Simplified Airflow CLI Tool for Lauching CeleryExecutor Deployment
Project description
Airflow Run
Python tool for deploying Airflow Multi-Node Cluster.
Requirements
Python >=3.6 (tested)
Goal
To provide a quick way to setup Airflow Multi-Node Cluster (a.k.a. Celery Executor Setup).
Steps
Generate config yaml file.
Run commands to start webserver, scheduler, worker, (rabbitmq, postgres).
Add dag files and run initdb.
Generate config file:
afr --generate_config
Running the tool in the same directory as config file:
afr --run postgresql
afr --run initdb
afr --run rabbitmq
afr --run webserver
afr --run scheduler
afr --run worker --queue {queue name}
afr --run flower
Or, running the tool specifying config path:
afr --run postgresql --config /path/config.yaml
Or, use this environment variable to set the config path:
export AIRFLOWRUN_CONFIG_PATH="/some_path/config.yaml"
After running webserver, scheduler and worker (postgres and rabbitmq if needed local instances), Add your dag files in the dags subdirectory in the directory you defined in the config file.
(* note: make sure you have the correct user permission in the dags, logs subdirectories.)
That is it!!
Default Config yaml file:
private_registry: False
registry_url: registry.hub.docker.com
username: ""
password: ""
repository: pkuong/airflow-run
image: airflow-run
tag: latest
local_dir: {local directory where you want to mount /dags and /logs folder}
webserver_port: 8000
flower_port: 5555
env:
AIRFLOW__CORE__EXECUTOR: CeleryExecutor
AIRFLOW__CORE__LOAD_EXAMPLES: "False"
AIRFLOW__CORE__DAGS_FOLDER: /usr/local/airflow/airflow/dags
AIRFLOW__CORE__LOGS_FOLDER: /usr/local/airflow/airflow/logs
AIRFLOW_HOME: /usr/local/airflow
rabbitmq:
name: rabbitmq
username: {username}
password: {password}
host: {IP}
virtual_host: /
image: rabbitmq:3-management
home: /var/lib/rabbitmq
ui_port: 15672
port: 5672
env:
RABBITMQ_DEFAULT_USER: {username}
RABBITMQ_DEFAULT_PASS: {password}
postgresql:
name: postgresql
username: {username}
password: {password}
host: {host}
image: postgres
data: /var/lib/postgresql/data
port: 5432
env:
PGDATA: /var/lib/postgresql/data/pgdata
POSTGRES_USER: {username}
POSTGRES_PASSWORD: {password}
Docker image
This tool is using the following public docker image by default.
https://hub.docker.com/repository/docker/pkuong/airflow-run
Building the image:
If you want to build your own image, you can run the following:
afd --build --config_path={absolute path to config.yaml} --dockerfile_path={absolute path to directory which contains Dockerfile}
Contributors
Paulo Kuong (@pkuong)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
airflow-run-0.2.0.tar.gz
(9.6 kB
view hashes)
Built Distribution
Close
Hashes for airflow_run-0.2.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 28dd17fd8c13ccd4e44e64b9c70288f3eed61075862d35420557fe796beadbef |
|
MD5 | 89b3ac477cd18e1a56bb62a629ad464b |
|
BLAKE2b-256 | ac901bfbfc6dd994a01663f478897be19eb7112138a326223aa6b61e1405a543 |