Merge pull request #552 from ykulazhenkov/nvidia-ipam-leader-elect

Update manifests to deploy nvidia-ipam in HA mode

Author: adrianchiris

README.md

@@ -217,7 +217,7 @@ spec:
   nvIpam:
     image: nvidia-k8s-ipam
     repository: ghcr.io/mellanox
-    version: v0.0.2
+    version: v0.0.3
     config: '{
       "pools":  {
         "my-pool": {"subnet": "192.168.0.0/24", "perNodeBlockSize": 100, "gateway": "192.168.0.1"}

deployment/network-operator/README.md

@@ -555,11 +555,11 @@ optionally deployed components:
 #### NVIDIA IPAM Plugin
 
 | Name                      | Type   | Default            | description                                                                            |
-| ------------------------- | ------ | ------------------ | -------------------------------------------------------------------------------------- |
+| ------------------------- | ------ |--------------------| -------------------------------------------------------------------------------------- |
 | `nvIpam.deploy`           | bool   | `false`            | Deploy NVIDIA IPAM Plugin                                                              |
 | `nvIpam.image`            | string | `nvidia-k8s-ipam`  | NVIDIA IPAM Plugin image name                                                          |
 | `nvIpam.repository`       | string | `ghcr.io/mellanox` | NVIDIA IPAM Plugin image repository                                                    |
-| `nvIpam.version`          | string | `v0.0.2`           | NVIDIA IPAM Plugin image version                                                       |
+| `nvIpam.version`          | string | `v0.0.3`           | NVIDIA IPAM Plugin image version                                                       |
 | `nvIpam.imagePullSecrets` | list   | `[]`               | An optional list of references to secrets to use for pulling any of the Plugin image   |
 | `nvIpam.config`           | string | `''`               | Network pool configuration as described in https://github.com/Mellanox/nvidia-k8s-ipam |
 

deployment/network-operator/values.yaml

@@ -231,7 +231,7 @@ nvIpam:
   deploy: false
   image: nvidia-k8s-ipam
   repository: ghcr.io/mellanox
-  version: v0.0.2
+  version: v0.0.3
   # imagePullSecrets: []
   # network pool configuration as described in https://github.com/Mellanox/nvidia-k8s-ipam
   config: ''

example/crs/mellanox.com_v1alpha1_nicclusterpolicy_cr-nvidia-ipam.yaml

@@ -61,7 +61,7 @@ spec:
   nvIpam:
     image: nvidia-k8s-ipam
     repository: ghcr.io/mellanox
-    version: v0.0.2
+    version: v0.0.3
     config: '{
       "pools":  {
         "my-pool": {"subnet": "192.168.0.0/24", "perNodeBlockSize": 100, "gateway": "192.168.0.1"}

hack/release.yaml

@@ -52,4 +52,4 @@ IpamPlugin:
 nvIpam:
   image: nvidia-k8s-ipam
   repository: ghcr.io/mellanox
-  version: v0.0.2
+  version: v0.0.3

manifests/state-nv-ipam-cni/020-cluster-role.yaml

@@ -47,4 +47,23 @@ rules:
       - get
       - list
       - watch
+  - apiGroups:
+      - coordination.k8s.io
+    resources:
+      - leases
+    verbs:
+      - get
+      - list
+      - watch
+      - create
+      - update
+      - patch
+      - delete
+  - apiGroups:
+      - ""
+    resources:
+      - events
+    verbs:
+      - create
+      - patch
 ---

manifests/state-nv-ipam-cni/040-nv-ipam-controller.yaml

@@ -27,7 +27,7 @@ metadata:
 spec:
   strategy:
     type: RollingUpdate
-  replicas: 1
+  replicas: 2
   selector:
     matchLabels:
       name: nv-ipam-controller
@@ -41,6 +41,18 @@ spec:
       priorityClassName: system-cluster-critical
       serviceAccountName: nv-ipam-controller
       affinity:
+        podAntiAffinity:
+          # use preferred(instead of required) here to support deployment to the singe node cluster
+          preferredDuringSchedulingIgnoredDuringExecution:
+          - weight: 100
+            podAffinityTerm:
+              labelSelector:
+                matchExpressions:
+                  - key: name
+                    operator: In
+                    values:
+                      - nv-ipam-controller
+              topologyKey: "kubernetes.io/hostname"
         nodeAffinity:
           preferredDuringSchedulingIgnoredDuringExecution:
           - weight: 1
@@ -81,6 +93,8 @@ spec:
           args:
             - --config-name=nvidia-k8s-ipam-config
             - --config-namespace=$(POD_NAMESPACE)
+            - --leader-elect=true
+            - --leader-elect-namespace=$(POD_NAMESPACE)
           env:
             - name: POD_NAMESPACE
               valueFrom: