For AI agents: The documentation index is at https://docs.digitalocean.com/llms.txt. Markdown versions of pages use the same URL with index.html.md in place of the HTML page (for example, append index.html.md to the directory path instead of opening the HTML document).

Limits

• After you enable the metrics agent, you cannot revert to basic Droplet graphs. If you want different visualizations, export metrics to an external tool such as Grafana and build a custom dashboard.

• You cannot export metrics directly from the do-agent service. The agent collects system metrics and sends them to DigitalOcean backend services, where they appear in the DigitalOcean Control Panel and API.

• You cannot set a separate alert email address for an individual account. To share alerts, use a team account email address or configure Slack notifications for a shared channel.

• The metrics agent collects system-level data only. It does not provide application performance monitoring (APM) and does not track application errors, website traffic, or application bottlenecks. To monitor application-level metrics, use a separate APM tool such as New Relic.

GPU Observability Limits

• The power throttling GPU metric is currently available only for AMD GPUs. NVIDIA support is planned but not yet available. To use AMD GPUs, contact support to request access.

• AMD GPUs are available by request only. Contact support to request access before you provision AMD GPU resources.

• You cannot enable or disable GPU Observability after Droplet creation. It is available only when you select Improved Metrics and monitoring (Free) during Droplet creation with an AI/ML Ready image.

• You currently cannot create alert policies for GPU metrics.

• GPU temperature limits follow either NVIDIA or AMD specifications. Sustained temperatures near the rated maximum may indicate thermal constraints.

• You cannot view power utilization as a percentage. Power usage is available in watts only.

• Metric availability varies by GPU model. We provide an extensive set of metrics for NVIDIA A30, H100, and H200 GPUs and for AMD MI300 and MI325 GPUs. Profiler metrics such as occupancy and tensor core utilization are not available for NVIDIA L40S, RTX 6000 Ada, and RTX 4000 Ada GPUs.

• You may not see GPU Occupancy (%) or Tensor Utilization (%) on RTX-class GPUs. Some models, including RTX 4000 Ada, RTX 6000 Ada, and L40S, do not expose DCGM profiling metrics because of NVIDIA exporter limitations.

• You must use an AI/ML Ready image to enable GPU Observability automatically. For Basic Images, you can enable GPU metrics by manually installing the exporter, binding it to 127.0.0.1, reconfiguring do-agent to scrape it, and restarting do-agent.