jmckenzie-dev commented on code in PR #2554: URL: https://github.com/apache/cassandra/pull/2554#discussion_r1287728520
########## .build/config/cassandra_ci.yaml: ########## @@ -0,0 +1,342 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +# +# Contains definitions of all pipelines and jobs (test suites) in Apache Cassandra's CI. +# + +# We have 2 concepts in CI: +# 1. Pipelines: a collection of test suites to be run +# 2. Jobs: a set of tests to be run with a corresponding command that executes said testing +# +# We also have a variety of parameterizable settings in tests that need to be exercised to confirm different configurations +# both work in isolation and work when combined with other params. + + +#----------------------------------------------------------------------------- +# PIPELINES +#----------------------------------------------------------------------------- +pipelines: + # All jobs in the pre-commit pipeline must run within constraints and pass + # before a commit is merged upstream. Committers are expected to validate + # and sign off on this if using non-reference CI environments. + # + # Failure to do so will lead to commits being reverted. + - name: pre-commit + jobs: + - unit + - jvm-dtest + - jvm-dtest-upgrade + - dtest + - dtest-large + - dtest-upgrade + - dtest-upgrade-large + - long-test + - cqlsh-test + + # The post-commit pipeline is a larger set of tests that will be matrixed + # across other configuration axes (see parameters below) where we expect + # different implementations or configurations to only fail rarely if at all. + # Failures in tests here will be made visible on JIRA tickets shortly after + # test run on reference CI and committers are expected to prioritize + # rectifying any failures introduced by their work. + - name: post-commit + jobs: + - unit + - cdc + - compression + - test-oa + - test-system-keyspace-directory + - test-tries + - jvm-dtest + - jvm-dtest-upgrade + - dtest + - dtest-novnode + - dtest-offheap + - dtest-large + - dtest-large-novnode + - dtest-upgrade + - dtest-upgrade-large + - long-test + - cqlsh-test + + # These are longer-term, much more rarely changing pieces of infrastructure or + # testing. We expect these to fail even more rarely than post-commit. + - name: nightly + jobs: + - stress-test + - fqltool-test + - test-burn + +#----------------------------------------------------------------------------- +# JOBS +# +# Parameters: +# job: the name +# runtime_env: which runtime_env above to base self off +# description: text description of the job +# cmd: command to run in shell to execute tests +# testlist: command to run in shell to generate full list of tests to run. Expected to output filename value to TESTLIST= env var. +# test_timeout: timeout for the test in ms +# multiplex: Number of times to multiplex a test of the given suite on addition or change +# +# Required env vars: +# DIST_DIR=<C*/build dir> +# CASSANDRA_DIR=<C* dir> +# TESTLIST=<populated by testlist: tag entry on a given job> +# JOB=<populated by job: tag entry on a given job> +# +# Jobs can replace any base params defined in their resource class, and map +# types will append new or overwriting keys. +# +#----------------------------------------------------------------------------- +# RUNTIME ENVIRONMENTS +# i.e. common resource classes +#----------------------------------------------------------------------------- +runtime_env: + - type: jvm + resources: {cpu=4, memory=4g, storage=25g} + env: + ANT_TEST_OPTS: "-Dno-build-test=true" + testlist: find test/unit -name '*Test.java' | sed "s;^test/unit/;;g" | sort + test_timeout: 480000 + multiplex: 500 + + - type: jvm-dtest + resources: {cpu=4, memory=16g, storage=25g} + env: + ANT_TEST_OPTS: "-Dno-build-test=true" + test_timeout: 900000 + multiplex: 500 + + - type: python + resources: {cpu=4, memory=6g, storage=50g} + testlist: echo "WARNING! You have to define a testlist entry for this job." && exit 1 + test_timeout: 900000 + multiplex: 500 + cmd: pytest ${PYTEST_OPTS} --cassandra-dir=${CASSANDRA_DIR} --keep-failed-test-dir ${DTEST_ARGS} ${TESTLIST} + env: + PYTHONIOENCODING: "utf-8" + PYTHONUNBUFFERED: true + CASS_DRIVER_NO_EXTENSIONS: true + CASS_DRIVER_NO_CYTHON: true + CCM_MAX_HEAP_SIZE: "1024M" + CCM_HEAP_NEWSIZE: "512M" + CCM_CONFIG_DIR: ${DIST_DIR}/.ccm + NUM_TOKENS: "16" + CASSANDRA_SKIP_SYNC: true + TMPDIR: "$(mktemp -d /tmp/run-python-dtest.XXXXXX)" + DTEST_ARGS: "--use-vnodes --num-tokens=${NUM_TOKENS} --skip-resource-intensive-tests" + # TODO: Confirm usage of --junit-prefix; see if we can just use the job name below or if these are specific (i.e. logging or control) + PYTEST_OPTS: "-vv --log-cli-level=DEBUG --junit-xml=${DIST_DIR}/test/output/nosetests.xml --junit-prefix=${JOB} -s" + + +#----------------------------------------------------------------------------- +# JVM based single node tests +#----------------------------------------------------------------------------- +test_suites: + - job: unit + runtime_env: jvm + description: Runs unit tests with default configuration + cmd: ant testclasslist ${TESTLIST} + + - job: cdc + runtime_env: jvm + description: Runs test suites with CDC enabled + cmd: ant testclasslist-cdc ${TESTLIST} + + - job: compression + runtime_env: jvm + description: Runs test suites with SSTable compression enabled + cmd: ant testclasslist-compression ${TESTLIST} + + - job: test-oa + runtime_env: jvm + description: Tests oa sstable format + cmd: ant testclasslist-oa ${TESTLIST} + + - job: test-system-keyspace-directory + runtime_env: jvm + description: Execute unit tests with a system keyspaces directory configured + + - job: test-tries + runtime_env: jvm + description: Execute unit tests with trie memtables + cmd: ant testclasslist-trie ${TESTLIST} + + + #----------------------------------------------------------------------------- + # JVM based multi-node tests + #----------------------------------------------------------------------------- + - job: jvm-dtest + runtime_env: jvm-dtest + description: Execute in-jvm dtests excepting upgrade tests + testlist: find test/distributed -name '*Test.java' | sed "s;^test/distributed/;;g" | grep -v "upgrade" sort + cmd: ant testclasslist -Dtest.classlistprefix=distributed -Dtest.timeout=${test_timeout} -Dtest.classlistfile=${testlist} ${ANT_TEST_OPTS} + multiplex: 500 + test_timeout: 900000 + + - job: jvm-dtest-upgrade + runtime_env: jvm-dtest + description: Execute upgrade in-jvm dtests + testlist: find test/distributed -name '*Test.java' | sed "s;^test/distributed/;;g" | grep "upgrade" sort + multiplex: 100 + cmd: ant test-jvm-upgrade-dtest ${ANT_TEST_OPTS} + + + #----------------------------------------------------------------------------- + # Python based tests + #----------------------------------------------------------------------------- + - job: dtest + type: python + description: Python based multi-node distributed tests with default configuration + + - job: dtest-novnode + runtime_env: python + description: Python based multi-node distributed tests without vnodes + env: + DTEST_ARGS: "--skip-resource-intensive-tests --keep-failed-test-dir" + + - job: dtest-offheap + runtime_env: python + description: Python based multi-node distributed tests with off-heap memtables + env: + DTEST_ARGS: "--skip-resource-intensive-tests --keep-failed-test-dir" + + - job: dtest-large + runtime_env: python + resources: {cpu=4, memory=16g, storage=32g} + description: Python based multi-node distributed tests with tests flagged as too large for the regular suite + env: + DTEST_ARGS: "--use-vnodes --num-tokens=${NUM_TOKENS} --only-resource-intensive-tests --force-resource-intensive-tests" + + - job: dtest-large-novnode + runtime_env: python + description: Python based multi-node distributed tests with tests flagged as too large for the regular suite with vnodes disabled + env: + DTEST_ARGS: "--only-resource-intensive-tests --force-resource-intensive-tests" + + - job: dtest-upgrade + parent: dtest + description: Python based multi-node distributed upgrade tests + multiplex: 25 + env: + DTEST_ARGS: "--use-vnodes --num-tokens=${NUM_TOKENS} --execute-upgrade-tests --execute-upgrade-tests-only --upgrade-target-version-only --upgrade-version-selection all" + + - job: dtest-upgrade-large + parent: dtest + description: Python based multi-node distributed upgrade tests flagged as too large for the regular upgrade suite + env: + DTEST_ARGS: "--use-vnodes --num-tokens=${NUM_TOKENS} --execute-upgrade-tests --execute-upgrade-tests-only --upgrade-target-version-only --upgrade-version-selection all --only-resource-intensive-tests --force-resource-intensive-tests" + + #----------------------------------------------------------------------------- + # Longer running soak / burn / benchmarking tests + #----------------------------------------------------------------------------- + - job: stress-test + runtime_env: jvm + resources: {cpu=4, memory=1g, storage=5g} + description: Runs stress tests + + - job: fqltool-test + runtime_env: jvm + resources: {cpu=4, memory=1g, storage=5g} + description: Tests to ensure cassandra-stress is operating correctly + + - job: microbench + runtime_env: jvm + resources: {cpu=4, memory=1g, storage=5g} + description: Microbenchmarks executing specific portions of the codebase + + - job: test-burn + runtime_env: jvm + resources: {cpu=4, memory=16g, storage=5g} + description: Long running and fuzz tests + test_timeout: 60000000 + + - job: long-test + runtime_env: jvm + resources: {cpu=4, memory=6g, storage=10g} + description: Execute long-running tests that are beyond the normal suite timeout + test_timeout: 600000 + multiplex: 100 + + - job: cqlsh-test + runtime_env: jvm + resources: {cpu=4, memory=1g, storage=5g} + description: Subset of tests exercising cqlsh only + cmd: ./pylib/cassandra-cqlsh-tests.sh $(pwd) + +#----------------------------------------------------------------------------- +# PARAMETERS +#----------------------------------------------------------------------------- +# Since JVM tests are generally quicker, we can afford to multiplex across more params +jdk_parameters: + - param: number + values: + - 11 + - 17 + - param: offheap + values: + - none + - objects + - all + - param: compression + values: + - on + - off + - param: memtables + values: + - trie + - btree + - param: sstables + values: + - bigtable + - tries + +# We can't afford to re-run all the python dtests across all the different +# possible combinations above. We stick with just JDK as defined above and +# vnode vs. novnode. +python_parameters: + - params: num_tokens + values: + - 0 + - 16 + +#----------------------------------------------------------------------------- +# Commands that can be pulled and run in pipelines +#----------------------------------------------------------------------------- +commands: Review Comment: Hah. Well, this was still in the circle config so I assumed it was something we relied on and could make use of. Very happy to delete this. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
