Bitzu

Associate Cloud Engineer ยท Diagnostics, health and recommendations

Diagnosing application issues on Google Cloud

Cloud Trace for latency, Cloud Profiler for CPU/memory hot spots, Error Reporting, Query Insights and index advisor for databases.

Key points

  • Cloud Trace shows where request latency is spent across services.
  • Cloud Profiler shows which functions burn CPU or memory in production.
  • Error Reporting groups exceptions so you see the top failures.
  • Query Insights and the index advisor point at slow database queries and missing indexes.

Source: https://docs.cloud.google.com/trace/docs/overview

Practice questions

Single choice

Which tool measures how long a request spends in each service it passes through?

  1. Cloud Profiler
  2. Cloud Trace
  3. Error Reporting
  4. Cloud NAT
Show answer

Answer: B. Cloud Trace

Cloud Trace is the distributed tracing system built for request latency. Profiler looks inside one process at CPU and memory, not across services.

Select all that apply

Which tools help you diagnose problems inside your running application?

  1. Cloud Trace
  2. Cloud NAT
  3. Cloud Profiler
  4. Error Reporting
  5. Cloud Armor
Show answer

Answer: A. Cloud Trace, C. Cloud Profiler, D. Error Reporting

Trace, Profiler and Error Reporting each expose application behavior. Cloud NAT and Cloud Armor are network and security services and do not diagnose app code.

True or false

True or false: Cloud Trace keeps spans in its default bucket for 365 days.

Show answer

Answer: False

False. Spans in the _Trace bucket are kept for 30 days, so export or investigate promptly if you need older data.

Practice 8 questions on this topic in Bitzu

Short lessons, instant explanations and spaced review of what you get wrong. Free on iOS and Android.

Download on the App StoreGet it on Google Play

More Associate Cloud Engineer topics in this module