Skip to content

FAQ

This topic lists the frequently asked questions for using NebulaGraph Dashboard. You can use the search box in the help center or the search function of the browser to match the questions you are looking for.

"What are Cluster, Node, and Service?"

  • Cluster: refers to a group of systems composed of nodes where multiple NebulaGraph services are located.
  • Node: refers to the physical or virtual machine hosting NebulaGraph services.
  • Service: refers to NebulaGraph services, including Metad, Storaged, and Graphd services.

"What is the cluster status?"

The status of a cluster is as follows:

  • installing: The cluster is being created. The process will take about 3 to 10 minutes.
  • healthy: All services in the cluster are healthy.
  • unhealthy: There is an unhealthy service in the cluster service.

"Why authorizing nodes?"

Managing clusters requires the SSH information of the corresponding node. Therefore, you need to have at least an SSH account and the corresponding password with executable permissions before performing operations on Dashboard.

"What is scaling?"

NebulaGraph is a distributed graph database that supports dynamic scaling services at runtime. Therefore, you can dynamically scale Storaged and Graphd services through Dashboard. The Metad service cannot be scaled.

"Why cannot operate on the Metad service?"

The Metad service stores the metadata of the NebulaGraph database. Once the Metad service fails to function, the entire cluster may break down. Besides, the amount of data processed by the Metad service is not much, so it is not recommended to scale the Metad service. And we directly disabled operating on the Metad service in Dashboard to prevent the cluster from being unavailable due to the misoperation of users.

"What impact will the scaling have on the data?"

  • Scale out the Storaged service: Dashboard will create and start the Storaged service on the specified machine, which will not affect the existing data. You can choose to perform Balance Leader in the Storage Leader Distribution area and Balance Data in the Partition Distribution area on the Information->Overview Info page according to your own needs.
  • Scale in the Storaged service: Dashboard will not scale in Storage services until you execute Balance Data Remove to migrate all the partitions from the specified Storage service to other Storage services in the Partition Distribution area on the Information->Overview Info page.
  • Scaling the Graphd service will not affect the data.

"Why Dashboard Enterprise Edition cannot be started?"

  • Make sure that the license file is copied to the Dashboard directory and sudo ./dashboard.service start all is executed.
  • Make sure that the license is not expired.

You can also execute cat logs/webserver.log in the Dashboard directory to view the startup information of each module. If the above conditions are met but Dashboard still cannot be started, go to NebulaGraph Official Forum for consultation.

"Can I add the NebulaGraph installation package manually?"

You can add the installation package manually in Dashboard. To download the system and RPM/DEB package you need, see How to download NebulaGraph and add the package to nebula-dashboard-ent/download/nebula-graph. And you can select the added package for deployment when creating and scaling out a cluster.

Why does it prompt “SSH connection error” when importing a cluster?

If Service Host shows 127.0.0.1, and your Dashboard and NebulaGraph are deployed on the same machine when authorizing service hosts, the system will prompt "SSH connection error”. You need to change the Host IP of each service to the real machine IP in the configuration files of all NebulaGraph services. For more information, see Configuration management.

If you import a cluster deployed with Docker, it also prompts "SSH connection error". Dashboard does not support importing a cluster deployed with Docker.


Last update: September 16, 2022