An organization operates a fleet of Compute Engine virtual machines running an in-memory database application. To establish proactive capacity planning, the operations team needs to collect detailed internal system memory utilization and disk space metrics within Google Cloud Observability. After deploying the legacy Stackdriver Monitoring agent to Debian 12 instances, the team observes that internal memory metrics are missing from Cloud Monitoring. Following Google-recommended best practices for Compute Engine operational management, which action should the team take to collect these metrics?
- Uninstall the legacy agent and install the unified Google Cloud Ops Agent on the virtual machine instances.Cevap
- BGrant the Compute Engine default service account the Editor primitive role to permit custom telemetry ingestion.
- CSubmit a request to increase the Cloud Monitoring API time-series ingestion quota in the GCP Console.
- DConvert the virtual machine instances into Spot VMs to enable automated system-level memory metric logging.
Cevap
Uninstall the legacy agent and install the unified Google Cloud Ops Agent on the virtual machine instances.
The Google Cloud Ops Agent is Google's recommended solution for collecting guest OS telemetry, such as internal memory utilization and detailed disk metrics, from Compute Engine virtual machines. Replacing legacy Monitoring and Logging agents with the Ops Agent restores metric reporting into Google Cloud Observability.
Adım Adım Çözüm
Anahtar Kavram
Telemetry collection for Compute Engine instances using the unified Google Cloud Ops Agent