github/kata-containers
1
0
Fork 2
mirror of https://github.com/kata-containers/kata-containers.git synced 2026-07-01 22:50:54 +00:00
Code Issues Projects Releases Wiki Activity

helm-chart: add priorityClassName to prevent kata-deploy eviction

Browse Source 
kata-deploy is a per-node infrastructure DaemonSet; if it gets evicted
under node memory/CPU pressure the node loses its Kata runtime until
the pod is rescheduled. Default to system-node-critical so the kubelet
evicts lower-priority workloads first.

The value is configurable via `priorityClassName` in values.yaml.

Fixes: #13068

Signed-off-by: Fabiano Fidêncio <ffidencio@nvidia.com>
This commit is contained in:
Fabiano Fidêncio
2026-05-18 15:12:56 +02:00
committed by Fabiano Fidêncio
parent f6fce19e01
commit 5d40ba66ff
2 changed files with 20 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
tools/packaging/kata-deploy/helm-chart/kata-deploy/templates/kata-deploy.yaml
View File

@@ -61,6 +61,9 @@ spec:
tolerations:
{{- toYaml . | nindent 8 }}
{{- end }}
{{- with .Values.priorityClassName }}
priorityClassName: {{ . | quote }}
{{- end }}
{{- if index .Values "node-feature-discovery" "enabled" }}
# When node-feature-discovery is managed by this chart (enabled: true), enforce virtualization requirements.
# Kata Containers requires hardware virtualization support to function.

17
tools/packaging/kata-deploy/helm-chart/kata-deploy/values.yaml
View File

@@ -38,6 +38,23 @@ containerd:
nodeSelector: {}
tolerations: []
# Priority class name for the kata-deploy DaemonSet pods.
#
# kata-deploy is an infrastructure DaemonSet that installs Kata runtime
# artifacts on every node. If it gets evicted under node pressure, the
# node loses its Kata runtime until the pod is rescheduled. Setting a
# high-priority class prevents the kubelet from evicting kata-deploy
# before lower-priority workloads.
#
# Kubernetes ships "system-node-critical" and "system-cluster-critical"
# PriorityClasses by default. You can also create your own.
#
# Examples:
# priorityClassName: system-node-critical
# priorityClassName: system-cluster-critical
# priorityClassName: my-custom-priority-class
priorityClassName: "system-node-critical"
# Update strategy for the kata-deploy DaemonSet
# Default is RollingUpdate with maxUnavailable: 1
# Examples: