我在气流中设置了一个条件任务here。它所做的就是检查是否存在hive分区。如果是,请继续执行其余任务,如果不是,请先继续添加分区,然后再继续。条件检查任务可能失败或成功,两者都可以。但是,我为dag设置了寻呼机职责电子邮件通知,因为我想知道下游任务何时失败。如何将该特定条件任务的失败通知静音,以便我不会在寻呼机上发生误报?
答案 0 :(得分:1)
email_on_failure
和on_failure_callback
等是任务(操作员)级别参数。它们继承自DAG对象,传递给DAG的default_args
值,但您也可以在初始化时覆盖它们。
YourOperator(task_id='task1', dag=dag, email_on_failure=None, on_failure_callback=None, ...)
这是源代码,当任务失败时,气流如何处理这些回调,让你更清楚它是如何工作的。
def handle_failure(self, error, test_mode=False, context=None):
self.log.exception(error)
task = self.task
session = settings.Session()
self.end_date = datetime.utcnow()
self.set_duration()
Stats.incr('operator_failures_{}'.format(task.__class__.__name__), 1, 1)
Stats.incr('ti_failures')
if not test_mode:
session.add(Log(State.FAILED, self))
# Log failure duration
session.add(TaskFail(task, self.execution_date, self.start_date, self.end_date))
# Let's go deeper
try:
# Since this function is called only when the TI state is running,
# try_number contains the current try_number (not the next). We
# only mark task instance as FAILED if the next task instance
# try_number exceeds the max_tries.
if task.retries and self.try_number <= self.max_tries:
self.state = State.UP_FOR_RETRY
self.log.info('Marking task as UP_FOR_RETRY')
if task.email_on_retry and task.email:
self.email_alert(error, is_retry=True)
else:
self.state = State.FAILED
if task.retries:
self.log.info('All retries failed; marking task as FAILED')
else:
self.log.info('Marking task as FAILED.')
if task.email_on_failure and task.email:
self.email_alert(error, is_retry=False)
except Exception as e2:
self.log.error('Failed to send email to: %s', task.email)
self.log.exception(e2)
# Handling callbacks pessimistically
try:
if self.state == State.UP_FOR_RETRY and task.on_retry_callback:
task.on_retry_callback(context)
if self.state == State.FAILED and task.on_failure_callback:
task.on_failure_callback(context)
except Exception as e3:
self.log.error("Failed at executing callback")
self.log.exception(e3)
if not test_mode:
session.merge(self)
session.commit()
self.log.error(str(error))
https://airflow.apache.org/_modules/airflow/models.html#BaseOperator