DOP-C02 Monitoring and Logging Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Invocations

  Why it's wrong here

  The Invocations metric increments each time Lambda is invoked by the Kinesis event source mapping, regardless of whether the execution succeeds or fails. It therefore reflects call volume, not the health or outcome of streaming data processing. A healthy, heavy workload can show a high Invocations count while a failing processor might actually show a decreasing count if the mapping stops invoking the function. Because it fails to distinguish successful processing from backend or data errors, it cannot serve as a reliable signal for detecting streaming data failures.

• ✓

  IteratorAge

  Why this is correct

  IteratorAge is the correct metric to monitor for Kinesis-triggered Lambda because it directly measures the age of the oldest unprocessed record, reported in milliseconds. When the Lambda consumer can't keep up with the shard's data rate, the iterator age grows, indicating that stream records are sitting unprocessed for longer — a clear sign of a processing bottleneck or backlog. A sustained increase in IteratorAge typically drives alarms for scaling out the Lambda function or increasing the number of shards, making it the definitive indicator of stream processing lag.

• ✗

  Errors

  Why it's wrong here

  The Errors metric only counts Lambda invocations that terminate due to unhandled exceptions or runtime crashes during the function execution. However, many Kinesis processing failures never surface as invocation errors: a batch can be successfully invoked but fail to checkpoint, or the function may intentionally swallow a record that can't be deserialized after retries. Furthermore, errors that occur in downstream systems or during partial batch failures are not reflected in this metric. Thus, a near-zero Errors count can still coexist with a significant stream backlog, making it an unreliable measure for this scenario.

• ✗

  Throttles

  Why it's wrong here

  Throttles occurs when Lambda refuses to run a new invocation because the function's reserved or account-level concurrency limit is already exhausted. In a Kinesis event-source mapping, a throttled event is retried by Lambda, so the record is not immediately lost, but repeated throttling can stall progress and eventually cause data to be pushed beyond the retention window. The metric specifically indicates a resource capacity constraint rather than a failure inherent to the data or processing logic. Therefore, Throttles alone does not reveal why records are failing or lagging; it only shows when the invocation was rejected before code could run.

Go deeper

Related to this question