Free Top Amazon/AWS DEA-C01 Actual Exam Questions - Question 5 Discussion
The company has enabled logging and monitoring for all AWS Glue jobs. One of the AWS Glue jobs
begins to fail. A data engineer investigates the error and wants to examine metrics for all individual
stages within the job. How can the data engineer access the stage metrics?
Maybe A here. The Spark UI is designed to show detailed metrics for each stage of a Glue job since Glue runs on Apache Spark. CloudWatch (B) usually shows overall job metrics but not the fine-grained stage details. CloudTrail (C) is more about API calls, so it wouldn’t have stage metrics. Run Insights (D) sounds useful but it depends on the Glue version, which isn’t specified, so it feels less certain. Given that, checking the Spark UI directly seems like the most reliable way to get stage-level info.
A Spark UI gives a direct look at each stage’s performance, so it’s handy if you want detailed metrics without extra features. It’s built for that kind of granular Spark info.
A/D? Run Insights is great if supported, but if the job’s running on an older Glue version, the Spark UI (A) is still a solid fallback for detailed stage metrics without extra setup.
Adeel E. imo Run Insights (D) is made for deep diagnostics, but if you want to cross-check or can't use it, the Spark UI (A) still gives detailed stage metrics directly from the job execution.
B tbh, CloudWatch gives you good overall metrics but not the fine-grained stage details. That level of detail usually needs something integrated with the job itself, so options A or D seem more likely.
If the Spark UI requires extra setup and isn’t always on, that might rule out A. CloudWatch usually handles aggregated metrics, so it might not show stage-level details clearly. Could D be the only option giving direct stage metrics?
Maybe D. From what I’ve seen, Run Insights is designed specifically for detailed job diagnostics, including stage-level metrics, and it doesn’t require extra setup like the Spark UI often does. CloudWatch usually only has higher-level job metrics, so it’s probably not the best for stage details. CloudTrail is definitely out since it logs API calls, not job stages. So between A and D, I’d pick D because it’s more straightforward and purpose-built for this kind of troubleshooting in Glue.
D imo, Run Insights is designed to give detailed stage-level metrics without needing extra setup, unlike Spark UI which can be tricky to access in Glue. CloudWatch won’t show that granular stage info.
I’m leaning towards A because the Spark UI usually shows detailed stage metrics for Glue jobs. CloudWatch is good for overall job metrics but not per stage, and CloudTrail is more about API calls, so C seems off. Not sure about D, haven’t used run insights much. Anyone else?