Kubernetes installation

eRPC can be deployed on Kubernetes using the following manifests. These examples provide a basic setup that you can customize based on your needs.

Configuration

First, create a ConfigMap and a Secret for your eRPC configuration:

apiVersion: v1
kind: ConfigMap
metadata:
  name: erpc-config
data:
  erpc.yaml: |
    logLevel: debug
    projects:
     - id: main
       upstreams:
       - endpoint: alchemy://${ALCHEMY_API_KEY}
       - endpoint: blastapi://${BLASTAPI_API_KEY}
       - endpoint: https://mynode-chain-1.svc.cluster.local
---
apiVersion: v1
kind: Secret
metadata:
  name: erpc-secrets
type: Opaque
stringData:
  ALCHEMY_API_KEY: your-alchemy-key-here
  BLASTAPI_API_KEY: your-blastapi-key-here

Deployment

Deploy eRPC with the following configuration:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: erpc
  labels:
    app: erpc
spec:
  replicas: 1
  selector:
    matchLabels:
      app: erpc
  template:
    metadata:
      labels:
        app: erpc
    spec:
      containers:
      - name: erpc
        image: ghcr.io/erpc/erpc:latest
        resources:
          # CPU limits removed as they can cause throttling issues
          requests:
            memory: "256Mi"
          limits:
            memory: "2Gi"
        env:
        - name: GOGC
          value: "40"
        - name: GOMEMLIMIT
          value: "1900MiB"
        envFrom:
        - secretRef:
            name: erpc-secrets
        ports:
        - containerPort: 4000
          name: http
        - containerPort: 4001
          name: metrics
        volumeMounts:
        - name: config
          mountPath: /erpc.yaml
          subPath: erpc.yaml
        startupProbe:
          httpGet:
            path: /healthcheck
            port: 4000
          initialDelaySeconds: 10
          periodSeconds: 10
          timeoutSeconds: 5
          failureThreshold: 6
        readinessProbe:
          httpGet:
            path: /healthcheck
            port: 4000
          initialDelaySeconds: 10
          periodSeconds: 5
          timeoutSeconds: 5
          failureThreshold: 2
          successThreshold: 1
        livenessProbe:
          tcpSocket:
            port: 4000
          initialDelaySeconds: 30
          periodSeconds: 10
          timeoutSeconds: 1
          failureThreshold: 3
          successThreshold: 1
      volumes:
      - name: config
        configMap:
          name: erpc-config
      # This must be same or greater than server.maxTimeout in erpc.yaml
      terminationGracePeriodSeconds: 180

Service

Expose eRPC using a Service:

apiVersion: v1
kind: Service
metadata:
  name: erpc
  labels:
    app: erpc
spec:
  ports:
  - port: 4000
    name: http
    targetPort: 4000
  - port: 4001
    name: metrics
    targetPort: 4001
  selector:
    app: erpc

Horizontal Pod Autoscaling

Configure automatic scaling based on CPU and memory usage:

apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: erpc
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: erpc
  minReplicas: 1
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 80
  - type: Resource
    resource:
      name: memory
      target:
        type: Utilization
        averageUtilization: 80

Installation

Apply the manifests using kubectl:

# Apply the manifests
kubectl apply -f erpc-configmap.yaml
kubectl apply -f erpc-secret.yaml
kubectl apply -f erpc-deployment.yaml
kubectl apply -f erpc-service.yaml
kubectl apply -f erpc-hpa.yaml
 
# Verify the deployment
kubectl get pods
kubectl get services
kubectl get hpa

The eRPC service will be available within your cluster at erpc:4000 for HTTP traffic and erpc:4001 for metrics.

Railway Cloud