Apache Airflow Certification Practice Test 04

Production Support Troubleshooting Scenarios

Question 1 of 20

📋 Multiple Choice

Question 61

A newly joined Airflow Support Engineer receives an alert at 3:00 AM. A critical Finance DAG failed and business reports were not generated. The Airflow UI shows: - DAG Run = Failed - 1 Task Failed - 7 Tasks Succeeded What should the engineer do NEXT?

✓

A. Review failed task logs

B. Restart all workers

C. Delete the DAG

D. Create a new schedule

Answer

✅ A. Review failed task logs

Answered: 0 / 20

Apache Airflow Certification Practice Test 04

Production Support Troubleshooting Scenarios

Insightful Saga — Modern Data Engineering Certification Preparation

Question 61

A. Review failed task logs

B. Restart all workers

C. Delete the DAG

D. Create a new schedule

Answer: ✅ A. Review failed task logs

Question 62

A retail company reports missing sales data. Investigation shows: - DAG ran successfully - Extraction task succeeded - Load task processed zero records What should the engineer investigate NEXT?

A. Source data availability

B. Pool configuration

C. Airflow version

D. Worker hostname

Answer: ✅ A. Source data availability

Question 63

An Airflow DAG was expected to run at 1:00 AM but no DAG run exists. What should the engineer verify FIRST?

A. DAG is unpaused

B. XCom values

C. Task Groups

D. Pool slots

Answer: ✅ A. DAG is unpaused

Question 64

A healthcare workflow has been stuck in "queued" state for 45 minutes. Many other workflows are running normally. What should be reviewed first?

A. Pool availability

B. Variable values

C. XCom messages

D. Trigger rules

Answer: ✅ A. Pool availability

Question 65

A banking pipeline is failing every day at the same task. The task calls an external API. Logs show repeated connection timeout errors. What should be recommended?

A. Retry strategy with backoff

B. More DAGs

C. More Variables

D. Remove dependencies

Answer: ✅ A. Retry strategy with backoff

Question 66

An airline company notices the Scheduler is healthy but newly added DAGs are not appearing in Airflow. What should be verified next?

A. DAG file deployment location

B. Pool count

C. XCom size

D. Worker memory only

Answer: ✅ A. DAG file deployment location

Question 67

A support engineer receives an SLA breach alert. The workflow completed successfully but exceeded the required completion time. What should be reviewed first?

A. Long-running tasks

B. Role hierarchy

C. Variable names

D. DAG description

Answer: ✅ A. Long-running tasks

Question 68

A retail company executes a DAG every hour. Business users report data duplication. Investigation reveals the DAG executed twice for the same schedule interval. What should be checked?

A. Trigger history

B. XCom cleanup

C. Pool count

D. Connection metadata

Answer: ✅ A. Trigger history

Question 69

A task waits for a file to arrive using a Sensor. The file never arrives. What is the MOST likely reason for the DAG remaining active for hours?

A. Sensor waiting condition

B. Variable corruption

C. Pool overflow

D. Trigger rule issue

Answer: ✅ A. Sensor waiting condition

Question 70

A manufacturing workflow is failing after a source system schema change. What should engineers review first?

A. Upstream schema modifications

B. Queue settings

C. Worker names

D. DAG documentation

Answer: ✅ A. Upstream schema modifications

Question 71

A task succeeds when executed manually but fails in scheduled execution. What should the support engineer investigate first?

A. Runtime environment differences

B. DAG color

C. Pool description

D. Airflow logo version

Answer: ✅ A. Runtime environment differences

Question 72

A company notices workers are running at 100% utilization continuously. Users report growing delays. What should be reviewed next?

A. Worker capacity planning

B. Variable values

C. Task Groups

D. DAG tags

Answer: ✅ A. Worker capacity planning

Question 73

A banking workflow completed successfully. However, customers complain that reports were never received. Which component should be checked first?

A. Notification task

B. Scheduler

C. Worker startup

D. DAG tags

Answer: ✅ A. Notification task

Question 74

A healthcare ETL workflow stopped processing records three days ago. The DAG still reports success. What should engineers investigate?

A. Business output validation

B. DAG descriptions

C. Airflow UI theme

D. Queue names

Answer: ✅ A. Business output validation

Question 75

An Airflow deployment suddenly processes historical schedules from two months ago. What configuration should be reviewed?

A. Catchup setting

B. Email settings

C. Pool slots

D. Trigger rules

Answer: ✅ A. Catchup setting

Question 76

A finance DAG fails intermittently. The same code sometimes succeeds and sometimes fails. What category of issue should be investigated first?

A. Environmental instability

B. DAG naming

C. Pool descriptions

D. Documentation accuracy

Answer: ✅ A. Environmental instability

Question 77

A support engineer sees thousands of queued tasks but very few running tasks. What should be reviewed immediately?

A. Worker availability

B. XCom payload size

C. DAG tags

D. Variables

Answer: ✅ A. Worker availability

Question 78

A business team reports yesterday's processing never occurred. No DAG runs exist for the missed date. What should be investigated first?

A. Scheduler activity

B. XCom cleanup

C. Trigger rules

D. Pool management

Answer: ✅ A. Scheduler activity

Question 79

A retail company wants quicker recovery from overnight failures. What should be implemented first?

A. Automated alerting

B. More DAGs

C. Additional Variables

D. More Task Groups

Answer: ✅ A. Automated alerting

Question 80

A newly promoted Airflow Support Lead is asked: "What is the first responsibility during an incident?" Choose the BEST answer.

A. Identify impact and isolate the failure

B. Rewrite the DAG immediately

C. Create a new Airflow environment

D. Deploy unrelated fixes

Answer: ✅ A. Identify impact and isolate the failure

Explanation:

Effective incident response starts with understanding business impact, identifying the failing component, and narrowing the scope of investigation before applying corrective actions.