Kubernetes Deployment - Nurosentrix Documentation

Overview

Nurosentrix supports deploying Odoo instances on Kubernetes clusters, providing better scalability and resource management for large deployments.

Prerequisites

A running Kubernetes cluster (1.24+)
kubectl configured to access the cluster
Helm 3.x installed
Storage class for persistent volumes
Ingress controller (nginx-ingress recommended)

Connecting Your Cluster

Step 1: Add Cluster in Nurosentrix

Go to Settings → Kubernetes
Click "Add Cluster"
Enter a name for your cluster
Upload your kubeconfig file or paste the configuration
Click "Connect"

Step 2: Verify Connection

Nurosentrix will verify it can connect to your cluster and list the available namespaces and resources.

The Kubernetes options appear in the instance wizard when a cluster is connected under Settings → Kubernetes.

Helm Chart

Nurosentrix deploys Odoo using a Helm chart. The chart includes:

Odoo Deployment
PostgreSQL StatefulSet
Persistent Volume Claims
Services
Ingress configuration
Horizontal Pod Autoscaler

Chart Values

# values.yaml
odoo:
  version: "17.0"
  edition: community
  replicas: 2

  resources:
    requests:
      cpu: 500m
      memory: 1Gi
    limits:
      cpu: 2000m
      memory: 4Gi

postgresql:
  enabled: true
  primary:
    persistence:
      size: 20Gi

persistence:
  filestore:
    size: 50Gi
    storageClass: standard

ingress:
  enabled: true
  hostname: odoo.example.com
  tls: true

autoscaling:
  enabled: true
  minReplicas: 2
  maxReplicas: 10
  targetCPU: 70

Deploying an Instance

Using Nurosentrix Dashboard

Go to Instances → New Instance
Select your Kubernetes cluster
Choose a namespace (or create new)
Configure Odoo settings
Set resource limits
Configure ingress/domain
Click "Deploy"

Using Helm Directly

# Add Nurosentrix Helm repository
helm repo add nurosentrix https://charts.nurosentrix.com
helm repo update

# Install Odoo
helm install my-odoo nurosentrix/odoo \
  --namespace odoo \
  --create-namespace \
  -f values.yaml

Kubernetes-backed instances show the target cluster and namespace in the Instances list.

Architecture

A typical Kubernetes deployment includes:

┌─────────────────────────────────────────┐
│              Ingress                     │
│         (nginx-ingress)                  │
└─────────────────┬───────────────────────┘
                  │
┌─────────────────▼───────────────────────┐
│              Service                     │
│           (ClusterIP)                    │
└─────────────────┬───────────────────────┘
                  │
┌─────────────────▼───────────────────────┐
│            Deployment                    │
│         (Odoo Pods x N)                  │
│    ┌──────┐ ┌──────┐ ┌──────┐          │
│    │ Pod  │ │ Pod  │ │ Pod  │          │
│    └──────┘ └──────┘ └──────┘          │
└─────────────────┬───────────────────────┘
                  │
┌─────────────────▼───────────────────────┐
│           StatefulSet                    │
│          (PostgreSQL)                    │
│              ┌──────┐                    │
│              │ Pod  │                    │
│              └──────┘                    │
└─────────────────────────────────────────┘

Scaling

Horizontal Pod Autoscaler

Enable HPA to automatically scale Odoo pods based on CPU usage:

autoscaling:
  enabled: true
  minReplicas: 2
  maxReplicas: 10
  targetCPU: 70
  targetMemory: 80

Manual Scaling

kubectl scale deployment my-odoo --replicas=5 -n odoo

Persistent Storage

Configure storage for filestore and database:

Component	Recommended Size	Storage Class
PostgreSQL	20-100 GB	Fast SSD (gp3, premium-ssd)
Filestore	50-500 GB	Standard or NFS

High Availability

For production deployments:

Run at least 2 Odoo replicas
Use PostgreSQL with replication
Configure pod anti-affinity
Use shared filestore (NFS or cloud storage)

affinity:
  podAntiAffinity:
    preferredDuringSchedulingIgnoredDuringExecution:
      - weight: 100
        podAffinityTerm:
          labelSelector:
            matchLabels:
              app: odoo
          topologyKey: kubernetes.io/hostname

Monitoring

If you have Prometheus installed, Nurosentrix can export metrics:

metrics:
  enabled: true
  serviceMonitor:
    enabled: true
    namespace: monitoring

Troubleshooting

Pods Not Starting

# Check pod status
kubectl get pods -n odoo

# View pod events
kubectl describe pod <pod-name> -n odoo

# Check logs
kubectl logs <pod-name> -n odoo

Common Issues

ImagePullBackOff - Check image name and pull secrets
CrashLoopBackOff - Check application logs
Pending - Insufficient resources or storage issues