eG Help

Typically, in a Kubernetes environment, when Pods run containers, they use the CPU and memory resources on the node to which they are scheduled. By default, a Pod in Kubernetes will run with no limits on CPU and memory. This means that a single Pod can end up hogging the resources of the node! To isolate such resource hogging Pods, it is necessary to periodically monitor the resource utilization of the Pods. The CorOSPodPerfTest helps administrators in this regard!

For each pod launched using the target CRIO container engine, this test reports the resource utilization and the total number of iNodes available in the pod. Using this test, administrators can identify resource-intensive pods, analyze the root-cause of such high resource utilization and rectify the same at the earliest.

Outputs of the test : One set of results for each container available in the CRIO Engine being monitored

Measurement	Description	Measurement Unit	Interpretation
Pod_container	Indicates the total number of containers available in this pod.	Number	The detailed diagnosis of this measure lists the Pod Namespace, Container name and the Container ID associated with each Pod.
Pod_cpu	Indicates the percentage of CPU utilized by this pod.	Percent	Comparing the value of this measure across the pods will enable you to accurately identify the pod that is utilizing maximum CPU resources. The detailed diagnosis of this measure lists the Pod Namespace, Container name, Container ID and the CPU Usage of each Pod.
Pod_memory	Indicates the amount of memory utilized by this pod.	MB	A high value for this measure indicates that the memory resource of the pod is depleting alarmingly. Compare the value of this measure across containers to figure out the pod that is using maximum amount of memory resources. The detailed diagnosis of this measure lists the Pod Namespace, Container name, Container ID and the memory used by each Pod.
Pod_disk	Indicates the amount of disk space utilized by this pod.	MB	Compare the value of this measure across containers to figure out the pod that is using maximum amount of disk space. The detailed diagnosis of this measure lists the Pod Namespace, Container name, Container ID and the amount of disk space utilized by each Pod.
Pod_inode	Indicates the total number of iNodes in this pod.	Number	The detailed diagnosis of this measure lists the Pod Namespace, Container name, Container ID and the total number of iNodes in each Pod.