Pod Inspector - CoreWeave Docs

The Pod Inspector dashboard provides a combined view of common Pod metrics, Kubernetes events, and Pod logs, filtered to a specific Pod or container in any namespace. Use it to investigate the runtime behavior of a single Pod or container when troubleshooting issues such as restarts, resource pressure, or scheduling problems. To view the dashboard, go to the Pod Inspector dashboard.

For instructions about accessing CoreWeave Grafana dashboards, see Access and use CoreWeave Grafana dashboards.

The following table describes each panel available in the dashboard.

Panel title	Description
Node History	Shows the history of the Node’s status or events.
Type	Displays the type of the monitored resource.
Node	Shows Node information.
Network Throughput	Shows the network throughput of the Node.
Container Waiting History	Shows the history of containers in a `Waiting` state.
CPU Percentage Usage	Shows the CPU utilization as a percentage of the total available.
CGNAT Traffic Classification	Shows the Carrier-Grade NAT (CGNAT) traffic.
CPU Usage (Cores)	Displays CPU usage as the number of cores in use.
Memory Percentage Usage	Shows the memory utilization as a percentage of the total available.
CPU Throttle Rate	Shows the rate at which the CPU is being throttled.
Memory Usage (RSS)	Displays the resident set size (RSS) memory usage.
OOM History	Shows the history of out-of-memory (OOM) events.
InfiniBand Bandwidth on Node (Active)	Shows the active bandwidth on the Node’s InfiniBand network interface.
Container Restarts	Shows the number of container restarts over time.
GPU Usage	Shows the utilization of the Node’s GPU.
GPU Memory Usage	Shows the memory usage of the Node’s GPU.
Logs	Displays log entries from the Node or its containers.
Kubernetes Events	Displays the Kubernetes events related to the monitored resource.

Last modified on June 10, 2026