Google Associate Cloud Engineer Practice Test 32
Question 310
A migration input is a published HTTP URL-list file naming retrievable objects, not an S3 bucket or a local directory. Which Storage Transfer Service source method matches this inventory?
- Configure a POSIX on-premises agent pool and point it to the web server's hostname as though it were a mounted filesystem.
- Use the supported URL-list transfer source with a correctly formatted accessible list and destination permissions.
- Enable Cloud VPN alone; a tunnel will discover and download every link on the public webpage automatically.
- Create a BigQuery external table for the URL text; query execution streams each asset to the target bucket by default.
Explanation
Correct Answer
B:Use the supported URL-list transfer source with a correctly formatted accessible list and destination permissions. Storage Transfer Service supports transfers using a URL list in addition to cloud buckets and filesystem sources, subject to format and access requirements.
Incorrect Answers
A:Configure a POSIX on-premises agent pool and point it to the web server's hostname as though it were a mounted filesystem. POSIX transfer agents read actual filesystem paths. An ordinary hosted URL manifest is an agentless URL-list input, not a mounted path.
C:Enable Cloud VPN alone; a tunnel will discover and download every link on the public webpage automatically. VPN connectivity does not turn a collection of HTTP links into an object-copy transfer job or perform retries.
D:Create a BigQuery external table for the URL text; query execution streams each asset to the target bucket by default. A BigQuery external table interprets data for queries; it does not automatically copy referenced HTTP objects into Cloud Storage.
Question 311
An administrator begins a large transfer without checking source and destination IAM, causing most objects to fail. What should happen before retrying?
- Configure a scheduled Storage Transfer Service job from the source Cloud Storage bucket to the destination bucket.
- Verify source read access, destination write access and transfer-service identity permissions.
- Rely on bucket labels alone to enforce object retention or deletion restrictions.
- Move objects into an unrelated project without first checking their location and policies.
Explanation
Correct Answer
B:Transfer jobs need authorized access on both sides; a successful transfer-job creation isn't proof data-plane permissions are sufficient.
Incorrect Answers
A:The option does not enforce the required object location, access or lifecycle behavior. The specific requirement concerns the object's required data or access behavior. The appropriate action here is: Verify source read access, destination write access and transfer-service identity permissions.
C:The option does not enforce the required object location, access or lifecycle behavior. The necessary distinction is the object's required data or access behavior. The appropriate action here is: Verify source read access, destination write access and transfer-service identity permissions.
D:The option does not enforce the required object location, access or lifecycle behavior. The case instead depends on the object's required data or access behavior. The appropriate action here is: Verify source read access, destination write access and transfer-service identity permissions.
Question 312
A migration completes with fewer objects than the source manifest and the team wants integrity evidence before cutover. What should it validate?
- Create an agentless Storage Transfer Service job from Amazon S3 to Cloud Storage with approved source and destination access.
- Reconcile object counts, size/checksum evidence and transfer-job errors against the approved source manifest.
- Increase the caller VM memory without modifying any bucket or object behavior.
- Turn off object versioning and assume previous generations remain available indefinitely.
Explanation
Correct Answer
B:Large transfers require completion and integrity validation, not only a green initial job-submission status.
Incorrect Answers
A:The option does not enforce the required object location, access or lifecycle behavior. The necessary distinction is the object's required data or access behavior. The appropriate action here is: Reconcile object counts, size/checksum evidence and transfer-job errors against the approved source manifest.
C:Memory sizing may change capacity but cannot replace the specified platform feature. The case instead depends on the object's required data or access behavior.
D:Object versioning and the requested retention or deletion safeguards are separate mechanisms. It leaves the object's required data or access behavior unaddressed. The appropriate action here is: Reconcile object counts, size/checksum evidence and transfer-job errors against the approved source manifest.
Question 313
An object migration has copied most of its files but the transfer operation reports several failed copies. Operators need to inspect failures and retry without replacing every object already copied. What settings and evidence matter?
- Delete every destination object before retrying so all successfully migrated data is recopied without inspecting failure details.
- Increase the project network tier; this guarantees Storage Transfer Service will overwrite only objects that failed.
- Use the Cloud SQL error log for PostgreSQL WAL replication, because object-copy failures are database log records.
- Review the transfer operation status and configured COPY failure logs, then use appropriate overwrite and retry options for the affected objects.
Explanation
Correct Answer
D:Review the transfer operation status and configured COPY failure logs, then use appropriate overwrite and retry options for the affected objects. Transfer operations and enabled per-action Cloud Logging identify failed copies; supported job options control replacement and repeat-transfer behavior.
Incorrect Answers
A:Delete every destination object before retrying so all successfully migrated data is recopied without inspecting failure details. Deleting completed objects discards successful work and can create avoidable downtime. The request specifically calls for selective recovery rather than full recopy.
B:Increase the project network tier; this guarantees Storage Transfer Service will overwrite only objects that failed. Network tier does not define Storage Transfer Service overwrite behavior or identify which source objects failed.
C:Use the Cloud SQL error log for PostgreSQL WAL replication, because object-copy failures are database log records. Cloud SQL logs describe database operations and do not contain the authoritative per-object outcomes of a Storage Transfer Service job.
Question 314
A globally deployed transactional application needs Spanner data replicated across multiple Google Cloud regions. The team currently selected a regional instance configuration. Which Spanner setting meets the multi-region requirement?
- Add another read-only application server in a different region while leaving the Spanner instance configured for one region.
- Use a different client-side SQL query timeout; this changes the geographic placement of Spanner replicas.
- Select an eligible Spanner multi-region instance configuration with an approved replication topology and data residency.
- Increase the regional instance processing units only, because more compute nodes automatically change the location topology.
Explanation
Correct Answer
C:Select an eligible Spanner multi-region instance configuration with an approved replication topology and data residency. Spanner instance configurations determine replication placement. An eligible multi-region configuration distributes replicas across more than one region to address the stated geography requirement.
Incorrect Answers
A:Add another read-only application server in a different region while leaving the Spanner instance configured for one region. Application compute placement does not change where the Spanner instance maintains its database replicas. The instance configuration still remains regional.
B:Use a different client-side SQL query timeout; this changes the geographic placement of Spanner replicas. Client timeouts affect request behavior but cannot alter geographic placement or replication of the Spanner database.
D:Increase the regional instance processing units only, because more compute nodes automatically change the location topology. More processing units scale capacity within the chosen instance configuration. They do not convert regional replication into multi-region replication.
Question 315
A critical time-series application uses one Bigtable cluster in a single region and needs inter-region resilience. What deployment option fits?
- Use Bigtable with row-key design optimized for the expected read/write patterns.
- Turn on a backup schedule and assume it also repairs live key or access patterns.
- Rebuild only the client VM image while leaving the Bigtable table configuration unchanged.
- Add an appropriate Bigtable replication cluster in another approved region and test failover behavior.
Explanation
Correct Answer
D:Bigtable supports replication across clusters and regions for higher availability and resilience.
Incorrect Answers
A:It does not correct the Bigtable row-key, cluster or data-access requirement. The specific requirement concerns the required key-oriented data access.
B:Backups protect recovery points but do not automatically change live database behavior. The necessary distinction is the required key-oriented data access.
C:It does not correct the Bigtable row-key, cluster or data-access requirement. The case instead depends on the required key-oriented data access.
Question 316
A relational Cloud SQL deployment needs recovery after operator error deletes important rows, but it has only daily full backups. What capability should be considered?
- Enable supported point-in-time recovery with sufficient retained transaction logs and test restore procedures.
- Deploy Cloud SQL for MySQL and configure backup and recovery settings.
- Modify the client connection timeout but leave the underlying Cloud SQL setting unchanged.
- Move the database to a Compute Engine VM and assume the managed database behavior remains unchanged.
Explanation
Correct Answer
A:Point-in-time recovery reduces the data loss window relative to relying only on coarse periodic backups.
Incorrect Answers
B:That change does not satisfy the database engine, read/write or recovery constraint. The necessary distinction is the database engine and managed SQL operation.
C:Timeouts control waiting time, not the missing platform feature or access right. The case instead depends on the database engine and managed SQL operation.
D:That change does not satisfy the database engine, read/write or recovery constraint. It leaves the database engine and managed SQL operation unaddressed. The appropriate action here is: Enable supported point-in-time recovery with sufficient retained transaction logs and test restore procedures.
Question 317
A bucket currently uses a single-region location. A media application must continue reading critical objects when an entire Google Cloud region becomes unavailable. Which new bucket placement addresses cross-region data resilience?
- Keep the single-region bucket and create two object prefixes; prefixes automatically replicate objects to another region.
- Select a supported dual-region or multi-region bucket location and review its replication, residency, and failover characteristics.
- Add more VM instance groups in the same region; compute replicas make a single-region bucket cross-region redundant.
- Switch the bucket storage class from Standard to Nearline without changing its single-region location.
Explanation
Correct Answer
B:Select a supported dual-region or multi-region bucket location and review its replication, residency, and failover characteristics. A dual-region stores replicas in two specified regions; multi-region placement supplies cross-region redundancy subject to its location and replication behavior.
Incorrect Answers
A:Keep the single-region bucket and create two object prefixes; prefixes automatically replicate objects to another region. Object prefixes are names within the same bucket and do not create a second geographic replica of that bucket's data.
C:Add more VM instance groups in the same region; compute replicas make a single-region bucket cross-region redundant. Additional compute instances cannot overcome a storage bucket's placement in an unavailable region if the data remains single-region.
D:Switch the bucket storage class from Standard to Nearline without changing its single-region location. Storage class determines access/cost behavior; changing class alone does not alter the bucket's geographic location type.
Question 318
An event pipeline writes immutable JSON payloads under unique object names. Several geographic services must fetch and retry downloads over HTTPS without mounting NFS or changing objects in place. Which interface should clients use?
- Mount a shared Cloud SQL primary data disk in read/write mode on every regional service instance.
- Use the Cloud Storage object HTTP/API interface with stable object names, appropriate IAM, and integrity checks.
- Use Kubernetes node-local emptyDir volumes as globally durable storage; content survives every node replacement.
- Put the payload in a DNS A record because DNS resolves object names and provides immutable file bodies.
Explanation
Correct Answer
B:Use the Cloud Storage object HTTP/API interface with stable object names, appropriate IAM, and integrity checks. Cloud Storage provides remotely addressable object data, authenticated reads, and object integrity fields. Stable object names support repeatable downloads.
Incorrect Answers
A:Mount a shared Cloud SQL primary data disk in read/write mode on every regional service instance. A relational database disk is not shared multi-region object storage; simultaneously mounting block storage read/write across independent servers is unsuitable.
C:Use Kubernetes node-local emptyDir volumes as globally durable storage; content survives every node replacement. emptyDir is ephemeral storage tied to a Pod's lifecycle and cannot be treated as globally durable object storage.
D:Put the payload in a DNS A record because DNS resolves object names and provides immutable file bodies. DNS records map names to network information and do not store or serve arbitrary JSON objects via the Cloud Storage API.
Question 319
A project has millions of obsolete objects and the team wants automatic retention-based transitions rather than a nightly script. Which service feature fits?
- Use a Filestore NFS share as though it implemented Cloud Storage object versioning and generation semantics.
- Use a VM-local Persistent Disk as the only copy of the object data without the required bucket controls.
- Use Cloud Storage Object Lifecycle Management with reviewed conditions and class or delete actions.
- Give anonymous users Storage Admin as a shortcut to object access management.
Explanation
Correct Answer
C:Bucket lifecycle rules automate object transitions and deletion under defined conditions.
Incorrect Answers
A:Object versioning and the requested retention or deletion safeguards are separate mechanisms. The specific requirement concerns retention or recoverability controls. The appropriate action here is: Use Cloud Storage Object Lifecycle Management with reviewed conditions and class or delete actions.
B:The option does not enforce the required object location, access or lifecycle behavior. The necessary distinction is retention or recoverability controls. The appropriate action here is: Use Cloud Storage Object Lifecycle Management with reviewed conditions and class or delete actions.
D:Broad Storage Admin access introduces privileges beyond the stated task. The case instead depends on retention or recoverability controls.
Question 320
A migration requires copying existing objects to a second cloud bucket but must not overwrite newer destination generations. What transfer design decision matters?
- Create an agentless Storage Transfer Service job from Amazon S3 to Cloud Storage with approved source and destination access.
- Choose supported overwrite conditions in the managed transfer job and validate with a small conflict test.
- Move objects into an unrelated project without first checking their location and policies.
- Change only the storage class and assume it adds recovery, locking or IAM rights.
Explanation
Correct Answer
B:Transfer overwrite policies affect whether an older source replaces newer destination object content.
Incorrect Answers
A:The option does not enforce the required object location, access or lifecycle behavior. The necessary distinction is the object's required data or access behavior. The appropriate action here is: Choose supported overwrite conditions in the managed transfer job and validate with a small conflict test.
C:The option does not enforce the required object location, access or lifecycle behavior. The case instead depends on the object's required data or access behavior. The appropriate action here is: Choose supported overwrite conditions in the managed transfer job and validate with a small conflict test.
D:Storage class changes cost and access behavior, not the required policy by itself. It leaves the object's required data or access behavior unaddressed. The appropriate action here is: Choose supported overwrite conditions in the managed transfer job and validate with a small conflict test.