Add LangKit Helm Chart

[AnchorArray]

Output of helm template --namespace langkit langkit .

---
# Source: langkit/templates/namespace.yaml
apiVersion: v1
kind: Namespace
metadata:
  name: langkit
---
# Source: langkit/templates/service.yaml
apiVersion: v1
kind: Service
metadata:
  name: langkit
  namespace: langkit
  labels:
    helm.sh/chart: langkit-0.1.0
    app.kubernetes.io/name: langkit
    app.kubernetes.io/instance: langkit
    app.kubernetes.io/version: "1.16.0"
    app.kubernetes.io/managed-by: Helm
spec:
  type: ClusterIP
  ports:
    - port: 8000
      targetPort: http
      protocol: TCP
      name: http
  selector:
    app.kubernetes.io/name: langkit
    app.kubernetes.io/instance: langkit
---
# Source: langkit/templates/deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: langkit
  namespace: langkit
  labels:
    helm.sh/chart: langkit-0.1.0
    app.kubernetes.io/name: langkit
    app.kubernetes.io/instance: langkit
    app.kubernetes.io/version: "1.16.0"
    app.kubernetes.io/managed-by: Helm
spec:
  replicas: 3
  selector:
    matchLabels:
      app.kubernetes.io/name: langkit
      app.kubernetes.io/instance: langkit
  template:
    metadata:
      labels:
        helm.sh/chart: langkit-0.1.0
        app.kubernetes.io/name: langkit
        app.kubernetes.io/instance: langkit
        app.kubernetes.io/version: "1.16.0"
        app.kubernetes.io/managed-by: Helm
    spec:
      serviceAccountName: default
      securityContext:
        {}
      containers:
        - name: langkit
          securityContext:
            {}
          image: "whylabs/whylogs:py-llm-latest"
          imagePullPolicy: IfNotPresent
          ports:
            - name: http
              containerPort: 8000
              protocol: TCP
          livenessProbe:
            null
          readinessProbe:
            null
          resources:
            limits:
              cpu: "8"
              memory: 16Gi
            requests:
              cpu: "4"
              memory: 8Gi
          envFrom:
            - secretRef:
                name: whylabs-api-key
            - secretRef:
                name: langkit-api-secret
      affinity:
        podAntiAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:
          - podAffinityTerm:
              labelSelector:
                matchExpressions:
                - key: app.kubernetes.io/name
                  operator: In
                  values:
                  - langkit
              topologyKey: kubernetes.io/hostname
            weight: 100
      tolerations:
        - effect: NoSchedule
          key: kubernetes.azure.com/scalesetpriority
          operator: Equal
          value: spot

[AnchorArray]

This Helm chart uses preferredDuringSchedulingIgnoredDuringExecution which will attempt to put all of the replicas on different nodes. Using preferred over required allows the Chart to be deployed, without issues, if there is a mismatch in the number of replicas/nodes or if a node is not able to support any additional workloads

[andyndang]

LGTM