When using reclaim core pod, the pod cannot be scheduled
Lan-ce-lot opened this issue · comments
What happened?
reclaimed-normal-pod.yaml
apiVersion: v1
kind: Pod
metadata:
annotations:
"katalyst.kubewharf.io/qos_level": reclaimed_cores
name: reclaimed-normal-pod
namespace: default
spec:
containers:
- name: stress
image: joedval/stress:latest
command:
- stress
- -c
- "1"
imagePullPolicy: IfNotPresent
resources:
requests:
"resource.katalyst.kubewharf.io/reclaimed_millicpu": "2k"
"resource.katalyst.kubewharf.io/reclaimed_memory": 1Gi
limits:
"resource.katalyst.kubewharf.io/reclaimed_millicpu": "2k"
"resource.katalyst.kubewharf.io/reclaimed_memory": 1Gi
schedulerName: katalyst-scheduler
What did you expect to happen?
Reclaim core pod completes scheduling.
How can we reproduce it (as minimally and precisely as possible)?
Use the default policy scheduler-policy-spread.yaml of the example folder.
Software version
$ <software> version
# paste output here