karmada/pkg/controllers/binding/common.go

360 lines
14 KiB
Go

package binding
import (
"context"
"reflect"
"sort"
apiextensionsv1 "k8s.io/apiextensions-apiserver/pkg/apis/apiextensions/v1"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/apimachinery/pkg/apis/meta/v1/unstructured"
"k8s.io/klog/v2"
"sigs.k8s.io/controller-runtime/pkg/builder"
"sigs.k8s.io/controller-runtime/pkg/client"
"sigs.k8s.io/controller-runtime/pkg/event"
"sigs.k8s.io/controller-runtime/pkg/predicate"
clusterv1alpha1 "github.com/karmada-io/karmada/pkg/apis/cluster/v1alpha1"
policyv1alpha1 "github.com/karmada-io/karmada/pkg/apis/policy/v1alpha1"
workv1alpha1 "github.com/karmada-io/karmada/pkg/apis/work/v1alpha1"
workv1alpha2 "github.com/karmada-io/karmada/pkg/apis/work/v1alpha2"
"github.com/karmada-io/karmada/pkg/util"
"github.com/karmada-io/karmada/pkg/util/helper"
"github.com/karmada-io/karmada/pkg/util/names"
"github.com/karmada-io/karmada/pkg/util/overridemanager"
)
var workPredicateFn = builder.WithPredicates(predicate.Funcs{
CreateFunc: func(e event.CreateEvent) bool {
return false
},
UpdateFunc: func(e event.UpdateEvent) bool {
var statusesOld, statusesNew []workv1alpha1.ManifestStatus
switch oldWork := e.ObjectOld.(type) {
case *workv1alpha1.Work:
statusesOld = oldWork.Status.ManifestStatuses
default:
return false
}
switch newWork := e.ObjectNew.(type) {
case *workv1alpha1.Work:
statusesNew = newWork.Status.ManifestStatuses
default:
return false
}
return !reflect.DeepEqual(statusesOld, statusesNew)
},
DeleteFunc: func(event.DeleteEvent) bool {
return true
},
GenericFunc: func(event.GenericEvent) bool {
return false
},
})
// ensureWork ensure Work to be created or updated.
// TODO(Garrybest): clean up the code to fix cyclomatic complexity
//nolint:gocyclo
func ensureWork(c client.Client, workload *unstructured.Unstructured, overrideManager overridemanager.OverrideManager, binding metav1.Object, scope apiextensionsv1.ResourceScope) error {
var targetClusters []workv1alpha2.TargetCluster
switch scope {
case apiextensionsv1.NamespaceScoped:
bindingObj := binding.(*workv1alpha2.ResourceBinding)
targetClusters = bindingObj.Spec.Clusters
case apiextensionsv1.ClusterScoped:
bindingObj := binding.(*workv1alpha2.ClusterResourceBinding)
targetClusters = bindingObj.Spec.Clusters
}
hasScheduledReplica, referenceRSP, desireReplicaInfos, err := getRSPAndReplicaInfos(c, workload, targetClusters)
if err != nil {
return err
}
var jobCompletions []workv1alpha2.TargetCluster
var jobHasCompletions = false
if workload.GetKind() == util.JobKind {
completions, found, err := unstructured.NestedInt64(workload.Object, util.SpecField, util.CompletionsField)
if err != nil {
return err
}
if found {
jobCompletions = util.DivideReplicasByTargetCluster(targetClusters, int32(completions))
jobHasCompletions = true
}
}
for i := range targetClusters {
targetCluster := targetClusters[i]
clonedWorkload := workload.DeepCopy()
cops, ops, err := overrideManager.ApplyOverridePolicies(clonedWorkload, targetCluster.Name)
if err != nil {
klog.Errorf("Failed to apply overrides for %s/%s/%s, err is: %v", clonedWorkload.GetKind(), clonedWorkload.GetNamespace(), clonedWorkload.GetName(), err)
return err
}
workNamespace, err := names.GenerateExecutionSpaceName(targetCluster.Name)
if err != nil {
klog.Errorf("Failed to ensure Work for cluster: %s. Error: %v.", targetCluster.Name, err)
return err
}
workLabel := mergeLabel(clonedWorkload, workNamespace, binding, scope)
if clonedWorkload.GetKind() == util.DeploymentKind && (referenceRSP != nil || hasScheduledReplica) {
err = applyReplicaSchedulingPolicy(clonedWorkload, desireReplicaInfos[targetCluster.Name], util.ReplicasField)
if err != nil {
klog.Errorf("failed to apply ReplicaSchedulingPolicy for %s/%s/%s in cluster %s, err is: %v",
clonedWorkload.GetKind(), clonedWorkload.GetNamespace(), clonedWorkload.GetName(), targetCluster.Name, err)
return err
}
}
if clonedWorkload.GetKind() == util.JobKind && hasScheduledReplica {
err = applyReplicaSchedulingPolicy(clonedWorkload, desireReplicaInfos[targetCluster.Name], util.ParallelismField)
if err != nil {
klog.Errorf("failed to apply Parallelism for %s/%s/%s in cluster %s, err is: %v",
clonedWorkload.GetKind(), clonedWorkload.GetNamespace(), clonedWorkload.GetName(), targetCluster.Name, err)
return err
}
// For a work queue Job that usually leaves .spec.completions unset, in that case, we skip setting this field.
// Refer to: https://kubernetes.io/docs/concepts/workloads/controllers/job/#parallel-jobs.
if jobHasCompletions {
err = applyReplicaSchedulingPolicy(clonedWorkload, int64(jobCompletions[i].Replicas), util.CompletionsField)
if err != nil {
klog.Errorf("failed to apply Completions for %s/%s/%s in cluster %s, err is: %v",
clonedWorkload.GetKind(), clonedWorkload.GetNamespace(), clonedWorkload.GetName(), targetCluster.Name, err)
return err
}
}
}
annotations := mergeAnnotations(clonedWorkload, binding, scope)
annotations, err = recordAppliedOverrides(cops, ops, annotations)
if err != nil {
klog.Errorf("failed to record appliedOverrides, Error: %v", err)
return err
}
workMeta := metav1.ObjectMeta{
Name: names.GenerateWorkName(clonedWorkload.GetKind(), clonedWorkload.GetName(), clonedWorkload.GetNamespace()),
Namespace: workNamespace,
Finalizers: []string{util.ExecutionControllerFinalizer},
Labels: workLabel,
Annotations: annotations,
}
if err = helper.CreateOrUpdateWork(c, workMeta, clonedWorkload); err != nil {
return err
}
}
return nil
}
func getRSPAndReplicaInfos(c client.Client, workload *unstructured.Unstructured, targetClusters []workv1alpha2.TargetCluster) (bool, *policyv1alpha1.ReplicaSchedulingPolicy, map[string]int64, error) {
if helper.HasScheduledReplica(targetClusters) {
return true, nil, transScheduleResultToMap(targetClusters), nil
}
referenceRSP, desireReplicaInfos, err := calculateReplicasIfNeeded(c, workload, helper.GetBindingClusterNames(targetClusters))
if err != nil {
klog.Errorf("Failed to get ReplicaSchedulingPolicy for %s/%s/%s, err is: %v", workload.GetKind(), workload.GetNamespace(), workload.GetName(), err)
return false, nil, nil, err
}
return false, referenceRSP, desireReplicaInfos, nil
}
func applyReplicaSchedulingPolicy(workload *unstructured.Unstructured, desireReplica int64, field string) error {
_, ok, err := unstructured.NestedInt64(workload.Object, util.SpecField, field)
if err != nil {
return err
}
if ok {
err := unstructured.SetNestedField(workload.Object, desireReplica, util.SpecField, field)
if err != nil {
return err
}
}
return nil
}
func mergeLabel(workload *unstructured.Unstructured, workNamespace string, binding metav1.Object, scope apiextensionsv1.ResourceScope) map[string]string {
var workLabel = make(map[string]string)
util.MergeLabel(workload, workv1alpha2.WorkNamespaceLabel, workNamespace)
util.MergeLabel(workload, workv1alpha2.WorkNameLabel, names.GenerateWorkName(workload.GetKind(), workload.GetName(), workload.GetNamespace()))
if scope == apiextensionsv1.NamespaceScoped {
util.MergeLabel(workload, workv1alpha2.ResourceBindingReferenceKey, names.GenerateBindingReferenceKey(binding.GetNamespace(), binding.GetName()))
workLabel[workv1alpha2.ResourceBindingReferenceKey] = names.GenerateBindingReferenceKey(binding.GetNamespace(), binding.GetName())
} else {
util.MergeLabel(workload, workv1alpha2.ClusterResourceBindingReferenceKey, names.GenerateBindingReferenceKey("", binding.GetName()))
workLabel[workv1alpha2.ClusterResourceBindingReferenceKey] = names.GenerateBindingReferenceKey("", binding.GetName())
}
return workLabel
}
func mergeAnnotations(workload *unstructured.Unstructured, binding metav1.Object, scope apiextensionsv1.ResourceScope) map[string]string {
annotations := make(map[string]string)
if scope == apiextensionsv1.NamespaceScoped {
util.MergeAnnotation(workload, workv1alpha2.ResourceBindingNamespaceLabel, binding.GetNamespace())
util.MergeAnnotation(workload, workv1alpha2.ResourceBindingNameLabel, binding.GetName())
annotations[workv1alpha2.ResourceBindingNamespaceLabel] = binding.GetNamespace()
annotations[workv1alpha2.ResourceBindingNameLabel] = binding.GetName()
} else {
util.MergeAnnotation(workload, workv1alpha2.ClusterResourceBindingLabel, binding.GetName())
annotations[workv1alpha2.ClusterResourceBindingLabel] = binding.GetName()
}
return annotations
}
func recordAppliedOverrides(cops *overridemanager.AppliedOverrides, ops *overridemanager.AppliedOverrides,
annotations map[string]string) (map[string]string, error) {
if annotations == nil {
annotations = make(map[string]string)
}
if cops != nil {
appliedBytes, err := cops.MarshalJSON()
if err != nil {
return nil, err
}
if appliedBytes != nil {
annotations[util.AppliedClusterOverrides] = string(appliedBytes)
}
}
if ops != nil {
appliedBytes, err := ops.MarshalJSON()
if err != nil {
return nil, err
}
if appliedBytes != nil {
annotations[util.AppliedOverrides] = string(appliedBytes)
}
}
return annotations, nil
}
func transScheduleResultToMap(scheduleResult []workv1alpha2.TargetCluster) map[string]int64 {
var desireReplicaInfos = make(map[string]int64, len(scheduleResult))
for _, clusterInfo := range scheduleResult {
desireReplicaInfos[clusterInfo.Name] = int64(clusterInfo.Replicas)
}
return desireReplicaInfos
}
func calculateReplicasIfNeeded(c client.Client, workload *unstructured.Unstructured, clusterNames []string) (*policyv1alpha1.ReplicaSchedulingPolicy, map[string]int64, error) {
var err error
var referenceRSP *policyv1alpha1.ReplicaSchedulingPolicy
var desireReplicaInfos = make(map[string]int64)
if workload.GetKind() == util.DeploymentKind {
referenceRSP, err = matchReplicaSchedulingPolicy(c, workload)
if err != nil {
return nil, nil, err
}
if referenceRSP != nil {
desireReplicaInfos, err = calculateReplicas(c, referenceRSP, clusterNames)
if err != nil {
klog.Errorf("Failed to get desire replicas for %s/%s/%s, err is: %v", workload.GetKind(), workload.GetNamespace(), workload.GetName(), err)
return nil, nil, err
}
klog.V(4).Infof("DesireReplicaInfos with replica scheduling policies(%s/%s) is %v", referenceRSP.Namespace, referenceRSP.Name, desireReplicaInfos)
}
}
return referenceRSP, desireReplicaInfos, nil
}
func matchReplicaSchedulingPolicy(c client.Client, workload *unstructured.Unstructured) (*policyv1alpha1.ReplicaSchedulingPolicy, error) {
// get all namespace-scoped replica scheduling policies
policyList := &policyv1alpha1.ReplicaSchedulingPolicyList{}
if err := c.List(context.TODO(), policyList, &client.ListOptions{Namespace: workload.GetNamespace()}); err != nil {
klog.Errorf("Failed to list replica scheduling policies from namespace: %s, error: %v", workload.GetNamespace(), err)
return nil, err
}
if len(policyList.Items) == 0 {
return nil, nil
}
matchedPolicies := getMatchedReplicaSchedulingPolicy(policyList.Items, workload)
if len(matchedPolicies) == 0 {
klog.V(2).Infof("No replica scheduling policy for resource: %s/%s", workload.GetNamespace(), workload.GetName())
return nil, nil
}
return &matchedPolicies[0], nil
}
func getMatchedReplicaSchedulingPolicy(policies []policyv1alpha1.ReplicaSchedulingPolicy, resource *unstructured.Unstructured) []policyv1alpha1.ReplicaSchedulingPolicy {
// select policy in which at least one resource selector matches target resource.
resourceMatches := make([]policyv1alpha1.ReplicaSchedulingPolicy, 0)
for _, policy := range policies {
if util.ResourceMatchSelectors(resource, policy.Spec.ResourceSelectors...) {
resourceMatches = append(resourceMatches, policy)
}
}
// Sort by policy names.
sort.Slice(resourceMatches, func(i, j int) bool {
return resourceMatches[i].Name < resourceMatches[j].Name
})
return resourceMatches
}
func calculateReplicas(c client.Client, policy *policyv1alpha1.ReplicaSchedulingPolicy, clusterNames []string) (map[string]int64, error) {
weightSum := int64(0)
matchClusters := make(map[string]int64)
desireReplicaInfos := make(map[string]int64)
// found out clusters matched the given ReplicaSchedulingPolicy
for _, clusterName := range clusterNames {
clusterObj := &clusterv1alpha1.Cluster{}
if err := c.Get(context.TODO(), client.ObjectKey{Name: clusterName}, clusterObj); err != nil {
klog.Errorf("Failed to get member cluster: %s, error: %v", clusterName, err)
return nil, err
}
for _, staticWeightRule := range policy.Spec.Preferences.StaticWeightList {
if util.ClusterMatches(clusterObj, staticWeightRule.TargetCluster) {
weightSum += staticWeightRule.Weight
matchClusters[clusterName] = staticWeightRule.Weight
break
}
}
}
if weightSum == 0 {
return desireReplicaInfos, nil
}
allocatedReplicas := int32(0)
for clusterName, weight := range matchClusters {
desireReplicaInfos[clusterName] = weight * int64(policy.Spec.TotalReplicas) / weightSum
allocatedReplicas += int32(desireReplicaInfos[clusterName])
}
if remainReplicas := policy.Spec.TotalReplicas - allocatedReplicas; remainReplicas > 0 && len(matchClusters) > 0 {
sortedClusters := helper.SortClusterByWeight(matchClusters)
for i := 0; remainReplicas > 0; i++ {
desireReplicaInfos[sortedClusters[i].ClusterName]++
remainReplicas--
if i == len(desireReplicaInfos) {
i = 0
}
}
}
for _, clusterName := range clusterNames {
if _, exist := matchClusters[clusterName]; !exist {
desireReplicaInfos[clusterName] = 0
}
}
return desireReplicaInfos, nil
}