA

Update Slack Cassandra topic —————————————-

Update topic in cassandra Slack room <slack>

/topic cassandra.apache.org | Latest releases: 4.1.0, 4.0.7, 3.11.4, 3.0.18 | ask, don’t ask to ask

Tweet from @Cassandra

Tweet the new release, from the @Cassandra account

Delete Old Releases

As described in When to Archive.

An example of removing old releases:

  1. svn co https://dist.apache.org/repos/dist/release/cassandra/ cassandra-dist
  2. svn rm <previous_version> debian/pool/main/c/cassandra/<previous_version>*
  3. svn st
  4. # check and commit