从python使用Airflow触发SQL时出现模板错误

Question

我正在尝试从Airflow运行SQL脚本。由于模板错误，它失败了。

该脚本基本上是尝试从Athena运行sql并加载到Redshift表中。

SQl位于：redshift/sql/public/flow/KN_AWS_RE_ShipmentData_dup_update.sql

我的气流代码

from datetime import datetime, timedelta
from airflow import DAG
from airflow.operators.postgres_operator import PostgresOperator
from utils import FAILURE_EMAILS
yesterday = datetime.combine(datetime.today() - timedelta(1), datetime.min.time())
default_args = {
'owner': 'airflow',
'depends_on_past': False,
'start_date': yesterday,
'email': FAILURE_EMAILS,
'email_on_failure': False,
'email_on_retry': False,
'retries': 1,
'retry_delay': timedelta(minutes=5)
}
dag = DAG('sample_sql', default_args=default_args, schedule_interval='0 4 * * *')
execute_notebook = PostgresOperator(
task_id='sample_sql',
postgres_conn_id='REDSHIFT_CONN',
sql="redshift/sql/public/flow/KN_AWS_RE_ShipmentData_dup_update.sql",
params={'limit': '50'},
dag=dag
)

错误

[2020-04-14 02:19:24,412] {{standard_task_runner.py:52}} INFO - Started process 23012 to run task
[2020-04-14 02:19:24,482] {{logging_mixin.py:112}} INFO - [2020-04-14 02:19:24,481] {{dagbag.py:403}} INFO - Filling up the DagBag from /usr/local/airflow/dags/Scripts/Sample.py
[2020-04-14 02:19:24,495] {{baseoperator.py:807}} ERROR - KN_AWS_RE_ShipmentData_dup_update.sql
Traceback (most recent call last):
File "/usr/local/lib/python3.7/site-packages/airflow/models/baseoperator.py", line 805, in resolve_template_files
setattr(self, field, env.loader.get_source(env, content)[0])
File "/usr/local/lib/python3.7/site-packages/jinja2/loaders.py", line 187, in get_source
raise TemplateNotFound(template)
jinja2.exceptions.TemplateNotFound: KN_AWS_RE_ShipmentData_dup_update.sql
[2020-04-14 02:19:24,545] {{logging_mixin.py:112}} INFO - Running %s on host %s <TaskInstance: sample_sql.sample_sql 2020-04-14T02:14:08.020072+00:00 [running]> 0ca54c719ff7
[2020-04-14 02:19:24,583] {{taskinstance.py:1088}} ERROR - KN_AWS_RE_ShipmentData_dup_update.sql
Traceback (most recent call last):
File "/usr/local/lib/python3.7/site-packages/airflow/models/taskinstance.py", line 940, in _run_raw_task
self.render_templates(context=context)

如何解决此问题？

Answer 1

您必须在DAG实例化时使用变量template_searchpath指定.sql模板文件的路径。默认情况下，Jinja会查看您的DAG文件夹。

请注意，您的DAG包含一个不好的做法，即使用动态的start_date。start_date应该是固定的(即datetime(2020,4,13)(而不是动态的(即，datetime.now()(。你可以在这里阅读更多关于它的信息。

这意味着我将尝试将DAG的定义更改为：

from datetime import datetime, timedelta
from airflow import DAG
from airflow.operators.postgres_operator import PostgresOperator
from utils import FAILURE_EMAILS
# Remove this
# yesterday = datetime.combine(datetime.today() - timedelta(1), datetime.min.time())
default_args = {
'owner': 'airflow',
'depends_on_past': False,
'start_date': datetime(2020,4,13),  # Change this
'email': FAILURE_EMAILS,
'email_on_failure': False,
'email_on_retry': False,
'retries': 1,
'retry_delay': timedelta(minutes=5)
}
dag = DAG(
'sample_sql', 
default_args=default_args, 
schedule_interval='0 4 * * *',
template_searchpath='/redshift/sql/public/flow/')
execute_notebook = PostgresOperator(
task_id='sample_sql',
postgres_conn_id='REDSHIFT_CONN',
sql='KN_AWS_RE_ShipmentData_dup_update.sql',
params={'limit': '50'},
dag=dag
)
execute_notebook  # Tell airflow the tasks dependencies, in this case no dependency

当然，您应该选择正确的绝对基本路径来分配给template_searchpath，比如/home/redshift/sql/public/flow。