Use appropriate logging levels

This commit is contained in:
Hakan Bostan 2023-05-31 15:03:11 +00:00
parent 7b8e0e62a7
commit c255aaa2a9
1 changed files with 1 additions and 2 deletions

View File

@ -448,7 +448,7 @@ func (o *ScaleUpOrchestrator) filterValidScaleUpNodeGroups(
}
if autoscalingOptions != nil && autoscalingOptions.AtomicScaleUp {
if o.autoscalingContext.MaxNodesTotal != 0 && currentNodeCount+nodeGroup.MaxSize() > o.autoscalingContext.MaxNodesTotal {
klog.Errorf("Skipping node group %s - atomic scale-up exceetds cluster node count limit", nodeGroup.Id())
klog.V(4).Infof("Skipping node group %s - atomic scale-up exceetds cluster node count limit", nodeGroup.Id())
skippedNodeGroups[nodeGroup.Id()] = NewSkippedReasons("atomic scale-up exceeds cluster node count limit")
continue
}
@ -505,7 +505,6 @@ func (o *ScaleUpOrchestrator) ComputeExpansionOption(
estimator := estimator.NewBinpackingNodeEstimator(o.autoscalingContext.PredicateChecker, o.autoscalingContext.ClusterSnapshot, limiter, estimator.NewDecreasingPodOrderer())
option.NodeCount, option.Pods = estimator.Estimate(pods, nodeInfo, nodeGroup)
if option.NodeCount != nodeGroup.MaxSize() {
klog.Warningf("Estimator NodeCount: %d setting to %d", option.NodeCount, nodeGroup.MaxSize())
option.NodeCount = nodeGroup.MaxSize()
}
return option