Questions to Determine

• What are the current node conditions (MemoryPressure, DiskPressure, PIDPressure)?
• Which pods are consuming the most memory on the node?
• Do the affected pods have memory requests and limits defined?
• What are the kubelet eviction thresholds configured on this node?
• Are evicted pods being rescheduled back to the same overloaded node?
• Is there a single pod causing the memory spike, or is it cumulative overcommitment?
• What is the total memory requested vs. allocatable on the node?
• Are there LimitRange or ResourceQuota policies in the affected namespaces?
• Is the cluster autoscaler enabled and are there pending pods?