To achieve high availability (HA) for OVN, deploy the ovn-central DPUService with the enableHA option enabled.
0. Prerequisites
Requires at least three schedulable control-plane nodes for quorum.
Use $HELM_REGISTRY_REPO_URL and TAG to specify the Helm chart registry and your desired version respectively.
## The repository URL for the NVIDIA Helm chart registry.
## Usually this is the NVIDIA Helm NGC registry. For development purposes, this can be set to a different repository.
export HELM_REGISTRY_REPO_URL=https://helm.ngc.nvidia.com/nvidia/doca
## The DPF TAG is the version of the OVN components which will be deployed in this guide.
export TAG=v25.10.1
1. Deploy ovn-central DPUService
cat examples/dpuservice-ovn-central.yaml | envsubst | kubectl apply -f -
This will deploy ovn-central DPUService with enableHA set to true, it will create 3 ovn-central pods on the control-plane nodes.
YAML
---
apiVersion: svc.dpu.nvidia.com/v1alpha1
kind: DPUService
metadata:
name: ovn-central
namespace: dpf-operator-system
spec:
deployInCluster: true
helmChart:
source:
repoURL: $HELM_REGISTRY_REPO_URL
version: $TAG
chart: ovn-chart
values:
exposedPorts:
ports:
ovnnb: true
ovnsb: true
management:
ovnCentral:
enabled: true
enableHA: true
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: "node-role.kubernetes.io/master"
operator: Exists
- matchExpressions:
- key: "node-role.kubernetes.io/control-plane"
operator: Exists
tolerations:
- key: node-role.kubernetes.io/master
operator: Exists
effect: NoSchedule
- key: node-role.kubernetes.io/control-plane
operator: Exists
effect: NoSchedule
Last updated: