1. 安装

airflow的安装非常简单，安装官方文档可以轻松完成，我是在macOS上安装的，Linux下应该是一样的，Windows就不清楚了。

# airflow needs a home, ~/airflow is the default,
# but you can lay foundation somewhere else if you prefer
# (optional)
export AIRFLOW_HOME=~/airflow

# install from pypi using pip
pip install apache-airflow

# initialize the database
airflow initdb

# start the web server, default port is 8080
airflow webserver -p 8080

airflow 需要设置一个环境变量，默认路径是～/airflow，当然你也可以选择其他路径
使用pip安装
初始化数据库
启动web服务，默认端口号是8080

2.配置

运行完上面的安装命令后，airflow会默认创建环境变量$AIRFLOW_HOME对应的文件目录，并创建文件“airflow.cfg”。你可以打开$AIRFLOW_HOME/airflow.cfg或者在web端通过菜单Admin->Configuration查看文件的内容。PID文件在$AIRFLOW_HOME/airflow-webserver.pid目录下或者如果被系统共享的话在/run/airflow/webserver.pid目录下。(我的是在$AIRFLOW_HOME/airflow-webserver.pid这里)

airflow使用sqlite数据库，不需要你进行后端操作，可以快速的应用airflow。它与SequentialExecutor一起工作，它会按顺序运行任务实例。虽然功能有限，但是利用它可以快速的开始体验airflow的功能，并且可以访问UI和命令行工具。

下面是几个命令，会触发几个任务实例。运行下面的命令，能在DAG中看到job的运行状态。

# run your first task instance
airflow run example_bash_operator runme_0 2015-01-01
# run a backfill over 2 days
airflow backfill example_bash_operator -s 2015-01-01 -e 2015-01-02

3.扩展包

airflow有很多扩展包，如下图：

subpackage	install command	enables
all	`pip install apache-airflow[all]`	All Airflow features known to man
all_dbs	`pip install apache-airflow[all_dbs]`	All databases integrations
async	`pip install apache-airflow[async]`	Async worker classes for gunicorn
devel	`pip install apache-airflow[devel]`	Minimum dev tools requirements
devel_hadoop	`pip install apache-airflow[devel_hadoop]`	Airflow + dependencies on the Hadoop stack
celery	`pip install apache-airflow[celery]`	CeleryExecutor
crypto	`pip install apache-airflow[crypto]`	Encrypt connection passwords in metadata db
druid	`pip install apache-airflow[druid]`	Druid.io related operators & hooks
gcp_api	`pip install apache-airflow[gcp_api]`	Google Cloud Platform hooks and operators (using google-api-python-client)
jdbc	`pip install apache-airflow[jdbc]`	JDBC hooks and operators
hdfs	`pip install apache-airflow[hdfs]`	HDFS hooks and operators
hive	`pip install apache-airflow[hive]`	All Hive related operators
kerberos	`pip install apache-airflow[kerberos]`	kerberos integration for kerberized hadoop
ldap	`pip install apache-airflow[ldap]`	ldap authentication for users
mssql	`pip install apache-airflow[mssql]`	Microsoft SQL operators and hook, support as an Airflow backend
mysql	`pip install apache-airflow[mysql]`	MySQL operators and hook, support as an Airflow backend
password	`pip install apache-airflow[password]`	Password Authentication for users
postgres	`pip install apache-airflow[postgres]`	Postgres operators and hook, support as an Airflow backend
qds	`pip install apache-airflow[qds]`	Enable QDS (qubole data services) support
rabbitmq	`pip install apache-airflow[rabbitmq]`	Rabbitmq support as a Celery backend
s3	`pip install apache-airflow[s3]`	`S3KeySensor, S3PrefixSensor`
samba	`pip install apache-airflow[samba]`	`Hive2SambaOperator`
slack	`pip install apache-airflow[slack]`	`SlackAPIPostOperator`
vertica	`pip install apache-airflow[vertica]`	Vertica hook support as an Airflow backend
cloudant	`pip install apache-airflow[cloudant]`	Cloudant hook
redis	`pip install apache-airflow[redis]`	Redis hooks and sensors