A DevOps engineer is configuring observability for a fleet of Compute Engine instances running a specialized media processing application. The operations team requires real-time memory usage metrics and operating system log collection in Google Cloud Observability, as standard hypervisor metrics do not capture OS-level memory consumption. Which action should the engineer take to meet these telemetry requirements?
- Install and configure the unified Google Cloud Ops Agent on all Compute Engine instances.Answer
- BInstall the legacy Stackdriver Monitoring Agent and Stackdriver Logging Agent on all Compute Engine instances.
- CGrant the Compute Instance Admin (v1) IAM role to the default Compute Engine service account.
- DEnable the Cloud Logging and Cloud Monitoring APIs in the organization node rather than the individual host project.
Answer
Install and configure the unified Google Cloud Ops Agent on all Compute Engine instances.
The Google Cloud Ops Agent is the primary unified agent for Compute Engine virtual machines. It combines logging and metric collection into a single agent, capturing guest OS telemetry such as detailed memory usage, swap space, disk metrics, and system logs, which are inaccessible via standard hypervisor metrics.
Step-by-Step Solution
Key Concept
Compute Engine VM Observability and Google Cloud Ops Agent Deployment