Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Added diagnostic logs for potential empty nodes #372

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

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

Merged
aaronfern merged 2 commits into from
Oct 21, 2025
Merged

Added diagnostic logs for potential empty nodes #372

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e64ab92
Added diagnostic logs for potential empty nodes
aaronfern Oct 1, 2025
4d4bc50
Address review comments
aaronfern Oct 21, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
34 changes: 34 additions & 0 deletions cluster-autoscaler/simulator/utilization/info.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,7 @@ package utilization

import (
"fmt"
"strings"
"time"

"k8s.io/autoscaler/cluster-autoscaler/cloudprovider"
Expand Down Expand Up @@ -106,28 +107,61 @@ func CalculateUtilizationOfResource(nodeInfo *framework.NodeInfo, resourceName a
// the same with the Mirror pod.
podsRequest := resource.MustParse("0")
daemonSetAndMirrorPodsUtilization := resource.MustParse("0")
// FORK-CHANGE: following variables added for diagnostic logs
// diagBuff - buffer to collect diagnostic logs
// maxDiagnosticPods - maximum number of pods to log details for
// podCount - total number of pods processed
// detailedPodCount - number of pods with details logged
var diagBuff strings.Builder
const maxDiagnosticPods = 20
podCount := 0
detailedPodCount := 0
for _, podInfo := range nodeInfo.Pods() {
podCount++
podRequests := podutils.PodRequests(podInfo.Pod)
resourceValue := podRequests[resourceName]

// factor daemonset pods out of the utilization calculations
if skipDaemonSetPods && podutils.IsDaemonSetPod(podInfo.Pod) {
daemonSetAndMirrorPodsUtilization.Add(resourceValue)
if detailedPodCount < maxDiagnosticPods {
fmt.Fprintf(&diagBuff, "pod %s/%s is a daemonset pod. Adding resource %q, value %v to daemonSetAndMirrorPodsUtilization\n", podInfo.Pod.Namespace, podInfo.Pod.Name, resourceName, resourceValue)
detailedPodCount++
}
continue
}

// factor mirror pods out of the utilization calculations
if skipMirrorPods && podutils.IsMirrorPod(podInfo.Pod) {
daemonSetAndMirrorPodsUtilization.Add(resourceValue)
if detailedPodCount < maxDiagnosticPods {
fmt.Fprintf(&diagBuff, "pod %s/%s is a mirror pod. Adding resource %q, value %v to daemonSetAndMirrorPodsUtilization\n", podInfo.Pod.Namespace, podInfo.Pod.Name, resourceName, resourceValue)
detailedPodCount++
}
continue
}

// ignore Pods that should be terminated
if drain.IsPodLongTerminating(podInfo.Pod, currentTime) {
if detailedPodCount < maxDiagnosticPods {
fmt.Fprintf(&diagBuff, "pod %s/%s is terminating.\n", podInfo.Pod.Namespace, podInfo.Pod.Name)
detailedPodCount++
}
continue
}

podsRequest.Add(resourceValue)
if detailedPodCount < maxDiagnosticPods {
fmt.Fprintf(&diagBuff, "pod %s/%s has requests %v for resource %q\n", podInfo.Pod.Namespace, podInfo.Pod.Name, resourceValue, resourceName)
detailedPodCount++
}
}

if podCount > maxDiagnosticPods {
fmt.Fprintf(&diagBuff, "... plus %d more pods (details omitted)\n", podCount-maxDiagnosticPods)
}
if podsRequest.MilliValue() == 0 {
klog.Warningf("Diagnostic 0 pod requests for node %s:\n%s\n", nodeInfo.Node().Name, diagBuff.String())
}

return float64(podsRequest.MilliValue()) / float64(nodeAllocatable.MilliValue()-daemonSetAndMirrorPodsUtilization.MilliValue()), nil
Expand Down
Loading