Why monitoring prevents silent pipeline failures in Apache Airflow - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to set the task failure callback in Airflow.

Apache Airflow

task = PythonOperator(task_id='my_task', python_callable=my_func, on_failure_callback=[1])

Drag options to blanks, or click blank then click option'

Asend_alert

Bsend_email

Clog_error

Dretry_task

Attempts:

3 left

💡 Hint

Common Mistakes

Using a function that retries the task instead of notifying.

The on_failure_callback parameter is used to specify a function like send_email that triggers when the task fails, helping to monitor failures.

2fill in blank

medium

Complete the code to add a sensor that waits for a file before running the task.

Apache Airflow

wait_for_file = [1](task_id='wait_for_file', filepath='/data/input.csv')

Drag options to blanks, or click blank then click option'

AHttpSensor

BTimeSensor

CFileSensor

DEmailSensor

Attempts:

3 left

💡 Hint

Common Mistakes

Using a sensor that waits for time or HTTP instead of a file.

FileSensor waits for a file to appear before the downstream task runs, preventing silent failures due to missing input.

3fill in blank

hard

Fix the error in the DAG definition to enable email alerts on failure.

Apache Airflow

default_args = {'owner': 'airflow', 'email': ['[e-mail]'], 'email_on_failure': [1]

dag = DAG('my_dag', default_args=default_args, schedule_interval='@daily')

Drag options to blanks, or click blank then click option'

AFalse

BNone

C0

DTrue

Attempts:

3 left

💡 Hint

Common Mistakes

Setting it to False or None disables email alerts.

Setting email_on_failure to True enables sending email alerts when a task fails.

4fill in blank

hard

Fill both blanks to create a task that retries 3 times with a 5-minute delay.

Apache Airflow

task = PythonOperator(task_id='retry_task', python_callable=my_func, retries=[1], retry_delay=timedelta(minutes=[2]))

Drag options to blanks, or click blank then click option'

A3

B5

C1

D10

Attempts:

3 left

💡 Hint

Common Mistakes

Using too few retries or zero delay.

Setting retries=3 and retry_delay=timedelta(minutes=5) makes the task retry 3 times with 5 minutes between attempts, helping catch transient failures.

5fill in blank

hard

Fill all three blanks to create a DAG with a start date, schedule, and catchup disabled.

Apache Airflow

dag = DAG('example_dag', start_date=datetime([1], 1, 1), schedule_interval='[2]', catchup=[3])

Drag options to blanks, or click blank then click option'

A2024

B@daily

CFalse

DTrue

Attempts:

3 left

💡 Hint

Common Mistakes

Setting catchup to True causes backfill runs.

Setting start_date=datetime(2024, 1, 1), schedule_interval='@daily', and catchup=False creates a DAG that runs daily from 2024 without backfilling missed runs, preventing silent failures from unexpected catchups.

Practice

1. Why is monitoring important in Airflow pipelines?

easy

A. It hides errors to keep the pipeline running smoothly.

B. It automatically fixes all errors without user input.

C. It alerts you immediately when a task fails.

D. It slows down the pipeline to prevent errors.

Solution

1. Step 1: Understand the role of monitoring

   Monitoring watches the pipeline and notices when something goes wrong.

2. Step 2: Identify the benefit of alerts

   Alerts notify you immediately so you can fix issues quickly.

3. Final Answer:

   It alerts you immediately when a task fails. -> Option C

4. Quick Check:

   Monitoring = Immediate alerts [OK]

Hint: Monitoring means getting alerts on failures fast [OK]

Common Mistakes:

• Thinking monitoring fixes errors automatically
• Believing monitoring hides errors
• Assuming monitoring slows down pipelines

2. Which Airflow feature helps you get notified when a task fails?

easy

A. DAG scheduling intervals

B. Failure callback functions

C. Task retries without notification

D. Task dependencies

Solution

1. Step 1: Identify notification features in Airflow

   Failure callback functions run when a task fails and can send alerts.

2. Step 2: Compare other options

   Retries try again silently, scheduling controls timing, dependencies control order but don't notify.

3. Final Answer:

   Failure callback functions -> Option B

4. Quick Check:

   Failure callbacks = Notifications [OK]

Hint: Failure callbacks notify on errors [OK]

Common Mistakes:

• Confusing retries with notifications
• Thinking scheduling sends alerts
• Believing dependencies notify failures

3. Given this Airflow task snippet, what happens if the task fails?

def notify_failure(context):
    print('Task failed:', context['task_instance'].task_id)

task = PythonOperator(
    task_id='example_task',
    python_callable=my_function,
    on_failure_callback=notify_failure
)

medium

A. The pipeline stops without any notification.

B. The task retries automatically without any message.

C. The task succeeds silently without any output.

D. The failure message prints immediately when the task fails.

Solution

1. Step 1: Understand the on_failure_callback usage

   The function notify_failure runs when the task fails, printing a message.

2. Step 2: Analyze the effect of the callback

   When failure occurs, the message 'Task failed: example_task' prints immediately.

3. Final Answer:

   The failure message prints immediately when the task fails. -> Option D

4. Quick Check:

   on_failure_callback triggers print on failure [OK]

Hint: on_failure_callback runs on task failure [OK]

Common Mistakes:

• Assuming retries happen without message
• Thinking task succeeds silently
• Believing pipeline stops silently

4. You set an on_failure_callback but never receive alerts when tasks fail. What is the most likely cause?

medium

A. The callback function is not properly linked to the task.

B. The task never actually fails.

C. Airflow disables failure callbacks by default.

D. The DAG schedule interval is too short.

Solution

1. Step 1: Check callback linkage

   If the callback is not linked correctly, it won't run on failure.

2. Step 2: Evaluate other options

   Tasks can fail, Airflow does not disable callbacks by default, and schedule interval doesn't affect callbacks.

3. Final Answer:

   The callback function is not properly linked to the task. -> Option A

4. Quick Check:

   Callback linkage = Alerts working [OK]

Hint: Ensure callback is linked to task correctly [OK]

Common Mistakes:

• Assuming tasks never fail
• Thinking Airflow disables callbacks
• Blaming schedule interval for alerts

5. How does setting up monitoring and failure callbacks in Airflow help prevent silent pipeline failures from causing bigger problems?

hard

A. They alert you early so you can fix issues before they cascade.

B. They automatically restart the entire pipeline on any failure.

C. They hide errors to keep the pipeline running without interruption.

D. They delay task execution to avoid failures.

Solution

1. Step 1: Understand the purpose of monitoring and callbacks

   Monitoring and callbacks send alerts immediately when failures happen.

2. Step 2: Connect early alerts to preventing bigger issues

   Early alerts let you fix problems quickly, stopping failures from silently causing more damage.

3. Final Answer:

   They alert you early so you can fix issues before they cascade. -> Option A

4. Quick Check:

   Early alerts prevent bigger failures [OK]

Hint: Early alerts stop small failures growing [OK]

Common Mistakes:

• Thinking monitoring restarts pipelines automatically
• Believing errors are hidden
• Assuming delays prevent failures