A global enterprise runs streaming workloads across multiple Google Cloud projects containing Google Kubernetes Engine (GKE) clusters and Compute Engine instances. The central Site Reliability Engineering (SRE) and security teams require a centralized observability pipeline. Specifically, high-severity security audit logs must be retained long-term in BigQuery for regulatory compliance, while operational application error logs must be routed in real time to an external Security Information and Event Management (SIEM) system via Cloud Pub/Sub. To optimize logging costs, ensure minimal log ingestion overhead, and enforce proper access security without dropping critical events, which TWO configuration steps should the cloud architect implement? Select TWO.
- Create an organization-level aggregated Log Router sink specifying BigQuery as the destination, filtered to include specific audit and error log types, and configure a table retention policy.Cevap
- Configure a Log Router sink with a real-time error log filter targeting a Cloud Pub/Sub topic, and grant the sink's writer service account the Pub/Sub Publisher role on that topic.Cevap
- CApply a global Cloud Logging exclusion filter with the query `severity >= ERROR` across all projects to suppress high-volume error log ingestion into Cloud Logging storage buckets.
- DGrant the primitive Owner role to the Log Router writer service accounts in each project to guarantee write permission to cross-project BigQuery datasets and Pub/Sub topics.
- ERely exclusively on Identity and Access Management (IAM) role bindings on the log sinks to prevent unauthorized users from exfiltrating log data to external Cloud Storage buckets.