Skip to content

Latest commit

 

History

History
288 lines (198 loc) · 15.3 KB

docker.rst

File metadata and controls

288 lines (198 loc) · 15.3 KB
Raw

Running Airflow in Docker

This quick-start guide will allow you to quickly start Airflow with CeleryExecutor </executor/celery> in Docker. This is the fastest way to start Airflow.

Before you begin

Follow these steps to install the necessary tools.

  1. Install Docker Community Edition (CE) on your workstation. Depending on the OS, you may need to configure your Docker instance to use 4.00 GB of memory for all containers to run properly. Please refer to the Resources section if using Docker for Windows or Docker for Mac for more information.
  2. Install Docker Compose v1.29.1 and newer on your workstation.

Older versions of docker-compose do not support all features required by docker-compose.yaml file, so double check that it meets the minimum version requirements.

Warning

Default amount of memory available for Docker on MacOS is often not enough to get Airflow up and running. If you have not enough memory available it might lead to airflow webserver continuously restarting. You should have at least 4GB memory allocated for the Docker Engine (ideally 8GB). You can check and change the amount of memory in Resources

You can also check if you have enough memory by running this command:

docker run --rm "debian:buster-slim" bash -c 'numfmt --to iec $(echo $(($(getconf _PHYS_PAGES) * $(getconf PAGE_SIZE))))'

docker-compose.yaml

To deploy Airflow on Docker Compose, you should fetch docker-compose.yaml.

quick_start_ctx

curl -LfO '{{ doc_root_url }}docker-compose.yaml'

This file contains several service definitions:

  • airflow-scheduler - The scheduler </concepts/scheduler> monitors all tasks and DAGs, then triggers the task instances once their dependencies are complete.
  • airflow-webserver - The webserver available at http://localhost:8080.
  • airflow-worker - The worker that executes the tasks given by the scheduler.
  • airflow-init - The initialization service.
  • flower - The flower app for monitoring the environment. It is available at http://localhost:5555.
  • postgres - The database.
  • redis - The redis - broker that forwards messages from scheduler to worker.

All these services allow you to run Airflow with CeleryExecutor </executor/celery>. For more information, see /concepts/overview.

Some directories in the container are mounted, which means that their contents are synchronized between your computer and the container.

  • ./dags - you can put your DAG files here.
  • ./logs - contains logs from task execution and scheduler.
  • ./plugins - you can put your custom plugins </plugins> here.

This file uses the latest Airflow image (apache/airflow). If you need install a new Python library or system library, you can build your image <docker-stack:index>.

Initializing Environment

Before starting Airflow for the first time, You need to prepare your environment, i.e. create the necessary files, directories and initialize the database.

On Linux, the mounted volumes in container use the native Linux filesystem user/group permissions, so you have to make sure the container and host computer have matching file permissions.

mkdir ./dags ./logs ./plugins
echo -e "AIRFLOW_UID=$(id -u)\nAIRFLOW_GID=0" > .env

SeeDocker Compose environment variables <docker-compose-env-variables>

On all operating systems, you need to run database migrations and create the first user account. To do it, run.

docker-compose up airflow-init

After initialization is complete, you should see a message like below.

airflow-init_1 | Upgrades done airflow-init_1 | Admin user airflow created airflow-init_1 | start_airflow-init_1 exited with code 0

The account created has the login airflow and the password airflow.

Running Airflow

Now you can start all services:

docker-compose up

In the second terminal you can check the condition of the containers and make sure that no containers are in unhealthy condition:

$ docker ps
CONTAINER ID   IMAGE            |version-spacepad| COMMAND                  CREATED          STATUS                    PORTS                              NAMES
247ebe6cf87a   apache/airflow:|version|   "/usr/bin/dumb-init …"   3 minutes ago    Up 3 minutes (healthy)    8080/tcp                           compose_airflow-worker_1
ed9b09fc84b1   apache/airflow:|version|   "/usr/bin/dumb-init …"   3 minutes ago    Up 3 minutes (healthy)    8080/tcp                           compose_airflow-scheduler_1
65ac1da2c219   apache/airflow:|version|   "/usr/bin/dumb-init …"   3 minutes ago    Up 3 minutes (healthy)    0.0.0.0:5555->5555/tcp, 8080/tcp   compose_flower_1
7cb1fb603a98   apache/airflow:|version|   "/usr/bin/dumb-init …"   3 minutes ago    Up 3 minutes (healthy)    0.0.0.0:8080->8080/tcp             compose_airflow-webserver_1
74f3bbe506eb   postgres:13      |version-spacepad| "docker-entrypoint.s…"   18 minutes ago   Up 17 minutes (healthy)   5432/tcp                           compose_postgres_1
0bd6576d23cb   redis:latest     |version-spacepad| "docker-entrypoint.s…"   10 hours ago     Up 17 minutes (healthy)   0.0.0.0:6379->6379/tcp             compose_redis_1

Accessing the environment

After starting Airflow, you can interact with it in 3 ways;

  • by running CLI commands </usage-cli>.
  • via a browser using the web interface </ui>.
  • using the REST API </stable-rest-api-ref>.

Running the CLI commands

You can also run CLI commands </usage-cli>, but you have to do it in one of the defined airflow-* services. For example, to run airflow info, run the following command:

docker-compose run airflow-worker airflow info

If you have Linux or Mac OS, you can make your work easier and download a optional wrapper scripts that will allow you to run commands with a simpler command.

quick_start_ctx

curl -LfO '{{ doc_root_url }}airflow.sh'
chmod +x airflow.sh

Now you can run commands easier.

./airflow.sh info

You can also use bash as parameter to enter interactive bash shell in the container or python to enter python container.

./airflow.sh bash
./airflow.sh python

Accessing the web interface

Once the cluster has started up, you can log in to the web interface and try to run some tasks.

The webserver available at: http://localhost:8080. The default account has the login airflow and the password airflow.

Sending requests to the REST API

Basic username password authentication <https://tools.ietf.org/html/rfc7617 https://en.wikipedia.org/wiki/Basic_access_authentication> is currently supported for the REST API, which means you can use common tools to send requests to the API.

The webserver available at: http://localhost:8080. The default account has the login airflow and the password airflow.

Here is a sample curl command, which sends a request to retrieve a pool list:

ENDPOINT_URL="http://localhost:8080/"
curl -X GET  \
    --user "airflow:airflow" \
    "${ENDPOINT_URL}/api/v1/pools"

Cleaning up

To stop and delete containers, delete volumes with database data and download images, run:

docker-compose down --volumes --rmi all

FAQ: Frequently asked questions

ModuleNotFoundError: No module named 'XYZ'

The Docker Compose file uses the latest Airflow image (apache/airflow). If you need install a new Python library or system library, you can customize and extend it <docker-stack:index>.

What's Next?

From this point, you can head to the /tutorial section for further examples or the /howto/index section if you're ready to get your hands dirty.

Environment variables supported by Docker Compose

Do not confuse the variable names here with the build arguments set when image is built. The AIRFLOW_UID and AIRFLOW_GID build args default to 50000 when the image is built, so they are "baked" into the image. On the other hand, the environment variables below can be set when the container is running, using - for example - result of id -u command, which allows to use the dynamic host runtime user id which is unknown at the time of building the image.

Variable Description Default
AIRFLOW_IMAGE_NAME Airflow Image to use. apache/airflow:
AIRFLOW_UID UID of the user to run Airflow containers as. Override if you want to use use non-default Airflow UID (for example when you map folders from host, it should be set to result of id -u call. If you change it from default 50000, you must set AIRFLOW_GID to 0. When it is changed, a 2nd user with the UID specified is dynamically created with default name inside the container and home of the use is set to /airflow/home/ in order to share Python libraries installed there. This is in order to achieve the OpenShift compatibility. See more in the Arbitrary Docker User <arbitrary-docker-user> 50000
AIRFLOW_GID Group ID in Airflow containers. It overrides the GID of the user. It is 50000 by default but if you want to use different UID than default it must be set to 0. 50000

Those additional variables are useful in case you are trying out/testing Airflow installation via docker compose. They are not intended to be used in production, but they make the environment faster to bootstrap for first time users with the most common customizations.

Variable Description Default
_AIRFLOW_WWW_USER_USERNAME Username for the administrator UI account. If this value is specified, admin UI user gets created automatically. This is only useful when you want to run Airflow for a test-drive and want to start a container with embedded development database. airflow
_AIRFLOW_WWW_USER_PASSWORD Password for the administrator UI account. Only used when _AIRFLOW_WWW_USER_USERNAME set. airflow
_PIP_ADDITIONAL_REQUIREMENTS If not empty, airflow containers will attempt to install requirements specified in the variable. example: lxml==4.6.3 charset-normalizer==1.4.1. Available in Airflow image 2.1.1 and above.