Answer:  https://spinnaker.io/guides/user/canary/best-practices/#compare-canary-against-baseline-not-against-production You might be tempted to compare the canary deployment against your current production deployment. Instead always compare the canary against an equivalent baseline, deployed at the same time. The baseline uses the same version and configuration that is currently running in production, but is otherwise identical to the canary: Same time of deployment Same size of deployment Same type and amount of traffic In this way, you control for version and configuration only, and you reduce factors that could affect the analysis, like the cache warmup time, the heap size, and so on.

Identify recommended reliability improvements to the service to be completed before handover. - If the app doesn’t meet the set SLOs, improvements need to be met or a revised SLO need to be agreed with relevant stakeholders.

This question is a little bit strange, but first we need to remove the invalid answers B: Incorrect An admission controller is a piece of code that intercepts requests to the Kubernetes API server prior to persistence of the object, but after the request is authenticated and authorized. (its for security but not "enforce approvals") C: Incorrect, we need to "enforce approvals" roles apply in the cluster and Ops always could push to production without approval. A: Incorrect, for me this answer sound well but this does not sound that an answer for a gcp exam and this do not enforce the use of the pipeline. D: Correct, they cannot push code to production without approval because their images are not signed.

Use Stackdriver Profiler to visualize the resources utilization throughout the application. High CPU usage can most definitely cause dropped or ignored connections. The database engine and underlying OS are fighting for resources and aren't able to respond to the connection in time. Finding out why the query is eating so much CPU usage and optimizing it. https://stackoverflow.com/questions/69919454/high-cpu-usage-on-cloud-sql-causing-timeouts # Cloud Profiler is a statistical, low-overhead profiler that continuously gathers CPU usage and memory-allocation information (supported profile types: CPU time, Heap, Allocated heap, Contention, Threads, Wall time) from your production applications. It attributes that information to the source code that generated it, helping you identify the parts of your application that are consuming the most resources, and otherwise illuminating your applications performance characteristics. https://cloud.google.com/profiler/docs/about-profiler

Google Cloud courses say: "Monitoring dashboard viewer can view dashboard settings while monitoring viewer can view charts and dashboards", so i think I need to see Dashboards and chart content, not settings only

The Horizontal Pod Autoscaler changes the shape of your Kubernetes workload by automatically increasing or decreasing the number of Pods in response to the workload's CPU or memory consumption B: Incorrect. It is not based on the CPU its based on the workload C: No, Hope is not an strategy D: No, have more resource than needed

A is correct choice. Why D is not correct, because even if you break steps into small chunks, you are just increasing the number of steps, and not solving the actual problem of reducing execution time. If step 1 takes 5min it will take 5min, if you break it down into three steps, combined Step 1+2+3 = 5min again, and they don't run in parallel. What did you solve? Nothing. If you cache ie. store intermediate artefacts in cloud storage, you do not need to recompile it in next steps, saves time and cost.

The keyword is "make several requests to dependent app". So you need trace for it. Cloud Trace Find performance bottlenecks in production. Cloud Profiler Continuous CPU and heap profiling to improve performance and reduce costs.

Answer Use Cloud Build to include the release version tag in the application image, Cloud Build provides the following default substitutions: $TAG_NAME: build.Source.RepoSource.Revision.TagName

Write answer is A as stated in the documentation here https://cloud.google.com/logging/docs/export/configure_export_v2#before-you-begin "Note that this guide describes creating and managing sinks at the Cloud project level, but you can create sinks (non-aggregated) for billing accounts, folders, and organizations. As you get started, ensure the following: You have a Google Cloud project with logs that you can see in the Logs Explorer. You have one of the following IAM roles for the source Cloud project from which you're routing logs. Owner (roles/owner) Logging Admin (roles/logging.admin) Logs Configuration Writer (roles/logging.configWriter) The permissions contained in these roles allow you to create, delete, or modify sinks. For information on setting IAM roles, see the Logging Access control guide."