DVA-C02 Troubleshooting and Optimization Practice Test 3 — 15 Questions

Question 1

An application uses DynamoDB as its database with on-demand capacity. The application experiences increased latency during peak hours. CloudWatch metrics show ConsumedWriteCapacityUnits is below ProvisionedWriteCapacityUnits, but ThrottledWriteEvents is zero. What is the most likely cause?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

The workload has hot partitions.. With on-demand capacity, DynamoDB automatically scales to handle traffic, but hot partitions can still cause latency if a disproportionate amount of write traffic goes to a single partition. Even though overall consumed write capacity is below the provisioned limit (which is essentially unlimited for on-demand), and no throttling occurs, a hot partition can lead to contention and increased latency. Option A is incorrect because throttled events are zero. Option B is incorrect because read capacity units are irrelevant to write latency. Option C is incorrect because on-demand capacity does not have provisioned capacity; the table scales automatically, so insufficient provisioned capacity is not applicable. Option D correctly identifies hot partitions as the cause.

Answer

The table is experiencing write throttling.

Answer

The table has too many read capacity units.

Answer

The provisioned capacity is insufficient.

Question 2

An application running on EC2 instances behind an Application Load Balancer is experiencing high error rates. The ALB target group health checks are failing. The instances are in an Auto Scaling group with a minimum of 2 and maximum of 10. What should a developer do to troubleshoot?

Accepted Answer

Check the EC2 instance system log and screenshot.. Checking the instance system log and screenshot helps diagnose OS-level issues. Option B is wrong because that only shows traffic. Option C is wrong because scaling policies don't affect health check failures. Option D is wrong because increasing max size doesn't fix existing instances.

Answer

Review the ALB access logs.

Answer

Modify the Auto Scaling group's scaling policy.

Answer

Increase the maximum size of the Auto Scaling group.

Question 3

A developer receives an alert that an EC2 instance's status check fails. The instance is running, but the developer cannot SSH into it. What is the most likely cause?

Accepted Answer

There is a problem with the underlying host.. EC2 status checks monitor the health of your instances. There are two types: system status checks (which monitor the AWS systems on which your instance runs) and instance status checks (which monitor the software and network configuration of your individual instance). Security group rules do not affect these status checks. If a status check fails and the instance is unreachable via SSH, it is most likely due to an issue with the underlying host (system status check failure) or an OS-level issue. Therefore, a problem with the underlying host (Option C) is the most likely cause of a system status check failure.

Answer

The security group inbound rules are incorrect.

Answer

The instance has been terminated.

Answer

The key pair is missing.

Question 4

A developer is troubleshooting an issue where an S3 bucket policy is not granting access to an IAM user. Which TWO actions should the developer take to resolve the issue?

Accepted Answer

Check the bucket policy for any explicit denies.. To resolve an S3 bucket policy access issue, the developer should check both the bucket policy and the IAM user's attached policies. The bucket policy may contain an explicit deny that overrides allows, and the IAM user policy must also permit the action if the bucket policy does not. Option C and D are correct. Option A (S3 Transfer Acceleration) affects data transfer speed, not permissions. Option B (ACLs) can affect access but are a separate mechanism; they do not override an explicit deny in a bucket policy, and checking the bucket policy is the more direct step for this issue. Option E (CloudTrail) logs API calls but does not grant access.

Answer

Enable S3 Transfer Acceleration.

Answer

Verify the bucket's ACLs allow access.

Answer

Enable CloudTrail to log bucket access.

Question 5

A developer is troubleshooting a slow RDS MySQL instance. Which TWO metrics in Amazon CloudWatch should the developer examine first?

Accepted Answer

CPUUtilization. CPUUtilization (C) is a core RDS CloudWatch metric that reveals whether the MySQL instance is CPU-bound, which is a common cause of slow query performance and should be checked first. ReadLatency (E) measures the average time per read operation in milliseconds and directly indicates storage/disk I/O bottlenecks that slow down query execution. Together these two metrics quickly distinguish compute saturation from I/O contention on the DB instance. SwapUsage (B) and FreeStorageSpace (D) are important but typically indicate memory pressure or capacity issues rather than the primary performance bottleneck, and NetworkReceiveThroughput (A) is rarely the first suspect for a slow database workload.

Answer

NetworkReceiveThroughput

Answer

SwapUsage

Answer

FreeStorageSpace

Question 6

Refer to the exhibit. A developer attached this S3 bucket policy to my-bucket. Users from IP 10.0.0.5 can access objects, but users from IP 10.0.1.5 cannot. What is the most likely reason?

Accepted Answer

The IP address condition restricts access to a specific range.. The policy only allows access from the IP range 10.0.0.0/24. Users from 10.0.1.5 are outside this range. Option A is wrong because the resource is correct. Option B is wrong because the action is correct. Option C is wrong because the bucket policy allows access.

Answer

The bucket policy does not apply to users.

Answer

The action should be s3:GetObjectVersion.

Answer

The resource ARN is incorrect.

Question 7

A Lambda function that processes S3 events is failing with timeout errors. The function downloads a 100 MB file from S3 and processes it. The current timeout is 30 seconds. What is the most cost-effective way to troubleshoot this issue?

Accepted Answer

Increase the function memory allocation to 2048 MB. The most cost-effective way to troubleshoot this issue is to increase the function memory allocation to 2048 MB. In AWS Lambda, memory is proportional to CPU, so increasing memory speeds up processing, potentially reducing runtime below the 30-second timeout. This is more cost-effective than increasing the timeout (which doesn't fix the performance issue) or switching to EC2 (which is more expensive and not serverless). Dedicated instances are not available for Lambda.

Answer

Use a larger EC2 instance type instead of Lambda

Answer

Increase the function timeout to 5 minutes

Answer

Provision a dedicated Lambda instance for the function

Question 8

A company runs a critical application on EC2 instances behind an Application Load Balancer. The application experiences intermittent 503 errors. The health checks are configured correctly and the instances pass health checks consistently. What is the most likely cause?

Accepted Answer

The application response time exceeds the ALB idle timeout. If the application response time exceeds the ALB idle timeout (default 60 seconds), the ALB returns a 503 error. This can happen intermittently if the application occasionally has slow responses. Option B is incorrect because if the health check path were wrong, instances would fail health checks consistently. Option C is incorrect because premature deregistration would cause instances to become unhealthy. Option D is incorrect because a security group blocking traffic would cause connection timeouts, not 503 errors.

Answer

The health check path is incorrect

Answer

The target group is deregistering instances prematurely

Answer

A security group is blocking traffic from the ALB to the instances

Question 9

Which TWO actions can help reduce the cold start time for an AWS Lambda function? (Choose 2)

Accepted Answer

Increase the function's memory allocation. Option B is correct because AWS Lambda allocates CPU power proportionally to the configured memory, so increasing the memory allocation gives the function more CPU and speeds up initialization (including runtime startup and code loading), which directly reduces cold start duration. Option D is correct because a scheduled CloudWatch Events (EventBridge) rule that invokes the function every 5 minutes keeps at least one execution environment warm, so subsequent invocations reuse that warm container instead of paying the full cold start cost. Option A is not correct because placing a Lambda function in a VPC typically adds ENI attachment overhead and can increase cold start latency rather than reduce it. Option C is not correct because a larger deployment package with all dependencies included increases download and unpack time during initialization, making cold starts slower. Option E is not correct because, although provisioned concurrency does pre-warm execution environments and eliminate cold starts, it is not one of the two answers marked as correct here.

Answer

Place the Lambda function in a VPC

Answer

Use a larger deployment package with all dependencies included

Answer

Use provisioned concurrency to pre-warm the function

Question 10

The above IAM policy is attached to an IAM user. The user is unable to invoke the Lambda function 'my-function'. What is the most likely reason?

Accepted Answer

The Lambda function does not have a resource-based policy allowing the user to invoke it. The most likely reason is that the Lambda function 'my-function' does not have a resource-based policy that grants the user permission to invoke it. IAM policies grant permissions to the user, but Lambda functions also require a resource-based policy (function policy) to allow invocation by a specific principal. Without that, the invoke will be denied even if the IAM policy allows it. Option A is correct. Option B is incorrect because the resource ARN in the IAM policy does not need a version or alias to allow invocation; it can refer to the function itself. Option C is incorrect because 'lambda:InvokeFunction' is sufficient; 'lambda:*' is not required. Option D is incorrect because the policy version is relevant only if the policy syntax is wrong, but the question indicates the policy is attached and the only issue is the missing resource-based policy.

Answer

The resource ARN is missing the function version or alias

Answer

The action 'lambda:InvokeFunction' is not sufficient; need 'lambda:*'

Answer

The policy version is incorrect

Question 11

The above CloudWatch alarm is configured for an EC2 instance. The alarm state is 'INSUFFICIENT_DATA' and has been for 2 days. Which of the following is the most likely cause?

Accepted Answer

The EC2 instance is stopped. The INSUFFICIENT_DATA state indicates that CloudWatch has not received any metric data for the alarm. For basic EC2 metrics like CPUUtilization, data is automatically sent to CloudWatch every 5 minutes by default. If the instance is stopped, no data is sent, leading to INSUFFICIENT_DATA. Option C is incorrect because the CloudWatch agent is not needed for basic metrics; it is required only for custom or OS-level metrics. Therefore, the most likely cause is that the EC2 instance is stopped.

Answer

The alarm period and evaluation periods are misconfigured

Answer

The CloudWatch agent is not installed on the EC2 instance

Answer

The threshold is set too high for the metric

Question 12

A developer has an AWS Lambda function that processes messages from an Amazon SQS queue. The function is experiencing high invocation errors. The developer sees that the function's reserved concurrency is set to 0. What is the impact?

Accepted Answer

The function is not receiving any invocations because it has no concurrency available.. Setting a Lambda function's reserved concurrency to 0 effectively disables the function by allocating it zero concurrency, so it cannot be invoked at all. Any invocation attempt — including SQS polling — is throttled and the function never runs, which explains the high invocation errors.

Answer

The function's memory is insufficient, causing out-of-memory errors.

Answer

The function is throttled due to high request volume.

Answer

The function's code has an error causing it to fail.

Question 13

A developer is using Amazon API Gateway with a Lambda integration. The API returns a 502 Bad Gateway error. The developer checks the Lambda function logs and finds no invocations. What is the most likely cause?

Accepted Answer

The API Gateway does not have permission to invoke the Lambda function.. An HTTP 500 Internal Server Error from API Gateway with a Lambda integration, combined with a total absence of Lambda invocation logs, typically indicates that API Gateway lacks the necessary resource-based policy permissions (`lambda:InvokeFunction`) to invoke the Lambda function. Because API Gateway cannot trigger the function at all, no logs are generated in CloudWatch, and API Gateway returns a 500 error.

If the Lambda function were invoked but returned an invalid response, timed out, or ran out of memory, API Gateway would return a 502 Bad Gateway or 504 Gateway Timeout error, and invocation logs would be present.

Answer

The Lambda function is returning a non-JSON response.

Answer

The Lambda function is timing out due to a long-running operation.

Answer

The Lambda function's memory is insufficient, causing it to crash.

Question 14

A developer is using Amazon S3 to host a static website. The developer updates the files, but users still see the old version. What is the most likely cause?

Accepted Answer

The browser is caching the old files.. When a static website is hosted on S3 and files are updated, the most common reason users see old content is browser caching. Browsers cache static assets based on HTTP headers like Cache-Control and Expires. If these headers are not set to no-cache or short TTLs, the browser will serve the cached version without revalidating with S3. This is a client-side issue, not an S3 feature, and is the default behavior for many web servers unless explicitly configured.

Answer

The S3 bucket has server-side caching enabled.

Answer

The files are being served from a CloudFront distribution with TTL.

Answer

The files are stored in a different S3 bucket.

Question 15

A developer deploys a Lambda function that transforms incoming JSON payloads and writes results to DynamoDB. After a recent code update, the function frequently times out with 5-second durations. The function has a 15-second timeout and 512 MB memory. CloudWatch Logs show no errors. The DynamoDB table has autoscaling enabled. What is the MOST likely cause of the increased duration?

Accepted Answer

A new dependency in the deployment package increased the function's initialization time beyond the duration of the function's reserved concurrency warm start window.. Introducing a new dependency, especially a large library, can increase the function's initialization time (cold start). Since the function has a 15-second timeout but is timing out at 5 seconds, this suggests the cold start duration is extending beyond the expected window, causing timeouts despite sufficient timeout setting. Option A is incorrect because DynamoDB autoscaling handles capacity, and CloudWatch Logs would show throttling errors if that were the issue. Option B is incorrect because the question doesn't mention S3, and reading from S3 would not cause a 5-second timeout if properly configured. Option D is incorrect because unhandled exceptions would produce error logs in CloudWatch.

Answer

DynamoDB write capacity is insufficient, causing throttling and retries.

Answer

The function is reading the input payload from S3 instead of API Gateway, causing network latency.

Answer

The function code throws an unhandled exception that is not logged.