You have an application that runs in Google Kubernetes Engine (GKE). Over the last 2 weeks, customers have reported that a specific part of the application returns errors very frequently. You currently have no logging or monitoring solution enabled on your GKE cluster. You want to diagnose the problem, but you have not been able to replicate the issue. You want to cause minimal disruption to the application. What should you do?

The correct option allows you to enable Cloud Operations for GKE (formerly Stackdriver) on your existing cluster with minimal disruption, as it's an in-place update. Once enabled, Cloud Operations automatically collects logs and metrics from your GKE cluster, including application logs. You can then use the GKE Monitoring dashboard, which integrates with Cloud Operations, to review logs from the affected pods and diagnose the intermittent issue without needing to replicate it. The incorrect options involving creating a new cluster and migrating pods would cause significant disruption and are unnecessary. While Prometheus can be a valuable monitoring tool, Cloud Operations for GKE provides comprehensive logging and monitoring capabilities out-of-the-box for this scenario, making it the more straightforward and less disruptive first step. Setting an alert is useful for future proactive monitoring but doesn't help with diagnosing a past or intermittent issue without underlying logging data.