Collabora + Nextcloud on Kubernetes with Hikube

Introduction: A Sovereign Alternative to Google Workspace and Microsoft 365

Nextcloud has established itself as the reference solution for self-hosted collaboration. Beyond simple file storage, Nextcloud offers a complete suite of services: calendar, contacts, task management, and real-time communication. When combined with Collabora Online, you gain a full office suite with collaborative document editing capabilities.

For organizations concerned with data sovereignty and GDPR compliance, this combination represents a robust alternative to Google Workspace or Microsoft 365. And when you deploy it on Kubernetes via Hikube, you gain the scalability, resilience, and flexibility that only container orchestration can provide.

This guide walks you through deploying this architecture on Hikube, Hidora's managed Kubernetes platform.

Why Kubernetes for Nextcloud + Collabora?

Three key reasons motivate a Kubernetes deployment over a simple virtual machine instance.

Elastic Scalability

With Kubernetes, you scale resources according to demand. Nextcloud and Collabora pods replicate horizontally. During traffic spikes, new pods start automatically. Once the peak passes, they scale down. You pay only for consumed resources.

Resilience and High Availability

Kubernetes ensures your services remain available even when nodes fail. Pods are automatically restarted. Updates occur without service interruption through rolling update strategies. Your critical collaboration data experiences no downtime.

Simplified Update Management

Through Helm charts, Nextcloud and Collabora updates become predictable and testable. You precisely control deployed versions and can rollback to a previous version in seconds if needed.

Architecture Deployed on Hikube

Here are the components of our Kubernetes architecture:

Frontend Tier

• Ingress Controller: Manages external HTTP/HTTPS access with automated TLS certificates
• Nextcloud Pods: Multiple replicas for load balancing
• Collabora Online Pods: Scalable document editing service

Application Tier

• Nextcloud Service: ClusterIP exposing Nextcloud replicas
• Collabora Service: ClusterIP for intra-cluster communication
• ConfigMaps: Configuration storage for Nextcloud
• Secrets: Sensitive data (credentials, API keys)

Data Tier

• PostgreSQL StatefulSet: Relational database with persistence
• Redis Cache: Improves session and caching performance
• Persistent Volumes (PV): Storage for Nextcloud files
• Persistent Volume Claims (PVC): Storage requests for data

Hidora provides managed Persistent Volumes across its distributed storage infrastructure, ensuring durability and resilience.

Deployment via Helm Charts

Helm is Kubernetes's package manager. Helm charts encapsulate all necessary configuration.

Installing Prerequisites

Start by initializing your Hikube context:

kubectl config use-context hikube-cluster
kubectl create namespace nextcloud

Adding Helm Repositories

helm repo add nextcloud https://nextcloud.github.io/helm/
helm repo add bitnami https://charts.bitnami.com/bitnami
helm repo update

Configuring Helm Values

Create a values-nextcloud.yaml file:

replicaCount: 3

image:
  repository: nextcloud
  tag: "28.0"

nextcloud:
  host: nextcloud.your-domain.ch
  username: admin
  password: change-me
  
persistence:
  enabled: true
  storageClass: "hikube-fast"
  size: 100Gi

redis:
  enabled: true
  auth:
    enabled: true
    password: redis-secure-password

postgresql:
  enabled: true
  postgresqlPassword: postgres-secure-password
  primary:
    persistence:
      enabled: true
      size: 50Gi

ingress:
  enabled: true
  className: nginx
  annotations:
    cert-manager.io/cluster-issuer: "letsencrypt-prod"
  hosts:
    - host: nextcloud.your-domain.ch
      paths:
        - path: /
          pathType: Prefix
  tls:
    - secretName: nextcloud-tls
      hosts:
        - nextcloud.your-domain.ch

Deploying Nextcloud

helm install nextcloud nextcloud/nextcloud \
  -n nextcloud \
  -f values-nextcloud.yaml

Configuring Collabora Online

For Collabora Online, use the official chart:

helm repo add collabora https://collaboraonline.github.io/helm-collabora-online
helm repo update

Create values-collabora.yaml:

replicaCount: 2

image:
  tag: "24.04"

persistence:
  enabled: true
  size: 20Gi

resources:
  limits:
    cpu: 2000m
    memory: 2Gi
  requests:
    cpu: 1000m
    memory: 1Gi

ingress:
  enabled: true
  className: nginx
  annotations:
    cert-manager.io/cluster-issuer: "letsencrypt-prod"
  hosts:
    - host: collabora.your-domain.ch
      paths:
        - path: /
          pathType: Prefix
  tls:
    - secretName: collabora-tls
      hosts:
        - collabora.your-domain.ch

Deploy:

helm install collabora collabora/collabora-online \
  -n nextcloud \
  -f values-collabora.yaml

Connecting Nextcloud to Collabora

After both services are running, you need to configure Nextcloud to use Collabora as its office document editor. Install the "Nextcloud Office" app (formerly Collabora Online) from the Nextcloud app store through the admin interface.

Navigate to Administration Settings > Nextcloud Office and set the Collabora Online server URL to https://collabora.your-domain.ch. Enable the "Use your own server" option. Nextcloud will test the connection and confirm that Collabora is reachable.

For environments where both services run in the same Kubernetes cluster, you can also use the internal ClusterIP service address (http://collabora-collabora-online.nextcloud.svc.cluster.local:9980) to avoid routing traffic through the public ingress. This reduces latency and keeps document editing traffic entirely within the cluster network.

To verify the integration, create a new document directly from the Nextcloud Files interface. The Collabora editor should open inline, allowing real-time collaborative editing. Test with multiple users simultaneously to confirm that changes sync correctly and that WebSocket connections are stable through your ingress configuration.

Persistent Volumes on Hikube

Hikube offers multiple storage classes optimized for different use cases.

For Nextcloud Files (frequent read/write, high availability)

Use hikube-fast, based on replicated SSD with node failure protection.

For PostgreSQL (intensive sequential access)

Use hikube-general, offering good balance between performance and cost with replication.

Configuring PVCs

Hikube automatically creates Persistent Volumes when launching Helm charts. Monitor their usage:

kubectl get pvc -n nextcloud
kubectl describe pvc nextcloud-nextcloud -n nextcloud

To dynamically increase size without interruption:

kubectl patch pvc nextcloud-nextcloud -n nextcloud \
  -p '{"spec":{"resources":{"requests":{"storage":"200Gi"}}}}'

Ingress Configuration with TLS

Hikube includes an Nginx Ingress Controller and cert-manager for automatic Let's Encrypt certificates.

To enable HTTPS:

kubectl apply -f - <<EOF
apiVersion: cert-manager.io/v1
kind: ClusterIssuer
metadata:
  name: letsencrypt-prod
spec:
  acme:
    server: https://acme-v02.api.letsencrypt.org/directory
    email: admin@your-domain.ch
    privateKeySecretRef:
      name: letsencrypt-prod
    solvers:
    - http01:
        ingress:
          class: nginx
EOF

Verify certificates:

kubectl get certificate -n nextcloud

Scaling and Performance Optimization

Horizontal Scaling

Adjust Nextcloud replicas based on your load:

kubectl scale deployment nextcloud --replicas=5 -n nextcloud

Or via Helm:

helm upgrade nextcloud nextcloud/nextcloud \
  -n nextcloud \
  --set replicaCount=5 \
  -f values-nextcloud.yaml

Autoscaling with HPA

For automatic scaling based on CPU load:

apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: nextcloud-hpa
  namespace: nextcloud
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: nextcloud
  minReplicas: 3
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 70

Resource Optimization

For optimal collaboration, Hikube lets you adjust resource limits:

kubectl set resources deployment nextcloud \
  --limits=cpu=2,memory=2Gi \
  --requests=cpu=500m,memory=512Mi \
  -n nextcloud

Collabora Online, resource-intensive for document editing, benefits from minimum 2 CPU and 2 GB RAM per pod.

Monitoring Your Deployment

A production Nextcloud + Collabora deployment needs active monitoring. Without visibility into pod health and resource consumption, performance issues go undetected until users complain.

Deploy Prometheus and Grafana in your cluster to collect metrics from both Nextcloud and Collabora pods. Nextcloud exposes health endpoints at /status.php and /ocs/v2.php/apps/serverinfo/api/v1/info that return JSON data on active users, storage usage and server load. Configure Prometheus to scrape these endpoints at regular intervals.

For Collabora, monitor the number of active document editing sessions, memory consumption per session and WebSocket connection counts. Collabora tends to consume significant memory when many users edit large spreadsheets simultaneously. Set alerts for when memory usage exceeds 80% of pod limits, so you can scale Collabora replicas before users experience degradation.

On Hikube, you can also leverage the built-in monitoring stack to track persistent volume usage. Running out of storage on your Nextcloud PVC is a preventable outage. Configure alerts when volume usage crosses 75%, giving your team time to expand the PVC before it fills completely.

Deployment Best Practices for Production Readiness

Before going live with Nextcloud and Collabora in production, a few operational practices will save you significant troubleshooting time. First, always pin your Helm chart versions and container image tags explicitly. Using latest or unpinned chart versions means that a routine helm upgrade could introduce breaking changes without warning. Lock your versions in your values.yaml files and only upgrade deliberately after testing in a staging namespace.

Second, configure Pod Disruption Budgets (PDBs) for both Nextcloud and Collabora deployments. A PDB ensures that Kubernetes never voluntarily evicts all your replicas simultaneously during node maintenance or cluster upgrades. For a three-replica Nextcloud deployment, setting minAvailable: 2 guarantees that at least two pods remain running at all times, preventing user-visible interruption during rolling updates or node drains.

Third, implement proper liveness and readiness probes. Nextcloud's /status.php endpoint is well-suited as a readiness probe target. If a Nextcloud pod fails to respond to the readiness check, Kubernetes automatically removes it from the service's load balancer, preventing users from being routed to an unhealthy instance. For Collabora, use the /hosting/capabilities endpoint as your readiness probe to confirm the document editing service is fully initialized before accepting traffic.

Compliance and Data Sovereignty

GDPR and Swiss Hosting

Your data remains hosted in Switzerland on Hidora's infrastructure. No third party accesses it. You maintain full control over access and user permissions.

Backups and Recovery

Hikube integrates Persistent Volume snapshots. Configure a backup policy:

kubectl apply -f - <<EOF
apiVersion: snapshot.storage.k8s.io/v1
kind: VolumeSnapshot
metadata:
  name: nextcloud-backup
  namespace: nextcloud
spec:
  volumeSnapshotClassName: hikube-snapshoter
  source:
    persistentVolumeClaimName: nextcloud-nextcloud
EOF

Audit and Logging

Enable Hikube audit logs to trace all access and modifications. Integrate with your SIEM system for continuous compliance.

Next Steps and Support

Deploying Nextcloud with Collabora Online on Kubernetes requires expertise and attention to detail. Certificate configuration, performance optimization, persistent volume management, monitoring, and automatic scaling form a non-trivial technical set.

Hikube simplifies Kubernetes infrastructure, but application configuration remains your responsibility. This is where our consulting services and our managed services deliver immense value.

Hidora can help you:

• Design optimal architecture for your collaboration needs
• Configure and deploy Nextcloud + Collabora on Hikube
• Establish high availability and backup strategies
• Optimize performance for your user load
• Manage secure updates
• Ensure GDPR and sovereignty compliance

Contact our team for a free consultation on your sovereign collaboration project.