Deploying to Kubernetes

Kubernetes is a powerful open-source platform for orchestrating containerized applications, automating their deployment, scaling, and management. In this guide, we'll leverage Helm, Kubernetes' package manager, to streamline our deployment process.

Technical Overview

Deploying Univer backend services on Kubernetes involves several key components:

• Kubernetes Cluster: The environment where your app resides.
• Helm: The deployment and management tool for apps on Kubernetes.
• Database and Messaging: PostgreSQL and RabbitMQ services.

Installing Helm

Head over to the Helm installation guide (opens in a new tab) to get started. Once installed, confirm Helm's presence by executing:

helm version

Deployment Steps

Step 1: Kubernetes Cluster Setup

Make sure you have an active Kubernetes cluster. For cloud providers like Google Kubernetes Engine (GKE) or Amazon Elastic Kubernetes Service (EKS), adhere to their specific instructions for cluster creation.

Step 2: Deploying Univer services

1. Use Helm to deploy Univer services:

helm install univer --create-namespace \
    --set global.istioNamespace=univer \
    oci://univer-acr-registry.cn-shenzhen.cr.aliyuncs.com/helm-charts/univer-stack
 
kubectl rollout restart deployment/collaboration-server -n univer
kubectl rollout restart deployment/universer -n univer

2. To check if your deployment is up and running:

kubectl get pods

Step 3: Deployment Verification

# For quick testing, you might edit your local hosts file to map the domain.
# Default domain: univer.example.com
 
Visit: http://univer.example.com

Troubleshooting

If your service isn't launching, verify the status of your Kubernetes Pods and consult the logs for clues.

If you're still having issues, feel free to contact us for assistance.