5.4 Monitoring and Troubleshooting Processes PROFESSIONAL-DATA-ENGINEER Practice Questions (Page 6)

Questions 26–30

1. 26application · medium

   Your company runs a nightly Dataflow pipeline that loads data into BigQuery. Recently, the pipeline has been failing intermittently at 2:00 AM, but the failures are not always the same. You need to quickly identify the root cause when the next failure occurs. What should you do?

   • AEnable BigQuery audit logs and review them the next morning to see which queries failed.
   • BCreate a Cloud Monitoring alert that triggers on any Dataflow job failure and includes a link to the job's logs in the alert notification.
   • CSet up a Cloud Scheduler job that runs every hour and checks the Dataflow job status, sending an email if the job is not successful.
   • DConfigure a Cloud Function that retries the failed Dataflow job automatically and logs the retry attempt.
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   Which strategy can help avoid hitting the BigQuery API request quota for a data pipeline?

   • ABatching multiple operations into fewer API calls
   • BCreating more datasets
   • CEnabling Cloud Logging for the pipeline
   • DUsing a larger machine type for the pipeline
   Reveal Solution Discussion Select an answer first
4. 28foundation · easy

   Which Cloud Monitoring resource is used to build a custom view of a data pipeline's CPU utilization and error rate on a single page?

   • ADashboard
   • BNotification channel
   • CLog sink
   • DAlerting policy
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A Dataflow job fails with the error 'Workflow failed. Causes: The Dataflow service is not available in the region.' You need to resolve this issue. What should you do?

   • AContact Google Cloud support to enable Dataflow in that region.
   • BRerun the job in a different region where Dataflow is available.
   • CCheck the job logs to see if there is a more specific error message.
   • DIncrease the number of workers for the job to overcome the regional limitation.
   Reveal Solution Discussion Select an answer first
6. 30expert · hard

   Your organization's BigQuery costs have increased by 40% over the past month, but the volume of queries has remained stable. You suspect that a specific team may be running expensive queries. You need to identify which team is responsible and take action to control costs. What should you do?

   • AContact Google Cloud support to request a detailed cost breakdown.
   • BUse BigQuery audit logs to analyze query costs by user and project, and then set up a budget alert to notify you when costs exceed a threshold.
   • CReview the Cloud Billing report to see the total cost by project, but do not investigate further.
   • DCreate a Cloud Monitoring alert on the BigQuery slot usage metric to identify when slots are heavily used.
   Reveal Solution Discussion Select an answer first