- local
Continuous Integration is important component of making Apache Airflow robust and stable. We are running a lot of tests for every pull request, for main and v2-*-test branches and regularly as CRON jobs.
Our execution environment for CI is GitHub Actions. GitHub Actions (GA) are very well integrated with GitHub code and Workflow and it has evolved fast in 2019/202 to become a fully-fledged CI environment, easy to use and develop for, so we decided to switch to it. Our previous CI system was Travis CI.
However part of the philosophy we have is that we are not tightly coupled with any of the CI environments we use. Most of our CI jobs are written as bash scripts which are executed as steps in the CI jobs. And we have a number of variables determine build behaviour.
Our builds on CI are highly optimized. They utilise some of the latest features provided by GitHub Actions environment that make it possible to reuse parts of the build process across different Jobs.
Big part of our CI runs use Container Images. Airflow has a lot of dependencies and in order to make sure that we are running tests in a well configured and repeatable environment, most of the tests, documentation building, and some more sophisticated static checks are run inside a docker container environment. This environment consist of two types of images: CI images and PROD images. CI Images are used for most of the tests and checks where PROD images are used in the Kubernetes tests.
In order to run the tests, we need to make sure that the images are built using latest sources and that it is done quickly (full rebuild of such image from scratch might take ~15 minutes). Therefore optimisation techniques have been implemented that use efficiently cache from the GitHub Docker registry - in most cases this brings down the time needed to rebuild the image to ~4 minutes. In some cases (when dependencies change) it can be ~6-7 minutes and in case base image of Python releases new patch-level, it can be ~12 minutes.
For the CI builds of our we are using Container Registry to store results of the "Build Image" workflow and pass it to the "CI Build" workflow.
Currently in main version of Airflow we run tests in 4 different versions of Python (3.6, 3.7, 3.8, 3.9) which means that we have to build 6 images (3 CI ones and 3 PROD ones). Yet we run around 12 jobs with each of the CI images. That is a lot of time to just build the environment to run. Therefore we are utilising pull_request_target feature of GitHub Actions.
This feature allows to run a separate, independent workflow, when the main workflow is run -this separate workflow is different than the main one, because by default it runs using main version of the sources but also - and most of all - that it has WRITE access to the repository.
This is especially important in our case where Pull Requests to Airflow might come from any repository, and it would be a huge security issue if anyone from outside could utilise the WRITE access to Apache Airflow repository via an external Pull Request.
Thanks to the WRITE access and fact that the 'pull_request_target' by default uses the 'main' version of the sources, we can safely run some logic there will checkout the incoming Pull Request, build the container image from the sources from the incoming PR and push such image to an GitHub Docker Registry - so that this image can be built only once and used by all the jobs running tests. The image is tagged with unique COMMIT_SHA of the incoming Pull Request and the tests run in the Pull Request can simply pull such image rather than build it from the scratch. Pulling such image takes ~ 1 minute, thanks to that we are saving a lot of precious time for jobs.
We can use either of the two available GitHub Container registries as cache:
- Legacy GitHub Package Registry which is not very stable, uses old infrastructure of GitHub and it lacks certain features - notably it does not allow us to delete the old image. The benefit of using GitHub Package Registry is that it works out-of-the-box (write authentication is done using
GITHUB_TOKENand users do not have to do any action to make it work in case they want to run build using their own forks. Also those images do not provide public access, so you need to login todocker.pkg.github.comdocker registry using your username and personal token to be able to pull those images. - The new GitHub Container Registry which is in Public Beta, has many more features (including permission management, public access and image retention possibility). Similarly as in case of GitHub Package Registry
GITHUB_TOKENis needed to push to the repositories. You also have to manually manage permissions of the images, i.e. after creating images for the first time, you need to set their visibility toPublicand addAdminpermissions to group of people managing the images (in our caseairflow-committersgroup). This makes it not very suitable to use GitHub container registry if you want to run builds of Airflow in your own forks (note - it does not affect pull requests from forks to Airflow).
Those two images have different naming schemas. See Images documentation for details.
You can interact with the GitHub Registry images (pull/push) via Breeze - you can pass --github-registry flag with either docker.pkg.github.com for GitHub Package Registry or ghcr.io for GitHub Container Registry and pull/push operations will be performed using the chosen registry, using appropriate naming convention. This allows building and pushing the images locally by committers who have access to push/pull those images.
The main goal of the CI philosophy we have that no matter how complex the test and integration infrastructure, as a developer you should be able to reproduce and re-run any of the failed checks locally. One part of it are pre-commit checks, that allow you to run the same static checks in CI and locally, but another part is the CI environment which is replicated locally with Breeze.
You can read more about Breeze in BREEZE.rst but in essence it is a script that allows you to re-create CI environment in your local development instance and interact with it. In its basic form, when you do development you can run all the same tests that will be run in CI - but locally, before you submit them as PR. Another use case where Breeze is useful is when tests fail on CI. You can take the full COMMIT_SHA of the failed build pass it as --github-image-id parameter of Breeze and it will download the very same version of image that was used in CI and run it locally. This way, you can very easily reproduce any failed test that happens in CI - even if you do not check out the sources connected with the run.
You can read more about it in BREEZE.rst and TESTING.rst
Depending whether the scripts are run locally (most often via Breeze) or whether they are run in "CI Build" or "Build Image" workflows they can take different values.
You can use those variables when you try to reproduce the build locally.
| Variable | Local development | Build Image CI workflow | Main CI Workflow | Comment |
|---|---|---|---|---|
|
bles | |||
PYTHON_MAJOR_MINOR_VERSION |
Major/Minor version of Python used. | |||
DB_RESET |
|
|
|
Determines whether database should be reset at the container entry. By default locally the database is not reset, which allows to keep the database content between runs in case of Postgres or MySQL. However, it requires to perform manual init/reset if you stop the environment. |
| Dockerhub variables | ||||
DOCKERHUB_USER |
|
Name of the DockerHub user to use | ||
DOCKERHUB_REPO |
|
Name of the DockerHub repository to use | ||
|
bles | |||
MOUNT_SELECTED_LOCAL_SOURCES |
|
|
|
Determines whether local sources are mounted to inside the container. Useful for local development, as changes you make locally can be immediately tested in the container. We mount only selected, important folders. We do not mount the whole project folder in order to avoid accidental use of artifacts (such as egg-info directories) generated locally on the host during development. |
MOUNT_ALL_LOCAL_SOURCES |
|
|
|
Determines whether all local sources are mounted to inside the container. Useful for local development when you need to access .git folders and other folders excluded when MOUNT_SELECTED_LOCAL_SOURCES is true. You might need to manually delete egg-info folder when you enter breeze and the folder was generated using different Python versions. |
|
bles | |||
FORCE_PULL_IMAGES |
|
|
|
Determines if images are force-pulled, no matter if they are already present locally. This includes not only the CI/PROD images but also the Python base images. Note that if Python base images change, also the CI and PROD images need to be fully rebuild unless they were already built with that base Python image. This is false for local development to avoid often pulling and rebuilding the image. It is true for CI workflow in case waiting from images is enabled as the images needs to be force-pulled from GitHub Registry, but it is set to false when waiting for images is disabled. |
FORCE_BUILD_IMAGES |
|
|
|
Forces building images. This is generally not very useful in CI as in CI environment image is built or pulled only once, so there is no need to set the variable to true. For local builds it forces rebuild, regardless if it is determined to be needed. |
FORCE_ANSWER_TO_QUESTIONS |
|
|
This variable determines if answer to questions during the build process should be automatically given. For local development, the user is occasionally asked to provide answers to questions such as - whether the image should be rebuilt. By default the user has to answer but in the CI environment, we force "yes" answer. | |
SKIP_CHECK_REMOTE_IMAGE |
|
|
|
Determines whether we check if remote image is "fresher" than the current image. When doing local breeze runs we try to determine if it will be faster to rebuild the image or whether the image should be pulled first from the cache because it has been rebuilt. This is slightly experimental feature and will be improved in the future as the current mechanism does not always work properly. |
|
les | |||
HOST_USER_ID |
User id of the host user. | |||
HOST_GROUP_ID |
Group id of the host user. | |||
HOST_OS |
|
|
OS of the Host (Darwin/Linux). | |
HOST_HOME |
Home directory on the host. | |||
|
rsion suffix v | ariables | ||
VERSION_SUFFIX_FOR_PYPI |
Version suffix used during provider package preparation for PyPI builds. | |||
|
les | |||
| COMMIT_SHA | GITHUB_SHA | GITHUB_SHA | SHA of the commit of the build is run | |
| Verbosity var | iables | |||
PRINT_INFO_FROM_SCRIPTS |
|
|
|
Allows to print output to terminal from running scripts. It prints some extra outputs if true including what the commands do, results of some operations, summary of variable values, exit status from the scripts, outputs of failing commands. If verbose is on it also prints the commands executed by docker, kind, helm, kubectl. Disabled in pre-commit checks. * set to false in pre-commits |
VERBOSE |
|
|
|
Determines whether docker, helm, kind, kubectl commands should be printed before execution. This is useful to determine what exact commands were executed for debugging purpose as well as allows to replicate those commands easily by copy&pasting them from the output. requires PRINT_INFO_FROM_SCRIPTS set to true. |
VERBOSE_COMMANDS |
|
|
|
Determines whether every command executed in bash should also be printed before execution. This is a low-level debugging feature of bash (set -x) and it should only be used if you are lost at where the script failed. |
| Image build va | riables | |||
UPGRADE_TO_NEWER_DEPENDENCIES |
|
|
|
Determines whether the build should attempt to upgrade Python base image and all PIP dependencies to latest ones matching Setting the value to commit SHA is best way to assure that constraints are upgraded even if there is no change to setup.py This way our constraints are automatically tested and updated whenever new versions of libraries are released.
|
CHECK_IMAGE_FOR_REBUILD |
|
|
|
Determines whether attempt should be made to rebuild the CI image with latest sources. It is true by default for local builds, however it is set to true in case we know that the image we pulled or built already contains the right sources. In such case we should set it to false, especially in case our local sources are not the ones we intend to use (for example when In CI builds it is set to true in case of the "Build Image" workflow or when waiting for images is disabled in the CI workflow.
|
SKIP_BUILDING_PROD_IMAGE |
|
|
|
Determines whether we should skip building the PROD image with latest sources. It is set to false, but in deploy app for kubernetes step it is set to "true", because at this stage we know we have good image build or pulled.
|
The following variables are automatically determined based on CI environment variables. You can locally by setting CI="true" and run the ci scripts from the scripts/ci folder:
provider_packages- scripts to build and test provider packagesconstraints- scripts to build and publish latest set of valid constraintsdocs- scripts to build documentationimages- scripts to build and push CI and PROD imageskubernetes- scripts to setup kubernetes cluster, deploy airflow and run kubernetes tests with ittesting- scripts that run unit and integration teststools- scripts that perform various clean-up and preparation tasks
Common libraries of functions for all the scripts can be found in libraries folder.
For detailed use of those scripts you can refer to .github/workflows/ - those scripts are used by the CI workflows of ours.
The default values are "sane" you can change them to interact with your own repositories or registries. Note that you need to set "CI" variable to true in order to get the same results as in CI.
| Variable | Default | Comment |
|---|---|---|
| CI | false |
If set to "true", we simulate behaviour of all scripts as if they are in CI environment |
| CI_TARGET_REPO | apache/airflow |
Target repository for the CI build. Used to compare incoming changes from PR with the target. |
| CI_TARGET_BRANCH | main |
Target branch where the PR should land. Used to compare incoming changes from PR with the target. |
| CI_BUILD_ID | 0 |
Unique id of the build that is kept across re runs (for GitHub actions it is GITHUB_RUN_ID) |
| CI_JOB_ID | 0 |
Unique id of the job - used to produce unique artifact names. |
| CI_EVENT_TYPE | pull_request |
Type of the event. It can be one of [pull_request, pull_request_target, schedule, push] |
| CI_REF | refs/head/main |
Branch in the source repository that is used to make the pull request. |
Our CI uses GitHub Registry to pull and push images to/from by default. You can however make it interact with DockerHub registry or change the GitHub registry to interact with and use your own repo by changing GITHUB_REPOSITORY and providing your own GitHub Username and Token.
Currently we are using GitHub Packages to cache images for the build. GitHub Packages are "legacy" storage of binary artifacts for GitHub and as of September 2020 they introduced GitHub Container Registry as more stable, easier to manage replacement for container storage. It includes complete self-management of the images including permission management, public access, retention management and many more.
More about it here:
https://github.blog/2020-09-01-introducing-github-container-registry/
Recently we started to experience unstable behaviour of the GitHub Packages ('unknown blob' and manifest v1 vs. v2 when pushing images to it). So together with ASF we proposed to enable GitHub Container Registry and it happened as of January 2020.
More about it in https://issues.apache.org/jira/browse/INFRA-20959
We are currently in the testing phase, especially when it comes to management of permissions -the model of permission management is not the same for Container Registry as it was for GitHub Packages (it was per-repository in GitHub Packages, but it is organization-wide in the Container Registry.
| Variable | Default | Comment |
|---|---|---|
| USE_GITHUB_REGISTRY | true | If set to "true", we interact with GitHub Registry registry not the DockerHub one. |
| GITHUB_REGISTRY | ghcr.io |
Name of the GitHub registry to use. Can be docker.pkg.github.com or ghcr.io |
| GITHUB_REPOSITORY | apache/airflow |
Prefix of the image. It indicates which. registry from GitHub to use |
| GITHUB_USERNAME | Username to use to login to GitHub | |
| GITHUB_TOKEN | Token to use to login to GitHub. This token is automatically set by GitHub CI to a to a READ-only token for PR builds from fork and to WRITE token for direct pushes and scheduled or workflow_run types of builds | |
| GITHUB_REGISTRY_WAIT_FOR_IMAGE | false |
Wait for the image to be available. This is useful if commit SHA is used as pull tag |
| GITHUB_REGISTRY_PULL_IMAGE_TAG | latest |
Pull this image tag. This is "latest" by default, can also be full-length commit SHA. |
| GITHUB_REGISTRY_PUSH_IMAGE_TAG | latest |
Pull this image tag. This is "latest" by default, can also be full-length commit SHA. |
We are currently in the process of testing using GitHub Container Registry as cache for our images during the CI process. The default registry is set to "GitHub Packages", but we are testing the GitHub Container Registry. In case of GitHub Packages, authentication uses GITHUB_TOKEN mechanism. Authentication is needed for both pushing the images (WRITE) and pulling them (READ) - which means that GitHub token is used in "main" build (WRITE) and in fork builds (READ). For container registry, our images are Publicly Visible and we do not need any authentication to pull them.
If USE_GITHUB_REGISTRY is set to "false" you can interact directly with DockerHub. By default you pull from/push to "apache/airflow" DockerHub repository, but you can change that to your own repository by setting those environment variables:
| Variable | Default | Comment |
|---|---|---|
| DOCKERHUB_USER | apache |
Name of the DockerHub user to use |
| DOCKERHUB_REPO | airflow-ci |
Name of the DockerHub repo to use |
The following components are part of the CI infrastructure
- Apache Airflow Code Repository - our code repository at https://github.com/apache/airflow
- Apache Airflow Forks - forks of the Apache Airflow Code Repository from which contributors make Pull Requests
- GitHub Actions - (GA) UI + execution engine for our jobs
- GA CRON trigger - GitHub Actions CRON triggering our jobs
- GA Workers - virtual machines running our jobs at GitHub Actions (max 20 in parallel)
- GitHub Private Image Registry- image registry used as build cache for CI jobs. It is at https://ghcr.io/apache/airflow/airflow
- DockerHub Public Image Registry - publicly available image registry at DockerHub. It is at https://hub.docker.com/r/apache/airflow-ci
- DockerHub Build Workers - virtual machines running build jibs at DockerHub
- Official Images (future) - these are official images that are prominently visible in DockerHub. We aim our images to become official images so that you will be able to pull them with
docker pull apache-airflow
The following CI Job run types are currently run for Apache Airflow (run by ci.yaml workflow) and each of the run types has different purpose and context.
Those runs are results of PR from the forks made by contributors. Most builds for Apache Airflow fall into this category. They are executed in the context of the "Fork", not main Airflow Code Repository which means that they have only "read" permission to all the GitHub resources (container registry, code repository). This is necessary as the code in those PRs (including CI job definition) might be modified by people who are not committers for the Apache Airflow Code Repository.
The main purpose of those jobs is to check if PR builds cleanly, if the test run properly and if the PR is ready to review and merge. The runs are using cached images from the Private GitHub registry -CI, Production Images as well as base Python images that are also cached in the Private GitHub registry. Also for those builds we only execute Python tests if important files changed (so for example if it is "no-code" change, no tests will be executed.
The workflow involved in Pull Requests review and approval is a bit more complex than simple workflows in most of other projects because we've implemented some optimizations related to efficient use of queue slots we share with other Apache Software Foundation projects. More details about it can be found in PULL_REQUEST_WORKFLOW.rst.
Those runs are results of direct pushes done by the committers or as result of merge of a Pull Request by the committers. Those runs execute in the context of the Apache Airflow Code Repository and have also write permission for GitHub resources (container registry, code repository). The main purpose for the run is to check if the code after merge still holds all the assertions - like whether it still builds, all tests are green.
This is needed because some of the conflicting changes from multiple PRs might cause build and test failures after merge even if they do not fail in isolation. Also those runs are already reviewed and confirmed by the committers so they can be used to do some housekeeping: - pushing most recent image build in the PR to the GitHub Private Registry (for caching) - upgrading to latest constraints and pushing those constraints if all tests succeed - refresh latest Python base images in case new patch-level is released
The housekeeping is important - Python base images are refreshed with varying frequency (once every few months usually but sometimes several times per week) with the latest security and bug fixes. Those patch level images releases can occasionally break Airflow builds (specifically Docker image builds based on those images) therefore in PRs we only use latest "good" Python image that we store in the private GitHub cache. The direct push/main builds are not using registry cache to pull the Python images - they are directly pulling the images from DockerHub, therefore they will try the latest images after they are released and in case they are fine, CI Docker image is build and tests are passing -those jobs will push the base images to the private GitHub Registry so that they be used by subsequent PR runs.
Those runs are results of (nightly) triggered job - only for main branch. The main purpose of the job is to check if there was no impact of external dependency changes on the Apache Airflow code (for example transitive dependencies released that fail the build). It also checks if the Docker images can be build from the scratch (again - to see if some dependencies have not changed - for example downloaded package releases etc. Another reason for the nightly build is that the builds tags most recent main with nightly-main tag so that DockerHub build can pick up the moved tag and prepare a nightly public main build in the DockerHub registry. The v1-10-test branch images are build in DockerHub when pushing v1-10-stable manually.
All runs consist of the same jobs, but the jobs behave slightly differently or they are skipped in different run categories. Here is a summary of the run categories with regards of the jobs they are running. Those jobs often have matrix run strategy which runs several different variations of the jobs (with different Backend type / Python version, type of the tests to run for example). The following chapter describes the workflows that execute for each run.
Those runs and their corresponding Build Images runs are only executed in main apache/airflow repository, they are not executed in forks - we want to be nice to the contributors and not use their free build minutes on GitHub Actions.
A general note about cancelling duplicated workflows: for the Build Images, CI Build and CodeQL workflows we use the concurrency feature of GitHub actions to automatically cancel "old" workflows of each time --meaning if you push a new commit to a branch or to a pull request and there is a workflow running GitHub Actions will cancel the old job automatically.
This workflow builds images for the CI Workflow.
It's a special type of workflow: pull_request_target which means that it is triggered when a pull request is opened. This also means that the workflow has Write permission to the Airflow repository and it can - for example - push to the GitHub registry the images used by CI Builds which means that the images can be built only once and reused by all the CI jobs (including the matrix jobs). We've implemented it in the way that the CI Build running will wait until the images are built by the "Build Images" workflow.
(This workflow is also triggered on normal pushes to our "main" branches, i.e. after a pull request is merged.)
It's possible to disable this feature and go back to the previous behaviour via GITHUB_REGISTRY_WAIT_FOR_IMAGE flag in the "Build Workflow image". Setting it to "false" switches back to the behaviour that each job builds own image.
You can also switch back to jobs building the images on its own on the fork level by setting AIRFLOW_GITHUB_REGISTRY_WAIT_FOR_IMAGE secret to false. This will disable pushing the "COMMIT_SHA" images to GitHub Registry and all the images will be built locally by each job. It is about 20% slower for the whole build on average, but it does not require to have access to push images to GitHub, which sometimes might be not available (depending on the account status).
The write permission also allows to cancel workflows. It is not possible for the pull request "CI Builds" workflow run from the forks as they have no Write permission and cannot cancel running workflows. In our case we perform several different cancellations:
- last, but not least - we cancel any of the "CI Build" workflow runs that failed in some important jobs. This is another optimisations - GitHub does not have "fail-fast" on the whole run and this cancelling effectively implements "fail-fast" of runs for some important jobs. Note that it only works when you submit new PRs or push new changes. In case the jobs failed and no new PR is pushed after that, the whole run will run to completion.
The workflow has the following jobs:
| Job | Description |
|---|---|
| Build Info | Prints detailed information about the build |
| Build CI/PROD images | Builds all configured CI and PROD images |
The images are stored in the GitHub Registry and the names of those images follow the patterns described in Naming conventions for stored images
Image building is configured in "fail-fast" mode. When any of the images fails to build, it cancels other builds and the source "CI Build" workflow run that triggered it.
This workflow is a regular workflow that performs all checks of Airflow code.
| Job | Description | PR | Push Merge | CRON (1) |
|---|---|---|---|---|
| Build info | Prints detailed information about the build | Yes | Yes | Yes |
| Helm tests | Runs tests for the Helm chart | Yes | Yes | Yes |
| Test OpenAPI client gen | Tests if OpenAPIClient continues to generate | Yes | Yes | Yes |
| CI Images | Waits for CI Images (3) | Yes | Yes | Yes |
| Static checks | Performs static checks | Yes | Yes | Yes |
| Build docs | Builds documentation | Yes | Yes | Yes |
| Spell check docs | Spell check for documentation | Yes | Yes | Yes |
| Trigger tests | Checks if tests should be triggered | Yes | Yes | Yes |
| Tests [Pg/Msql/Sqlite] | Run all the Pytest tests for Python code | Yes(2) | Yes | Yes |
| Quarantined tests | Flaky tests that we need to fix (5) | Yes(2) | Yes | Yes |
| Upload coverage | Uploads test coverage from all the tests | Yes | Yes | Yes |
| PROD Images | Waits for CI Images (3) | Yes | Yes | Yes |
| Tests Kubernetes | Run Kubernetes test | Yes(2) | Yes | Yes |
| Push PROD images | Pushes PROD images to GitHub Registry (4) | Yes | ||
| Push CI images | Pushes CI images to GitHub Registry (4) | Yes | ||
| Constraints | Upgrade constraints to latest ones (4) | Yes | Yes | |
| Constraints push | Pushes all upgraded constraints (4) | Yes | Yes | |
| Tag Repo nightly | Tags the repository with nightly tag (6) | Yes |
Comments:
- CRON jobs builds images from scratch - to test if everything works properly for clean builds
- The tests are run when the Trigger Tests job determine that important files change (this allows for example "no-code" changes to build much faster)
- The jobs wait for CI images if
GITHUB_REGISTRY_WAIT_FOR_IMAGEvariable is set to "true". You can set it to "false" to disable using shared images - this is slower though as the images are rebuilt in every job that needs them. You can also set your own fork's secretAIRFLOW_GITHUB_REGISTRY_WAIT_FOR_IMAGEtofalseto trigger the same behaviour.- PROD and CI images are pushed as "latest" to DockerHub registry and constraints are upgraded only if all tests are successful. Note that images are not pushed in CRON jobs because they are rebuilt from scratch and we want to push incremental changes to the DockerHub registry.
- Flaky tests never fail in regular builds. See the next chapter where our approach to flaky tests is explained.
- Nightly tag is pushed to the repository only in CRON job and only if all tests pass. This causes the DockerHub images are built automatically and made available to developers.
This is manually triggered workflow (via GitHub UI manual run) that should only be run in GitHub forks. When triggered, it will force-push the "apache/airflow" main to the fork's main. It's the easiest way to sync your fork main to the Apache Airflow's one.
This workflow is introduced, to delete old artifacts from the GitHub Actions build. We set it to delete old artifacts that are > 7 days old. It only runs for the 'apache/airflow' repository.
We also have a script that can help to clean-up the old artifacts: remove_artifacts.sh
The CodeQL security scan uses GitHub security scan framework to scan our code for security violations. It is run for JavaScript and Python code.
Documentation from the main branch is automatically published on Amazon S3.
To make this possible, GitHub Action has secrets set up with credentials for an Amazon Web Service account - DOCS_AWS_ACCESS_KEY_ID and DOCS_AWS_SECRET_ACCESS_KEY.
This account has permission to write/list/put objects to bucket apache-airflow-docs. This bucket has public access configured, which means it is accessible through the website endpoint. For more information, see: Hosting a static website on Amazon S3
Website endpoint: http://apache-airflow-docs.s3-website.eu-central-1.amazonaws.com/
The images produced during the CI builds are stored in the GitHub Registry
The images are stored with both "latest" tag (for last main push image that passes all the tests as well with the tags indicating the origin of the image.
The image names follow the patterns:
| Image | Name pattern | Tag for format | Comment |
|---|---|---|---|
| Python image | Python | <X.Y>-slim-buster-<COMMIT_SHA> | Base Python image used by both production and CI image. Python maintainer release new versions of those image with security fixes every few weeks. |
| CI image | <BRANCH>-python<X.Y>-ci | <COMMIT_SHA> | CI image - this is the image used for most of the tests. |
| PROD Build image | <BRANCH>-python<X.Y>-build | <COMMIT_SHA> | Production Build image - this is the "build" segment of production image. It contains build-essentials and all necessary packages to install PIP packages. |
| PROD image | <BRANCH>-python<X.Y> | <COMMIT_SHA> | Production image. This is the actual production image - optimized for size. It contains only compiled libraries and minimal set of dependencies to run Airflow. |
- <BRANCH> might be either "main" or "v1-10-test" or "v2-*-test"
- <X.Y> - Python version (Major + Minor). For "main" and "v2-*-test" should be in ["3.6", "3.7", "3.8", "3.9"].
- <COMMIT_SHA> - for images that get merged to "main", "v2-*-test" of "v1-10-test", or built as part of a pull request the images are tagged with the (full length) commit SHA of that particular branch. For pull requests the SHA used is the tip of the pull request branch.
Since we store images from every CI run, you should be able easily reproduce any of the CI build problems locally. You can do it by pulling and using the right image and running it with the right docker command, For example knowing that the CI build was for commit cd27124534b46c9688a1d89e75fcd137ab5137e3:
docker pull ghcr.io/apache/airflow-main-python3.6-ci:cd27124534b46c9688a1d89e75fcd137ab5137e3
docker run -it ghcr.io/apache/airflow-main-python3.6-ci:cd27124534b46c9688a1d89e75fcd137ab5137e3But you usually need to pass more variables and complex setup if you want to connect to a database or enable some integrations. Therefore it is easiest to use Breeze for that. For example if you need to reproduce a MySQL environment with kerberos integration enabled for commit cd27124534b46c9688a1d89e75fcd137ab5137e3, in python 3.8 environment you can run:
./breeze --github-image-id cd27124534b46c9688a1d89e75fcd137ab5137e3 --github-registry ghcr.io --python 3.8You will be dropped into a shell with the exact version that was used during the CI run and you will be able to run pytest tests manually, easily reproducing the environment that was used in CI. Note that in this case, you do not need to checkout the sources that were used for that run - they are already part of the image - but remember that any changes you make in those sources are lost when you leave the image as the sources are not mapped from your host machine.
Sequence diagrams are shown of the flow happening during the CI builds.
In 2.0 line we currently support Python 3.6, 3.7, 3.8.
In order to add a new version the following operations should be done (example uses Python 3.9)
- copy the latest constraints in
constraints-mainbranch from previous versions and name it using the new Python version (constraints-3.9.txt). Commit and push - add the new Python version to breeze-complete and _initialization.sh - tests will fail if they are not in sync.
- build image locally for both prod and CI locally using Breeze:
./breeze build-image --python 3.9- push image as cache to DockerHub and both registries:
./breeze push-image --python 3.9
./breeze push-image --python 3.9 --github-registry ghcr.io
./breeze push-image --python 3.9 --github-registry docker.pkg.github.com- Find the 3 new images (main, ci, build) created in GitHub Container registry go to Package Settings and turn on
Public Visibilityand addairflow-committersgroup asAdmin Roleto all of them. - In DockerHub create three entries for automatically built nightly-tag and release images:
| Source type | Source | Docker Tag | Dockerfile location | Build Context | Autobuild | Build caching | Comment |
|---|---|---|---|---|---|---|---|
| Tag | nightly-main | main-python3.9 | Dockerfile | / | x | Nightly CI/PROD images from successful scheduled main nightly builds | |
| Branch | v2-*-stable | v2-*-stable-python3.9 | Dockerfile | / | x | CI/PROD images automatically built pushed stable branch | |
| Tag | /^([1-2].*)$/ | {1}-python3.9 | Dockerfile | / | x | CI/PROD images automatically built from pushed release tags |