This test was conducted by going into a pod by kubectl exec and running a script makes an array of extreme size, this causes pods memory usage to increase by a huge amount.
if a pod consumes excessive memory resulting in exhaustion of memory for that pod. It goes into a state “ContainerStatusUnknown” and becomes unresponsive
In case of a Deployment another pod for the deployment is started.
memory resources exhaustion by a pod does not result in node getting killed or becoming unresponsive. Only the pod is becomes unresponsive with status “ContainerStatusUnknown”
/ For Nodes
This test was conducted by connecting to a node with ssh and running a script makes an array of extreme size, this causes node’s memory to get full.
In this scenario the node becomes unresponsive for 15-20 mins and is not reachable by any means, through ssh etc. GKE control plane detects this unresponsiveness and takes measures against it
Node becomes available after the above mentioned unresponsiveness period
Want to print your doc? This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
