Airflow只能使用Celery、CeleryKubernetes或KubernetesExecutors _编程开发

创始人

2024-08-02 15:31:46

0次

使用Celery executor：将“executor = SequentialExecutor”更改为“executor = CeleryExecutor”并在airflow.cfg文件的[celery]部分中进行必要的配置，例如broker_url和result_backend等。
使用CeleryKubernetes executor：将“executor = SequentialExecutor”更改为“executor = CeleryKubernetesExecutor”并在airflow.cfg文件的[celery_kubernetes]部分中配置必要的信息，例如namespace、image、worker_concurrency等。
使用Kubernetes executor：将“executor = SequentialExecutor”更改为“executor = KubernetesExecutor”，并在airflow.cfg文件中的[kubernetes]部分进行必要的配置，例如kube_config、namespace、image_pull_secrets等。示例代码： executor = CeleryKubernetesExecutor [callback] hostname_callable = airflow.example.com port = 8080 [celery] broker_url = redis://redis-master:6379/0 celery_result_backend = db+postgresql://airflow:airflow@pg-airflow/airflow [dask] framework = kubernetes [celery_kubernetes] namespace = airflow image_pull_policy = Always image_pull_secrets = NameOfYourImagePullSecrets worker_concurrency = 1 worker_memory = '2Gi' reqs_cpu = '100m' reqs_memory = '256Mi' limit_cpu = '1' limit_memory = '2Gi' tolerations = resources = {'driver-memory': '4G', 'executor-memory': '4G', 'executor-cores': 2}