Asking for help? Comment out what you need so we can get more information to help you!
Kubernetes version: 1.19.5
Cloud being used: (put bare-metal if not on a public cloud) bare-metal
Installation method: n/a
Host OS: centos
CNI and version: calico
CRI and version: docker
In a k8s cluster, in one of node, if it has 80 cpu and 74 are used for pod, like this:
kubectl describe node xxxx-0|grep "Capacity\|Allocatable" -A 1 Capacity: cpu: 80 -- Allocatable: cpu: 74
If the pod in this node has requested 64 cpu, like this:
kubectl describe node xxxx|grep "Allocated resources" -A 4 Allocated resources: (Total limits may be over 100 percent, i.e., overcommitted.) Resource Requests Limits -------- -------- ------ cpu 64356m (86%) 83140m (112%)
And if I run some other task in this node, how many cpu can be used for this taks? In my test, if I run a stress tool which will cost cpu, it only can be run in 6 CPU. (Capacity 80 - Allocatable 74)? For the assigned 64 cpu for pod, even the pod is idle(doesn’t really use it), can my task in node(stress tool) use these 64 cpu?