Quick, which node is hogging all the CPU? This one-liner tells you everything at a glance.

1
2
3
4
5
6
7
8

kubectl get pod -A -o wide > /tmp/pods.a; \
kubectl top nodes | while read line; do \
  export NODE=`echo $line | awk '{print $1}'`; \
  export NOS=`cat /tmp/pods.a | grep $NODE | grep Running | wc -l`; \
  export CPU=`echo $line | awk '{print $3}'`; \
  export MEM=`echo $line | awk '{print $5}'`; \
  echo NODE: $NODE, NOS=$NOS, CPU=$CPU, MEM=$MEM; \
done

Output format:

1

NODE: node-name, NOS=12, CPU=45%, MEM=67%

Where:

  • NOS: Number of running pods on the node
  • CPU: CPU usage percentage
  • MEM: Memory usage percentage