Search <book_title>...

Important Update: Cohesity Products Documentation

All Cohesity product documentation are now managed via the Cohesity Docs Portal: https://docs.cohesity.com/HomePage/Content/home.htm. Some documentation available here may not reflect the latest information or may no longer be accessible.

InfoScale™ for Kubernetes 9.1.0 - Linux

Last Published: 2026-01-07

Product(s): InfoScale & Storage Foundation (9.1)

Platform: Linux

Enabling kubelet inhibitor with systemd

To enable kubelet's graceful shutdown inhibitor on worker nodes, users must apply two configurations in the following order:

systemd change with machine config
Extends InhibitDelayMaxSec so systemd honors kubelet's shutdown inhibitor for the full grace period.
Kubelet Configuration shutdown grace periods
Defines how long kubelet will wait for regular and critical pods during shutdown.

Both configurations trigger MCP updates and node reboots, so they should be applied deliberately.

Note:

This must be performed only once in the lifetime of the cluster.

Perform these steps

Ensure the worker MCP contains the required label so MachineConfigs target the correct nodes:
oc label mcp worker machineconfiguration.openshift.io/role=worker
Verify labels
oc get mcp worker --show-labels

Apply the systemd override MachineConfig

This config sets InhibitDelayMaxSec=900 (15 minutes), allowing kubelet to delay shutdown properly.

Copy this butane file, change the version to OpenShift major version that is present on your cluster.

:: worker.butane

variant: openshift
version: 4.19.0
metadata:
  name: 99-worker-sysd-conf-override
  labels:
    machineconfiguration.openshift.io/role: worker
storage:
  files:
    - path: /etc/systemd/logind.conf.d/50-max-delay.conf # << systemd drop in
      mode: 0644
      overwrite: true
      contents:
        inline: |
          # The Machine Config Operator manages this file.
          [Login]
          InhibitDelayMaxSec=900 # << 15m

Convert the butane file to MachineConfig:
butane worker.bu -o 99-worker-sysd-conf-override.yaml
Apply MachineConfig to the cluster:
oc apply -f 99-worker-sysd-conf-override.yaml
Check MCP rollout with below commands:
oc get mcp
oc get nodes
Nodes will reboot as the new systemd config is applied.
Check if status of MachineConfigPool shows as:
```
Updated=True

Degraded=False
```
Apply the kubelet configuration for shutdown grace periods.
This config sets:
```
shutdownGracePeriod: 15m

shutdownGracePeriodCriticalPods: 5m
```

Copy below YAML and apply:

:: mc-kubelet-config.yaml
---
 apiVersion: machineconfiguration.openshift.io/v1
 kind: KubeletConfig
 metadata:
   name: custom-kubelet-config
 spec:
   machineConfigPoolSelector:
     matchLabels:
       machineconfiguration.openshift.io/role: worker
   kubeletConfig:
     shutdownGracePeriod: "15m"
     shutdownGracePeriodCriticalPods: "5m"

oc apply -f mc-kubelet-config.yaml

Verify custom kubeletconfig.
# oc describe kubeletconfigs.machineconfiguration.openshift.io custom-kubelet-config
Wait for MCP rollout to complete and nodes to update.
oc get mcp
oc get nodes
Check if status of MachineConfigPool shows as:
```
Updated=True

Degraded=False
```
After nodes return to Ready, login to worker nodes and run:
systemd-inhibit --list
Expected output:
```
A kubelet entry with WHAT=shutdown and MODE=delay.
```