Exam Professional Cloud Developer topic 1 question 2 discussion

C The task does not indicate that we should get rid of the old software. The pain point is slowness for time critical problems only. Thus we would use Stackdriver for the time critical alerts and still utilize the old platform for further analysis/storing of logs or whatever its business case is.

we can directly omit B and D. B. Just installing Stackdriver wont do any good. we need notification system D. Testing is irrelevant here. based on A&C, i would go with C

Addresses Immediate Problem: * Solves the critical notification speed issue * Maintains existing monitoring infrastructure * Provides a hybrid approach that leverages the best of both systems Benefits of this approach: * Uses Stackdriver's real-time alerting capabilities * Preserves investment in existing monitoring platform * Minimizes disruption to current operations * Allows for gradual transition if needed

the most balanced and least disruptive approach would likely be option C, "Use Stackdriver to capture and alert on logs, then ship them to your existing platform". This method allows for a seamless integration of Stackdriver's real-time alerting capabilities into your existing monitoring workflow without the need for a complete overhaul of your system.

B is correct answer

Option C. Notifications from on-prem monitoring system are too slow & applications are in GCP now => Stackdriver alerts on logs. There's no mentioning whether the apps have been migrated to GCE, GKE, App Engine or Cloud Run, so "Compute Engine instances" come from an assumption.

I would go with C.

Option B , Install stack driver agents on the compute instances is the correct answer . using stackdriver and shipping it to existing platform will have some delay

C you have problems with notifications. C option allows you to use stackdriver to send alerts immediately and straight away after sends all this data to your on-prem monitoring platform

B is the correct answer. https://cloud.google.com/monitoring/agent/monitoring/installation

Think twice. You have working an expensive monitoring system i.e Splunk and you have the problem with unacceptable delay time between incident and notification. You need to fix this problem, not doing a revolution (changing monitoring system). You can leverage GCP Monitoring with alerting system which is out-of-the-box with no huge effort, because if you want or not logs are in cloud logging. Simply implement alerts and push logs to Splunk. Simples.

A is correct

It's the rightest answer. C cannot work without Agent and there's not sense to send log to old monitoring

It's A 'cos you cannot use Stackdriver if you don't install Stackdriver agent on your compute engine.

Community choice is C

I think it should be A than C Apps have been migrated and why would you invest on C to send data back to existing system instead fix old system using direct connect or something

The answer is C. The point is notification problem with low performance, not monitoring.