SOA-C03 · Sample questions
10 real questions from the CertifyDeck SOA-C03 bank, spread across the domains of the official AWS Certified CloudOps Engineer - Associate skills outline, with the correct answer and a full explanation of why the other options are wrong. They are original questions written from the public syllabus, not exam dumps, and they are the same style you will find in the practice tests and mock exams.
Correct answer: A. Pass the encoded message to aws sts decode-authorization-message with permission for that STS action.
Answer each question before opening the explanation, and write down why you discarded the other options. On SOA-C03, most mistakes come from choosing an option that is technically possible but does not meet every requirement in the scenario.
These 10 questions are a taste of the format, not a diagnostic. To know which domains you fail, practice by domain for free and take a timed mock exam that mirrors the official weights and duration.
Start for free, no sign-up or card. Every question includes an explanation like the ones above, and the mock exam reproduces the official SOA-C03 format.
The decode-authorization-message command, which requires permission for sts:DecodeAuthorizationMessage, decodes the message and shows whether the request was denied by an explicit deny or by the absence of an allow, the principal, the action, the resource and the condition key values. The message is encoded because it can contain privileged information, so it is not plain Base64 policy JSON, CloudTrail records the call but not the decoded authorization details, and the simulator tests policies, not encoded messages.
Correct answer: D. The compliance score of the conformance pack that the AWS Config conformance packs page shows.
The compliance score is the percentage of compliant rule-resource combinations in a conformance pack, and AWS documents it as the way to track remediation progress and compare packs. An aggregator count covers every rule in every account rather than this pack's rules as a percentage. Trusted Advisor results are based on its own checks, not on the rules of the pack, and the configuration history is not a compliance measure.
Correct answer: B. Turn on S3 Versioning for the bucket by running put-bucket-versioning with Status=Enabled.
With S3 Versioning turned on, an overwrite creates a new version and a simple delete only inserts a delete marker, so every earlier version stays in the bucket and can be restored. A lifecycle transition only changes the storage class, and access logs and event notifications record that a change happened but keep no copy of the earlier data.
Correct answer: C. Increase the desired capacity of the Auto Scaling group in the recovery Region to production level.
In warm standby everything is already deployed and running at reduced capacity, so the failover procedure only needs to scale up, for example by raising the desired capacity of the Auto Scaling group, after checking that service quotas in the recovery Region allow it. Creating a new Auto Scaling group duplicates infrastructure that already exists, restoring backups does not add application capacity, and Transfer Acceleration only speeds up long-distance uploads to S3.
Correct answer: B. Create a geolocation record with the Default location for the same name and type.
A default geolocation record handles both queries from locations that have no geolocation record and queries from IP addresses that are not mapped to a location; without it, Route 53 returns a no answer response for those queries. Records for every remaining continent would still leave unmapped IP addresses without an answer, and health checks or a lower TTL do not add coverage.
Correct answer: A. Add bw_in_allowance_exceeded to the ethtool subsection of the CloudWatch agent configuration file.
The ENA metric bw_in_allowance_exceeded counts packets queued or dropped because inbound aggregate bandwidth exceeded the maximum for the instance, and the CloudWatch agent collects it only when it is listed in the ethtool subsection of metrics_collected. The pps_allowance_exceeded metric is about the packets-per-second allowance, not bandwidth, and linklocal_allowance_exceeded covers traffic to local proxy services such as the Amazon DNS server and the instance metadata service. Detailed monitoring increases the frequency of existing metrics but does not add the ENA allowance counters.
Correct answer: B. The accountId and region properties set to 111122223333 and eu-west-1 in the widget.
For cross-account cross-Region dashboards, the dashboard body accepts accountId and region parameters at the widget or metric level; without them a widget uses the signed-in account and the current Region. Adding an AccountId dimension does not change which account CloudWatch reads from and would not match the metrics' real dimensions. append_dimensions is a CloudWatch agent setting that only accepts EC2 keys such as InstanceId, and a shared dashboard link does not put the data into the monitoring account's dashboard.
Correct answer: A. Turn on Enhanced Monitoring for the DB instance and review the OS process list and its metrics.
Enhanced Monitoring gathers operating system metrics from an agent on the DB instance in real time and is designed to show how different processes or threads use the CPU. The CloudWatch CPUUtilization metric comes from the hypervisor and gives one value for the whole instance, however it is graphed, so it cannot break usage down by process.
Correct answer: B. Sequential Region concurrency with a failure tolerance of 2 stacks.
Sequential Region concurrency deploys to one Region at a time, and the failure tolerance is the number of failures allowed per Region before CloudFormation stops the operation and cancels it in the remaining Regions, so a value of 2 stops on the third failure. A tolerance of 0 would stop at the first failure, which is stricter than the requirement. Parallel concurrency deploys to all Regions at once, and the maximum concurrent accounts setting controls speed, not when to stop.
Correct answer: B. Use the L3 construct ApplicationLoadBalancedFargateService from the ECS patterns module.
L3 constructs, or patterns, create several resources that are configured to work together with opinionated defaults, and ApplicationLoadBalancedFargateService builds a Fargate service on an ECS cluster fronted by an Application Load Balancer with the fewest inputs. Separate L2 constructs have good defaults but still require each resource to be declared and wired. L1 constructs offer no abstraction, and the IaC generator works from resources that already exist.