Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Professional Cloud DevOps Engineer practice bank has 1,455 more like it. Try 10 free: instant score, no account.
A DevOps engineer is investigating intermittent HTTP 504 Gateway Timeout errors across a distributed microservices application composed of a Cloud Run frontend service and a Cloud SQL backend instance. The application logs thousands of events per minute in Cloud Logging, making manual correlation challenging.
The engineer needs to formulate an effective natural language prompt in Gemini Cloud Assist to analyze the log data, correlate events across both services, identify performance degradation trends, and determine the root cause.
Which prompting strategy should the engineer use to achieve the most precise and actionable analysis from Gemini Cloud Assist?
This approach uses Gemini Database Migration or Cloud SQL Studio code generation prompts to generate SQL schemas and DDL scripts rather than conducting log analysis.
The DevOps team was actively migrating a schema from SQL Server to AlloyDB or PostgreSQL and needed AI assistance to fix object conversion errors.
Single-tier prompting isolates the investigation to the boundary where symptoms first appear (the frontend Cloud Run HTTP 504 status codes) while excluding downstream infrastructure metadata and logs.
The root cause was definitively isolated to the Cloud Run container configuration (such as memory limits or cold starts) with zero dependency on backend data stores.
Broad prompting relies on generic, high-level natural language questions (such as asking the AI assistant to inspect an entire project for any failure) without bounding the search space by resource, time, or error signature.
You were conducting an initial high-level exploratory audit of overall project health rather than actively diagnosing an acute performance degradation across specific services.
Precise prompting involves giving Gemini Cloud Assist specific contextual anchors—including exact resource identifiers, timestamps or operational windows, observed error indicators, and explicit analytical objectives. When troubleshooting complex multi-tier systems, anchoring the model with specific resource contexts allows it to synthesize relevant logs, metrics, and configuration data effectively.
Iterative and detailed prompt formulation provides the model with the necessary parameters to narrow down millions of log entries into a focused timeline of cause and effect, outperforming broad, ambiguous prompts.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated