100% Latest & Updated Cisco 350-901 Practice Test Questions, Exam Dumps & Verified Answers!
30 Days Free Updates, Instant Download!
Question 150
The client floods the API after throttling and loses monitoring data. Independently, reviewers flag this potentially unsafe practice for assessment: 'Replace failed sites with zero health alerts and label all healthy'. Which verification addresses that practice without assuming it caused the incident?
- Repeat every create request after a timeout without checking completion
- Inspect returned Retry-After and retry scheduler
- Review retry count and elapsed-time bounds
- Review missing-site report and data confidence flags
Explanation
Correct Answer
D:Silent substitution hides monitoring gaps. The intended behavior is: Partial coverage should not masquerade as a clean report.
Incorrect Answers
A:The initial write might have succeeded despite lost response.
B:This checks respect rate-limit guidance, not the named failure. Server pacing guidance prevents immediate repeat throttling.
C:This checks use bounded retry budget, not the named failure. Bounded exponential delay avoids synchronized retry storms.
Question 151
A Python automation service changes ACLs across sites over unreliable WAN links. The specific implementation objective is to set explicit request timeout. Which configuration best meets the stated requirement?
- Interpret all timeouts as proof no change was made
- Rerun every successful change blindly after one failed device
- Catch every exception and report configuration completed
- Use a justified connect/read timeout per API call
Explanation
Correct Answer
D:Bounded waits prevent indefinitely hung tasks.
Incorrect Answers
A:A device can process a request while the response is lost.
B:Unscoped replay can duplicate disruptive operations.
C:Broad exception swallowing creates false success.
Question 152
A Python automation service changes ACLs across sites over unreliable WAN links. The specific implementation objective is to classify network versus HTTP failures. Which design action is most appropriate for this case?
- Interpret all timeouts as proof no change was made
- Distinguish DNS/TLS/connection failures from an HTTP response error
- Leave HTTP requests with unlimited blocking waits
- Rerun every successful change blindly after one failed device
Explanation
Correct Answer
B:Different failure classes need different recovery actions.
Incorrect Answers
A:A device can process a request while the response is lost.
C:One unresponsive network path can stall the workflow.
D:Unscoped replay can duplicate disruptive operations.
Question 153
A Python automation service changes ACLs across sites over unreliable WAN links. The specific implementation objective is to keep partial success traceable. What should the engineer implement for this specific objective?
- Catch every exception and report configuration completed
- Record which devices changed, failed or were skipped
- Interpret all timeouts as proof no change was made
- Leave HTTP requests with unlimited blocking waits
Explanation
Correct Answer
B:A transaction log supports safe resumption.
Incorrect Answers
A:Broad exception swallowing creates false success.
C:A device can process a request while the response is lost.
D:One unresponsive network path can stall the workflow.
Question 154
A Python automation service changes ACLs across sites over unreliable WAN links. The specific implementation objective is to stop on unsafe uncertainty. Which adjustment would satisfy this particular constraint?
- Catch every exception and report configuration completed
- Rerun every successful change blindly after one failed device
- Leave HTTP requests with unlimited blocking waits
- Pause destructive follow-on actions when a timed-out write's result is unknown
Explanation
Correct Answer
D:A timeout does not prove the server rolled back.
Incorrect Answers
A:Broad exception swallowing creates false success.
B:Unscoped replay can duplicate disruptive operations.
C:One unresponsive network path can stall the workflow.
Question 155
A Python automation service changes ACLs across sites over unreliable WAN links. The deployment plan must both (1) classify network versus HTTP failures and (2) stop on unsafe uncertainty. Which TWO actions together satisfy those requirements? Choose TWO.
- Distinguish DNS/TLS/connection failures from an HTTP response error
- Catch every exception and report configuration completed
- Interpret all timeouts as proof no change was made
- Rerun every successful change blindly after one failed device
- Pause destructive follow-on actions when a timed-out write's result is unknown
Explanation
Correct Answers
A:Different failure classes need different recovery actions.
E:A timeout does not prove the server rolled back.
Incorrect Answers
B:Broad exception swallowing creates false success.
C:A device can process a request while the response is lost.
D:Unscoped replay can duplicate disruptive operations.
Question 156
A Python automation service changes ACLs across sites over unreliable WAN links. Three conditions must be met: (1) set explicit request timeout; (2) classify network versus HTTP failures; and (3) stop on unsafe uncertainty. Which THREE actions form the narrowest complete plan? Choose THREE.
- Distinguish DNS/TLS/connection failures from an HTTP response error
- Interpret all timeouts as proof no change was made
- Leave HTTP requests with unlimited blocking waits
- Pause destructive follow-on actions when a timed-out write's result is unknown
- Rerun every successful change blindly after one failed device
- Use a justified connect/read timeout per API call
Explanation
Correct Answers
A:Different failure classes need different recovery actions.
D:A timeout does not prove the server rolled back.
F:Bounded waits prevent indefinitely hung tasks.
Incorrect Answers
B:A device can process a request while the response is lost.
C:One unresponsive network path can stall the workflow.
E:Unscoped replay can duplicate disruptive operations.
Question 157
One stalled device blocks the entire maintenance window because a requests call has no timeout. Independently, reviewers flag this potentially unsafe practice for assessment: 'Leave HTTP requests with unlimited blocking waits'. Which inspection most directly evaluates that practice?
- Inspect configured request timeout and elapsed calls
- Read back target state before another mutation
- Inspect exception types and per-device outcomes
- Rerun every successful change blindly after one failed device
Explanation
Correct Answer
A:One unresponsive network path can stall the workflow. The intended behavior is: Bounded waits prevent indefinitely hung tasks.
Incorrect Answers
B:This checks stop on unsafe uncertainty, not the named failure. A timeout does not prove the server rolled back.
C:This checks classify network versus HTTP failures, not the named failure. Different failure classes need different recovery actions.
D:Unscoped replay can duplicate disruptive operations.
Question 158
One stalled device blocks the entire maintenance window because a requests call has no timeout. Independently, reviewers flag this potentially unsafe practice for assessment: 'Catch every exception and report configuration completed'. What diagnostic evidence should first be collected to assess that practice?
- Compare completed-device records with retry targets
- Inspect exception types and per-device outcomes
- Leave HTTP requests with unlimited blocking waits
- Read back target state before another mutation
Explanation
Correct Answer
B:Broad exception swallowing creates false success. The intended behavior is: Different failure classes need different recovery actions.
Incorrect Answers
A:This checks keep partial success traceable, not the named failure. A transaction log supports safe resumption.
C:One unresponsive network path can stall the workflow.
D:This checks stop on unsafe uncertainty, not the named failure. A timeout does not prove the server rolled back.
Question 159
One stalled device blocks the entire maintenance window because a requests call has no timeout. Independently, reviewers flag this potentially unsafe practice for assessment: 'Rerun every successful change blindly after one failed device'. Which focused check best tests this specific assumption?
- Compare completed-device records with retry targets
- Catch every exception and report configuration completed
- Inspect configured request timeout and elapsed calls
- Read back target state before another mutation
Explanation
Correct Answer
A:Unscoped replay can duplicate disruptive operations. The intended behavior is: A transaction log supports safe resumption.
Incorrect Answers
B:Broad exception swallowing creates false success.
C:This checks set explicit request timeout, not the named failure. Bounded waits prevent indefinitely hung tasks.
D:This checks stop on unsafe uncertainty, not the named failure. A timeout does not prove the server rolled back.
Question 160
One stalled device blocks the entire maintenance window because a requests call has no timeout. Independently, reviewers flag this potentially unsafe practice for assessment: 'Interpret all timeouts as proof no change was made'. Which verification addresses that practice without assuming it caused the incident?
- Read back target state before another mutation
- Rerun every successful change blindly after one failed device
- Inspect configured request timeout and elapsed calls
- Inspect exception types and per-device outcomes
Explanation
Correct Answer
A:A device can process a request while the response is lost. The intended behavior is: A timeout does not prove the server rolled back.
Incorrect Answers
B:Unscoped replay can duplicate disruptive operations.
C:This checks set explicit request timeout, not the named failure. Bounded waits prevent indefinitely hung tasks.
D:This checks classify network versus HTTP failures, not the named failure. Different failure classes need different recovery actions.
Designing, Deploying, and Managing Network Automation Systems (350-901) Practice Tests