We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
What happened: graceful-eviction does not work.
depoly.yaml
apiVersion: apps/v1 kind: Deployment metadata: name: nginx labels: app: nginx spec: replicas: 4 selector: matchLabels: app: nginx template: metadata: labels: app: nginx spec: containers: - image: docker.m.daocloud.io/nginx name: nginx resources: limits: cpu: 250m memory: 512Mi requests: cpu: 250m memory: 512Mi
pp.yaml
apiVersion: policy.karmada.io/v1alpha1 kind: PropagationPolicy metadata: name: nginx-propagation spec: resourceSelectors: - apiVersion: apps/v1 kind: Deployment name: nginx placement: clusterAffinities: - affinityName: default clusterNames: - member1 - member2 replicaScheduling: replicaSchedulingType: Divided replicaDivisionPreference: Weighted weightPreference: dynamicWeight: AvailableReplicas
Add taints.
./_output/bin/darwin/arm64/karmadactl taint clusters member1 key1=value1:NoExecute
RB.yaml
spec: clusters: - name: member2 replicas: 4 gracefulEvictionTasks: - fromCluster: member1 producer: TaintManager reason: TaintUntolerated replicas: 2 placement: clusterAffinities: - affinityName: default clusterNames: - member1 - member2 clusterTolerations: - effect: NoExecute key: cluster.karmada.io/not-ready operator: Exists tolerationSeconds: 30 - effect: NoExecute key: cluster.karmada.io/unreachable operator: Exists tolerationSeconds: 30 replicaScheduling: replicaDivisionPreference: Weighted replicaSchedulingType: Divided weightPreference: dynamicWeight: AvailableReplicas replicaRequirements: resourceRequest: cpu: 250m memory: 512Mi replicas: 4 resource: apiVersion: apps/v1 kind: Deployment name: nginx namespace: default resourceVersion: "6942" uid: 78aff1f4-ddf4-40f9-9880-5d5f8ea23f33 schedulerName: default-scheduler status: aggregatedStatus: - applied: true clusterName: member1 health: Healthy status: availableReplicas: 2 readyReplicas: 2 replicas: 2 updatedReplicas: 2 - applied: true clusterName: member2 health: Healthy status: availableReplicas: 4 readyReplicas: 4 replicas: 4 updatedReplicas: 4 conditions: - lastTransitionTime: "2023-04-26T03:20:45Z" message: Binding has been scheduled reason: BindingScheduled status: "True" type: Scheduled - lastTransitionTime: "2023-04-26T03:20:46Z" message: All works have been successfully applied reason: FullyAppliedSuccess status: "True" type: FullyApplied schedulerObservedGeneration: 3 schedulerObservingAffinityName: default
gracefulEvictionTasks: - fromCluster: member1 producer: TaintManager reason: TaintUntolerated replicas: 2
The gracefulEvictionTasks queue is never empty.
gracefulEvictionTasks
Restart karmada-controller-manager doesn't work either.
karmada-controller-manager
What you expected to happen:
graceful-eviction can work.
How to reproduce it (as minimally and precisely as possible):
Anything else we need to know?:
Environment:
kubectl-karmada version
karmadactl version
The text was updated successfully, but these errors were encountered:
Please @XiShanYongYe-Chang @jwcesign take a look if it is similar to #3456.
Sorry, something went wrong.
Hi, can you check if there is a log for this function?
karmada/pkg/controllers/gracefuleviction/rb_graceful_eviction_controller.go
Line 39 in 0f84dd7
If there is not, #3455 will solve your problem.
The root cause of this problem is that rb.status.SchedulerObservedGeneration does not equal rb.Generation.
https://github.com/karmada-io/karmada/blob/master/pkg/controllers/gracefuleviction/rb_graceful_eviction_controller.go#L95
...n does not equal rb.Generation.
Yes, @Poor12 is taking care of it.
Successfully merging a pull request may close this issue.
What happened:
graceful-eviction does not work.
depoly.yaml
pp.yaml
Add taints.
RB.yaml
The
gracefulEvictionTasks
queue is never empty.Restart
karmada-controller-manager
doesn't work either.What you expected to happen:
graceful-eviction can work.
How to reproduce it (as minimally and precisely as possible):
Anything else we need to know?:
Environment:
kubectl-karmada version
orkarmadactl version
):The text was updated successfully, but these errors were encountered: