McAfee-Secured Website

Exam Bundle

Exam Code: Associate Cloud Engineer

Exam Name Associate Cloud Engineer

Certification Provider: Google

Corresponding Certification: Associate Cloud Engineer

Google Associate Cloud Engineer Bundle $44.99

100% Google Associate Cloud Engineer Guaranteed Success With Testking Training Materials

Get Real Associate Cloud Engineer Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

  • Questions & Answers

    Associate Cloud Engineer Questions & Answers

    391 Questions & Answers

    The ultimate exam preparation tool, Questions & Answers mirrors the actual Associate Cloud Engineer exam while allowing you to hone and measure every aspect of your test-taking ability.

  • Associate Cloud Engineer Video Course

    Associate Cloud Engineer Video Course

    234 Video Lectures

    Associate Cloud Engineer Video Course is developed by Google Professionals to help you pass the Associate Cloud Engineer exam.

    Description

    This course will improve your knowledge and skills required to pass Associate Cloud Engineer exam.
  • Study Guide

    Associate Cloud Engineer Study Guide

    849 PDF Pages

    Developed by industry experts, this 849-page guide spells out in painstaking detail all of the information you need to ace Associate Cloud Engineer exam.

Associate Cloud Engineer Practice Tests

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Total Cost: $194.97
Bundle Price: $149.98

Purchase Individually

  • Questions & Answers

    Questions & Answers

    391 Questions

    $124.99
  • Associate Cloud Engineer Video Course

    Video Course

    234 Video Lectures

    $39.99
  • Study Guide

    Study Guide

    849 PDF Pages

    $29.99

Google Associate Cloud Engineer Practice Test 20

Question 190

A legacy Compute Engine VM writes important logs to a local file and emits custom system telemetry. The Google Cloud dashboard lacks those guest-level signals. Which supported collection tool should the engineer install and configure?

  1. Enable Cloud NAT and assume NAT packets include local application log-file contents
  2. Increase the VM's Cloud Billing budget to enable guest log shipping automatically
  3. Build a Cloud Monitoring dashboard before configuring any guest-file log and metric collection on the VM
  4. Deploy the Google Cloud Ops Agent with appropriate logs and metrics receivers on the VM

Explanation

Correct Answer

D

Ops Agent collects supported guest logs and metrics and exports them to Google Cloud operations products.

Incorrect Answers

A:

NAT traffic translation does not read and ingest operating-system log files.

B:

Budget configuration does not install or configure a guest telemetry agent.

C:

A dashboard cannot visualize signals that the guest has not exported to the Monitoring and Logging backends.

Question 191

During an infrastructure change review, two requirements are being evaluated together. The two controls belong to distinct workload environments. Requirement A: a service must be prevented from a sensitive operation despite inherited allows. Requirement B: an operator needs to know why a principal can access a project. Which option assigns the right action to the first requirement and then the second?

  1. evaluate an applicable IAM deny policy for the prohibited permission; then inspect only VM labels
  2. add another broad allow binding; then inspect effective IAM policies across project ancestors
  3. add another broad allow binding; then inspect only VM labels
  4. evaluate an applicable IAM deny policy for the prohibited permission; then inspect effective IAM policies across project ancestors

Explanation

Correct Answer

D

IAM deny can constrain access even when an allow exists. IAM inheritance propagates permissions through resource ancestry.

Incorrect Answers

A:

The first action is valid: IAM deny can constrain access even when an allow exists. But the second fails: VM labels do not list effective user IAM grants.

B:

The second action is valid: IAM inheritance propagates permissions through resource ancestry. But the first fails: Additional allow grants cannot revoke inherited effective access.

C:

Both requirements remain unmet: Additional allow grants cannot revoke inherited effective access; VM labels do not list effective user IAM grants.

Question 192

A site reliability engineer is reviewing two service requirements from the runbook. The two controls belong to distinct workload environments. The first requirement is that an external CI system needs Google Cloud access using OIDC tokens. The second requirement is that an external workload must not store a Google private key. Which option correctly addresses requirement one and then requirement two?

  1. store a permanent organization Owner key in a CI variable; then download a service account JSON key into source control
  2. set up Workload Identity Federation with a trusted external issuer; then prefer Workload Identity Federation or a supported keyless credential flow
  3. set up Workload Identity Federation with a trusted external issuer; then download a service account JSON key into source control
  4. store a permanent organization Owner key in a CI variable; then prefer Workload Identity Federation or a supported keyless credential flow

Explanation

Correct Answer

B

Federation exchanges trusted external identity for short-lived access. Federation reduces static key handling.

Incorrect Answers

A:

Both requirements remain unmet: Owner private keys increase blast radius and rotation burden; Committed service account private keys create security exposure.

C:

The first action is valid: Federation exchanges trusted external identity for short-lived access. But the second fails: Committed service account private keys create security exposure.

D:

The second action is valid: Federation reduces static key handling. But the first fails: Owner private keys increase blast radius and rotation burden.

Question 193

A cloud administrator is implementing a two-part operational checklist before a scheduled release. These two items involve different cloud resources. First, a developer must switch gcloud operations from the sandbox project to production. Also, an operator must verify which identity gcloud uses for local commands. Which ordered action pair resolves the first and second items respectively?

  1. set the active gcloud project to the production project ID; then inspect the active gcloud account configuration
  2. set the active gcloud project to the production project ID; then inspect only the Cloud Storage bucket location
  3. change the default Compute Engine zone only; then inspect the active gcloud account configuration
  4. change the default Compute Engine zone only; then inspect only the Cloud Storage bucket location

Explanation

Correct Answer

A

Project selection controls the resource project for commands. Active account configuration identifies CLI caller credentials.

Incorrect Answers

B:

The first action is valid: Project selection controls the resource project for commands. But the second fails: Storage location cannot identify gcloud authentication identity.

C:

The second action is valid: Active account configuration identifies CLI caller credentials. But the first fails: Changing zones does not change the active project.

D:

Both requirements remain unmet: Changing zones does not change the active project; Storage location cannot identify gcloud authentication identity.

Question 194

A budget has a Pub/Sub topic configured, and the consuming service receives status updates even when no new email threshold alert occurred. Operators suspect duplicate alerts. What explains this behavior?

  1. Pub/Sub delivers only when a budget threshold crosses exactly 100 percent
  2. Every Pub/Sub message means Cloud Billing automatically stopped a resource instance
  3. Budget Pub/Sub notifications report periodic budget status separately from threshold-triggered email alerts
  4. Interpret each Pub/Sub budget status message as equivalent to a threshold-triggered email notification

Explanation

Correct Answer

C

Pub/Sub budget updates can occur multiple times daily independently of threshold email events.

Incorrect Answers

A:

Programmatic budget notifications are not restricted to the single full-budget threshold.

B:

Notifications provide spend state; service stoppage requires separately implemented authorized automation.

D:

Programmatic status messages and email threshold notifications use different delivery cadence and semantics.

Question 195

An application uses a managed instance group but each instance must keep a unique persistent data disk across supported repairs and updates. The current group treats all instance disks as disposable. What instance-group capability should be assessed?

  1. Configure a stateful managed instance group with supported stateful disks and preservation policies
  2. Add more stateless autoscaling replicas and assume each VM's local data follows its replacement
  3. Replace all persistent disks with Local SSD and disable autohealing
  4. Use a stateless MIG with persistent data disks but leave them unmarked as preserved state in instance-group configuration

Explanation

Correct Answer

A

Stateful MIG settings preserve designated per-instance state across supported lifecycle events.

Incorrect Answers

B:

Additional replicas do not preserve unique per-instance data across normal stateless recreation.

C:

Local SSD is ephemeral and disabling recovery does not make unique data durable.

D:

Persistent disk media alone does not instruct stateless MIG repair to retain the unique instance-to-disk mapping.

Question 196

During an infrastructure change review, two requirements are being evaluated together. These two items involve different cloud resources. First, an app needs a shared POSIX-style file system across VMs. Also, a database requires reliable managed backups and point-in-time recovery. Which option assigns the right action to the first requirement and then the second?

  1. provision Filestore or another matching managed file service; then disable transaction logging permanently
  2. use object-only Cloud Storage as a mounted POSIX replacement without constraints; then configure available Cloud SQL automated backups and PITR settings
  3. use object-only Cloud Storage as a mounted POSIX replacement without constraints; then disable transaction logging permanently
  4. provision Filestore or another matching managed file service; then configure available Cloud SQL automated backups and PITR settings

Explanation

Correct Answer

D

Managed file storage exposes file shares to clients. Backup and recovery settings preserve restoration options.

Incorrect Answers

A:

The first action is valid: Managed file storage exposes file shares to clients. But the second fails: Disabling logs undermines point-in-time recovery.

B:

The second action is valid: Backup and recovery settings preserve restoration options. But the first fails: Object APIs are not universally equivalent to POSIX file systems.

C:

Both requirements remain unmet: Object APIs are not universally equivalent to POSIX file systems; Disabling logs undermines point-in-time recovery.

Question 197

During a routine project audit, engineers discover two unrelated control gaps. Each requirement refers to its own target workload. First, public UDP game traffic must preserve client IP through backend handling. Also, a shared network platform must host common subnets for service projects. Which ordered remediation addresses the first gap and then the second gap?

  1. use an HTTPS Application Load Balancer; then attach service projects to a Shared VPC host
  2. use an HTTPS Application Load Balancer; then create an unrelated independent default VPC in each project
  3. use an appropriate external passthrough Network Load Balancer; then attach service projects to a Shared VPC host
  4. use an appropriate external passthrough Network Load Balancer; then create an unrelated independent default VPC in each project

Explanation

Correct Answer

C

Passthrough Network Load Balancing supports UDP and preserves packet addresses. Shared VPC centralizes network ownership and administration.

Incorrect Answers

A:

The second action is valid: Shared VPC centralizes network ownership and administration. But the first fails: HTTP load balancing does not directly handle arbitrary UDP.

B:

Both requirements remain unmet: HTTP load balancing does not directly handle arbitrary UDP; Separate networks do not share the host-project VPC.

D:

The first action is valid: Passthrough Network Load Balancing supports UDP and preserves packet addresses. But the second fails: Separate networks do not share the host-project VPC.

Question 198

The architecture review has recorded two mandatory controls for the next cloud deployment. These two items involve different cloud resources. First, teams need declarative repeatable provisioning from versioned files. Also, Kubernetes package installation must use a versioned chart. Which ordered pair configures the first control and then the second control?

  1. use Terraform configuration under version control; then edit the Cloud Billing account display name
  2. record resource settings only in screenshots; then deploy the application through Helm with the intended chart and values
  3. record resource settings only in screenshots; then edit the Cloud Billing account display name
  4. use Terraform configuration under version control; then deploy the application through Helm with the intended chart and values

Explanation

Correct Answer

D

Terraform configuration records desired infrastructure state. Helm manages templated Kubernetes application releases.

Incorrect Answers

A:

The first action is valid: Terraform configuration records desired infrastructure state. But the second fails: Billing metadata cannot install Kubernetes workloads.

B:

The second action is valid: Helm manages templated Kubernetes application releases. But the first fails: Screenshots cannot enforce reproducible declarative infrastructure.

C:

Both requirements remain unmet: Screenshots cannot enforce reproducible declarative infrastructure; Billing metadata cannot install Kubernetes workloads.

Question 199

An application may run only on workers in a licensed node pool labeled workload=licensed. The scheduler currently places replicas on general-purpose nodes as well. Which placement constraint expresses the requirement?

  1. Set required node affinity or a supported node selector matching workload=licensed
  2. Add tolerations for possible node taints without using any required node selector or affinity constraint
  3. Assign a Cloud Billing budget to the licensed pool and rely on it for scheduling
  4. Rename the Kubernetes Service to licensed without changing pod scheduling specifications

Explanation

Correct Answer

A

Required node selection constrains pod placement to nodes carrying the specified labels.

Incorrect Answers

B:

Tolerations allow eligibility but do not restrict pods exclusively to the licensed node pool.

C:

Cost budgets have no scheduling predicate for individual Kubernetes workloads.

D:

Service names do not restrict which worker nodes the scheduler can select.

Question 200

Two clients update the same Cloud Storage object concurrently, and one must not silently overwrite a newer version written after it last read the object. Which request safeguard helps prevent a lost update?

  1. Assign a static public IP address to each client so writes are automatically serialized
  2. Use a generation-match precondition tied to the observed object generation during the write
  3. Use only an object metageneration precondition while ignoring changes to the content generation being overwritten
  4. Increase the client's VM CPU count because faster writes cannot conflict with concurrent writers

Explanation

Correct Answer

B

Generation preconditions implement optimistic concurrency, rejecting stale writes against changed objects.

Incorrect Answers

A:

Source addressing cannot replace Cloud Storage conditional write preconditions during an overlapping object update.

C:

Metageneration detects metadata state, whereas content generation-match guards the intended object version against lost writes.

D:

Compute speed does not establish an atomic guard against object version changes.