devops-engineer video for as the SRE engineer for a microservice deployed to a Google Kubernetes Engine (GKE) Autopilot cluster, you oversee an online store's
As the SRE engineer for a microservice deployed to a Google Kubernetes Engine (GKE) Autopilot cluster, you oversee an online store's functionality. The system publishes order messages to Pub/Sub, and a microservice processes these messages to update stock information in the warehousing system. Due to a surge in orders during a sales event, stock updates are lagging, leading to orders being accepted for out-of-stock products. Reviewing the microservice metrics against typical levels reveals: Current Microservice Metrics: A. Average CPU usage across all Pods: 30% of Pod limit (typical state: 20% of Pod limit) B. Average memory usage across all Pods: 10% of Pod limit (typical state: 10% of Pod limit) C. Pub/Sub subscription:A. Average oldest unacknowledged message age: 8074 milliseconds (typical state: 347 milliseconds) D. Average undelivered messages: 14705 messages (typical state: 5 messages) E. Average acknowledgment latency: 354 milliseconds (typical state: 312 milliseconds) Your objective is to ensure accurate inventory representation in the warehouse system during order placement while minimizing customer impact. What course of action should you take to resolve the current situation?