Question bank
Professional Cloud DevOps Engineer
41 Professional Cloud DevOps Engineer questions with the correct answer and a written explanation, organised across 5 exam domains.
- 41 questions
- 5 domains
- Foundation level
certilumQuestion index
Jump straight to any question. Answers and explanations are inline.
Applying Site Reliability Engineering Practices
8 questions in this domain
- Question 1Single Choice
In the context of Site Reliability Engineering at Google, which statement correctly describes the primary goal of the discipline?
- AAutomate every operational task
- BMaximize feature velocity without limits
- CEliminate all software failures
- DBalance reliability and feature velocityCorrect answer
Why this is the answer
SRE exists to balance the risk of unreliable systems with the need for fast feature development.
- Question 2Multiple Select
Which two measurement types are used together in Site Reliability Engineering to define a service's reliability target?
- AService Level IndicatorCorrect answer
- BService Level ObjectiveCorrect answer
- CService Level Agreement
- DTotal Cost of Ownership
Why this is the answer
SLIs measure service behavior, and SLOs specify target values for those indicators; SLAs are external contracts.
- Question 3Interactive
Based on the SLO configuration shown above, what reliability percentage is defined as the target for the service?
- A90 percent
- B99 percentCorrect answer
- C99.9 percent
- D99.99 percent
Why this is the answer
The goal value 0.99 expressed as a percentage is 99 percent.
- Question 4Single Choice
In Site Reliability Engineering, what is the term for the acceptable amount of failure a service can experience?
- AError budgetCorrect answer
- BLatency target
- CRollback window
- DCapacity buffer
Why this is the answer
The error budget is the acceptable amount of failure a service can experience while still meeting its SLO.
- Question 5Multiple Select
Within Site Reliability Engineering, which two characteristics are commonly used to identify toil in operational work?
- AManual repetitive tasks with no enduring valueCorrect answer
- BWork that provides no lasting improvementCorrect answer
- COne-time strategic automation projects
- DCreative development of new product features
Why this is the answer
Toil is manual, repetitive work with no long-term benefit; strategic projects and product development are not toil.
- Question 6Interactive
Given the Python code above, what total amount of downtime is allowed for a service with a 99 percent monthly target?
- A25920 secondsCorrect answer
- B43200 seconds
- C86400 seconds
- D259200 seconds
Why this is the answer
At a 99 percent target, one percent of a 30-day month equals 25,920 seconds, or 432 minutes.
- Question 7Single Choice
Which deployment strategy should a team use to gradually shift a small percentage of traffic to a new version?
- ACanary deploymentCorrect answer
- BBlue/green deployment
- CRecreate deployment
- DShadow deployment
Why this is the answer
A canary deployment releases changes to a small subset first and measures impact before the full rollout.
- Question 8Multiple Select
Which two operational practices are most effective for reducing toil and manual workload on an SRE team?
- AAutomating repetitive tasksCorrect answer
- BDocumenting and eliminating toilCorrect answer
- CCreating more manual checklists
- DIncreasing alerting thresholds
Why this is the answer
Automating repetitive tasks and documenting toil help eliminate manual effort and allow SRE teams to focus on reliability improvements.
Bootstrapping and Maintaining a Google Cloud Organization
8 questions in this domain
- Question 9Single Choice
What is the highest-level node in the Google Cloud resource hierarchy?
- AOrganization nodeCorrect answer
- BFolder
- CProject
- DBilling account
Why this is the answer
The organization node is the root of the Google Cloud resource hierarchy. It enables centralized administration, IAM policy assignment, and organization policy enforcement across all folders and projects.
- Question 10Multiple Select
Select all items that are part of the Google Cloud resource hierarchy.
- AOrganization nodeCorrect answer
- BFolderCorrect answer
- CProjectCorrect answer
- DBilling account
Why this is the answer
The Google Cloud resource hierarchy consists of the organization node, folders, and projects. Billing accounts are important resources but are not part of the resource hierarchy.
- Question 11Interactive
A cloud engineer executes the command shown below during initial organization bootstrap. What action does it perform?
- AGrants the Organization Administrator role at the organization levelCorrect answer
- BGrants Project Owner on a project
- CCreates a new service account key
- DAssigns the Billing Administrator role at the organization level
Why this is the answer
The command uses gcloud organizations add-iam-policy-binding to assign the roles/organizationAdmin role to a user at the organization level.
- Question 12Single Choice
Which IAM role should be assigned to a user who is responsible for managing the financial aspects of all Google Cloud accounts?
- ABilling Account AdministratorCorrect answer
- BProject Owner
- COrganization Administrator
- DService Account Admin
Why this is the answer
The Billing Account Administrator role, roles/billing.admin, allows a user to manage billing accounts, payment methods, and related financial settings.
- Question 13Multiple Select
A consultant is starting to bootstrap a new Google Cloud organization. Which two initial steps should be performed?
- ACreate a Cloud Identity account and verify the domainCorrect answer
- BCreate the organization node associated with the Cloud Identity domainCorrect answer
- CCreate a project for every department
- DApply a uniform organization policy to all resources
Why this is the answer
Bootstrapping starts with a Cloud Identity account and then creates the organization node. Projects and policies can be created later, but the Cloud Identity domain and organization node are foundational.
- Question 14Interactive
The following command is used in a bootstrap process. Which resource does it create?
- AA folder named Production under the organizationCorrect answer
- BA project named Production inside the organization
- CAn organization policy called Production
- DA billing account for Production
Why this is the answer
The command gcloud resource-manager folders create creates a folder with the display name Production under the specified organization node.
- Question 15Single Choice
Where are organization policies evaluated in Google Cloud?
- AAt every level of the resource hierarchyCorrect answer
- BOnly on the organization node
- COnly on projects
- DOnly on folders
Why this is the answer
Organization policies are enforced at the organization, folder, and project levels, inheriting down through the resource hierarchy.
- Question 16Multiple Select
Which IAM roles are commonly used to manage a Google Cloud organization? Select all that apply.
- AOrganization AdministratorCorrect answer
- BBilling Account AdministratorCorrect answer
- CProject CreatorCorrect answer
- DCloud Functions Admin
Why this is the answer
Organization Administrator, Billing Account Administrator, and Project Creator are common organization-level or enterprise-level roles. Cloud Functions Admin is service-specific and not typically used to administer the whole organization.
Building and Implementing CI/CD Pipelines
10 questions in this domain
- Question 17Single Choice
In Google Cloud, which fully managed service is designed for building, testing, and deploying artifacts in a CI/CD pipeline?
- ACloud BuildCorrect answer
- BCloud Run
- CCloud Deploy
- DArtifact Registry
Why this is the answer
Cloud Build is Google Cloud's fully managed CI/CD service. Cloud Run is a compute platform, Cloud Deploy orchestrates delivery, and Artifact Registry stores artifacts.
- Question 18Multiple Select
Which of the following are considered core stages in a CI/CD pipeline? Select all that apply.
- AExecuting unit testsCorrect answer
- BCompiling source codeCorrect answer
- CDeploying to an environmentCorrect answer
- DOpening a support ticket
Why this is the answer
A CI/CD pipeline includes source checkout, build or compile, test, and deploy. Opening a support ticket is not a pipeline stage.
- Question 19Interactive
In the Cloud Build config snippet, what does the 'name' field in a build step define?
- AThe container image used to execute the stepCorrect answer
- BThe name of the build result
- CThe Google Cloud project ID
- DThe file path to the Dockerfile
Why this is the answer
In Cloud Build, each step uses a container image specified in the 'name' field to run the defined arguments and commands.
- Question 20Single Choice
Which Google Cloud service is a fully managed artifact registry for storing container images and software packages?
- AArtifact RegistryCorrect answer
- BCloud Functions
- CCloud Source Repositories
- DCloud Run
Why this is the answer
Artifact Registry stores Docker images, Maven packages, and other artifacts. Cloud Source Repositories hosts git repositories, and Cloud Functions is a compute service.
- Question 21Multiple Select
Which of the following can be used to trigger a Cloud Build? Select all that apply.
- AA push to a Cloud Source RepositoryCorrect answer
- BA Pub/Sub messageCorrect answer
- CSubmitting a build with the gcloud commandCorrect answer
- DCreating a Compute Engine VM
Why this is the answer
Cloud Build supports triggers from repository events, Pub/Sub messages, manual gcloud builds, and webhooks. Creating a Compute Engine VM does not trigger a build.
- Question 22Interactive
When the following Cloud Build step runs, what output is written to the build log?
- AHello, Cloud BuildCorrect answer
- Bubuntu
- Cecho
- DBuild success
Why this is the answer
The step runs the 'ubuntu' image with arguments 'echo' and 'Hello, Cloud Build', so the log displays the text 'Hello, Cloud Build'.
- Question 23Single Choice
Which IAM role is required to submit builds in Cloud Build?
- Aroles/cloudbuild.builds.editorCorrect answer
- Broles/storage.objectViewer
- Croles/cloudfunctions.developer
- Droles/iam.serviceAccountAdmin
Why this is the answer
Cloud Build uses IAM roles to control permissions. The roles/cloudbuild.builds.editor role grants permission to create and run builds.
- Question 24Multiple Select
Which practices are recommended for implementing reliable CI/CD pipelines on Google Cloud? Select all that apply.
- AStore pipeline definitions in version controlCorrect answer
- BUse secrets management for sensitive valuesCorrect answer
- CKeep build configuration files in the repositoryCorrect answer
- DDelete build logs to improve security
Why this is the answer
Infrastructure as code, version control, and secrets management are best practices. Retaining build logs is important for auditability and troubleshooting.
- Question 25Interactive
In the Cloud Build config snippet, what is `_REGION` in the substituted variable?
- AA user-defined substitution variableCorrect answer
- BA predefined Cloud Build substitution variable
- CA software package manager
- DA Google Cloud service
Why this is the answer
In Cloud Build, variables beginning with `_` are user-defined substitutions. They are replaced at build time with the values provided.
- Question 26Single Choice
Which Google Cloud service provides managed continuous delivery and rollout strategies such as canary analysis for GKE and Cloud Run?
- ACloud DeployCorrect answer
- BCloud Build
- CCloud Scheduler
- DCloud Composer
Why this is the answer
Cloud Deploy is Google Cloud's managed continuous delivery service for GKE, Cloud Run, and Anthos, supporting progressive rollout strategies such as canary analysis.
Implementing Observability Practices and Troubleshooting
10 questions in this domain
- Question 27Single Choice
Which Google Cloud service provides dashboards, alerting, and monitoring for applications and infrastructure?
- ACentralized metrics, dashboards, and alertingCorrect answer
- BLong-term storage and querying of logs
- CProfiling application code and memory use
- DTracing requests across distributed services
Why this is the answer
Cloud Monitoring collects metrics, builds dashboards, and sends alerts; Cloud Logging handles logs, Cloud Profiler profiles resources, and Cloud Trace measures request latency.
- Question 28Multiple Select
Which services are included in the Google Cloud operations suite?
- ACloud MonitoringCorrect answer
- BCloud LoggingCorrect answer
- CCloud TraceCorrect answer
- DCloud Composer
Why this is the answer
Cloud Monitoring, Cloud Logging, and Cloud Trace are core operations suite components. Cloud Composer orchestrates workflows and is not an observability service.
- Question 29Interactive
Look at the Cloud Logging filter in the code block. What log entries does it select?
- APayments namespace logs with severity ERROR or higherCorrect answer
- BError logs from all Kubernetes namespaces
- CAll logs from the payments cluster
- DInfo and ERROR logs in the payments namespace
Why this is the answer
The filter requires a k8s container in the payments namespace and severity greater than or equal to ERROR, so only matching errors or worse entries are selected.
- Question 30Single Choice
What does Cloud Trace primarily analyze for a request?
- ALatency across the request call path and its spansCorrect answer
- BNumber of log entries generated per service
- CDisk usage of each application instance
- DHTTP status codes returned by load balancers
Why this is the answer
Cloud Trace samples requests and records latency for the full path and individual spans, helping locate performance bottlenecks.
- Question 31Multiple Select
Which practices are recommended when designing Cloud Monitoring alert policies?
- AAlert on user-visible symptoms rather than every internal fluctuation.Correct answer
- BUse SLO burn-rate alerts to detect imminent reliability issues.Correct answer
- CPage engineers for every warning-level condition.
- DRoute all alerts through a single communication channel.
Why this is the answer
Good alerting focuses on symptoms and SLO burn rates. Paging on every warning creates alert fatigue, and a single channel is fragile.
- Question 32Interactive
A log-based counter metric is created with the filter in the code block. What events will it count?
- AGlobal log entries with severity ERROR or higherCorrect answer
- BOnly log entries that contain an error message
- CAll log entries from every resource type
- DGlobal logs with WARNING severity
Why this is the answer
The filter selects global log entries with severity ERROR or above, so the counter metric increments only for those matching entries.
- Question 33Single Choice
What is an uptime check in Google Cloud Monitoring?
- AA synthetic request that verifies availability and responseCorrect answer
- BA policy that auto-scales managed instance groups
- CA method for storing monitoring time series data
- DA log sink that exports health data
Why this is the answer
Uptime checks send synthetic requests to a public endpoint and verify that it is reachable and responding, supporting availability monitoring.
- Question 34Multiple Select
Which capabilities are provided by Cloud Logging?
- AIngest logs from applications and Google Cloud servicesCorrect answer
- BQuery logs with Logs ExplorerCorrect answer
- CExport logs to Cloud StorageCorrect answer
- DRestart unhealthy virtual machines
Why this is the answer
Cloud Logging ingests, stores, queries, and exports logs. It does not execute recovery actions on resources.
- Question 35Interactive
The filter in the code block identifies which monitoring resource and metric?
- ACPU utilization for one Compute Engine instanceCorrect answer
- BDisk throughput for all instances in a project
- CNetwork egress for a specific VPC network
- DMemory usage for a named virtual machine
Why this is the answer
The filter references the Compute Engine CPU utilization metric type and a single instance ID, so it selects CPU data for that instance.
- Question 36Single Choice
What is the main purpose of Cloud Profiler?
- ACPU and memory usage dataCorrect answer
- BLog entries and stack traces
- CRequest latency traces
- DDeployment artifact versions
Why this is the answer
Cloud Profiler uses low-overhead statistical sampling to capture CPU and memory usage from production, helping identify performance issues.
Optimizing Performance and Cost
5 questions in this domain
- Question 37Single Choice
In the context of Google Cloud cost management, what is the primary purpose of setting a Cloud Billing budget?
- AGet alerts before spending exceeds the budgetCorrect answer
- BAutomatically scale down compute resources
- CCreate a new billing account
- DApply committed use discounts to invoices
Why this is the answer
A Cloud Billing budget tracks actual and forecast costs. Alerts help DevOps teams keep spending aligned with expected operating costs.
- Question 38Multiple Select
As a DevOps engineer, which two practices are most effective for reducing compute costs in Google Cloud workloads?
- ARight-size machine types to fit application demandCorrect answer
- BUse preemptible VMs for fault-tolerant batch jobsCorrect answer
- CKeep all VMs running constantly
- DAlways choose the largest machine type available
Why this is the answer
Right-sizing matches capacity to actual utilization, and preemptible VMs cut costs for interruptible batch work. Keeping all VMs running constantly increases idle cost.
- Question 39Interactive
A DevOps Engineer runs the command shown in the code block. What outcome should they expect?
- ASets $500 budget with 50% and 90% alertsCorrect answer
- BEnables automatic shutdown after $500 are spent
- CCreates two separate budgets for 50% and 90%
- DRenames the billing account to Monthly-budget
Why this is the answer
gcloud billing budgets create defines a budget amount and threshold rules. Alerts are sent when forecast or actual spend reaches 50% and 90%.
- Question 40Single Choice
In Google Cloud, what does a Committed Use Discount provide to customers who commit to steady baseline usage for one or three years?
- ADiscounted price for committed resource useCorrect answer
- BMonthly credits for early payments
- CAutomatically converts all VMs to spot VMs
- DApplies only to Cloud Storage transfer operations
Why this is the answer
Committed Use Discounts offer lower prices when you commit to a minimum level of vCPU, memory, or GPU in return for predictable savings.
- Question 41Multiple Select
Which two Google Cloud services are most commonly used to improve response time for globally distributed users?
- ACloud CDNCorrect answer
- BCloud Load BalancingCorrect answer
- CCloud Billing
- DCloud Logging
Why this is the answer
Cloud CDN caches content at edge points; Cloud Load Balancing sends traffic to the closest healthy backend. Together they reduce latency.
Frequently asked questions
Everything about the Professional Cloud DevOps Engineer question bank and how to use it.