Skip to content

Deploy LM

Enterpriseonly

The LM service is only used to manage the NebulaGraph Enterprise license. If you are using the Community Edition of NebulaGraph, you do not need to deploy LM.

Before deploying NebulaGraph Enterprise 3.5.0 or later using Operator, you first need to deploy License Manager (LM) and configure the NebulaGraph Enterprise License in LM. LM is a standalone service used to manage the NebulaGraph license. LM checks the validity of the license when NebulaGraph Enterprise database starts. If the License is invalid, the database will not be able to start.

Deployment instructions

Operator does not currently support the deployment of LM. You need to deploy LM themselves.

As LM needs to store data and is a stateful service. You can deploy LM through the StatefulSet resource type or outside the Kubernetes cluster.

Deploy LM outside K8s cluster

For information on how to deploy LM on a machine outside the K8s cluster, see License Manager.

Caution

If LM is deployed outside the K8s cluster, make sure that the port of the LM service (default is 9119) can be accessed by all nodes in the K8s cluster.

Deploy LM in K8s using StatefulSet

Prerequisites

  • Prepare the LM image.
  • A StorageClass has been created to store LM data. For more information, see Storage Class.

Steps

  1. Create a namespace.

    # Create the nebula-license-manager namespace.
    kubectl create namespace nebula-license-manager
    
  2. Create a Secret for pulling the LM image from a private repository.

    kubectl -n nebula-license-manager create secret docker-registry <image-pull-secret> \
    --docker-server=DOCKER_REGISTRY_SERVER \
    --docker-username=DOCKER_USER \
    --docker-password=DOCKER_PASSWORD
    
    • <image-pull-secret>: Specify the name of the Secret.
    • DOCKER_REGISTRY_SERVER: Specify the address of the private repository server from which the image is pulled, for example, reg.example-inc.com.
    • DOCKER_USER: Image repository username.
    • DOCKER_PASSWORD: Image repository password.
  3. Create a StatefulSet resource configuration file for LM. Here is an example:

    apiVersion: v1
    kind: Service
    metadata:
      name: nebula-license-manager
      namespace: nebula-license-manager
      labels:
        app: nebula-license-manager
    spec:
      ports:
      - port: 9119
      selector:
        app: nebula-license-manager
    ---
    apiVersion: apps/v1
    kind: StatefulSet
    metadata:
      name: nebula-license-manager
      namespace: nebula-license-manager
    spec:
      replicas: 1
      selector:
        matchLabels:
          app: nebula-license-manager
      serviceName: nebula-license-manager
      template:
        metadata:
          labels:
            app: nebula-license-manager
        spec:
          containers:
          - name: nebula-license-manager    
            image: # Fill in the corresponding LM image address.
            ports:
            - containerPort: 9119
            volumeMounts:
            - name: data
              mountPath: /usr/local/nebula-license-manager/data
            livenessProbe:
              httpGet:
                path: /health
                port: 9119
            readinessProbe:
              httpGet:
                path: /health
                port: 9119
          # Used to specify one or more Secret names for pulling private images.      
          imagePullSecrets:
            - name: image-pull-secret # The name of the Secret.
      volumeClaimTemplates:
      - metadata:
          name: data
        spec:
          accessModes:
          - ReadWriteOnce
          storageClassName: "local-path" # Storage Class name.
          resources:
            requests:
              storage: 2Gi  
    
  4. Create LM.

    kubectl apply -f nebula-license-manager.yaml
    
  5. Verify that LM has been successfully deployed.

    kubectl -n nebula-license-manager get pods
    

Monitor LM

You can use monitoring tools, such as Dashboard Enterprise or Prometheus, to monitor the running status and metrics of LM. For more information, see Monitor LM.

Use LM to manage the license

  • For commands related to using LM deployed outside the K8s cluster to manage License, see License Manager.
  • Commands for managing the license using LM deployed within the K8s cluster are as follows:

    # View license information.
    kubectl -n nebula-license-manager exec -it nebula-license-manager-0 -- \
    /usr/local/nebula-license-manager/nebula-license-manager-cli info
    
    # Load the License Key. 
    # You must load the License Key before starting the NebulaGraph database.
    kubectl -n nebula-license-manager exec -it nebula-license-manager-0 -- \
    /usr/local/nebula-license-manager/nebula-license-manager-cli load --key XXXXX-XXXXX-...
    
    # View license quota usage.
    kubectl -n nebula-license-manager exec -it nebula-license-manager-0 -- \
      /usr/local/nebula-license-manager/nebula-license-manager-cli usage
    

Next to do

After deploying LM and loading the License Key, you need to configure the address and port of LM in the NebulaGraph Enterprise cluster through the licenseManagerURL parameter.

For more information, see Deploying Using Kubectl or Deploying Using Helm.


Last update: August 10, 2023