Deploy Studio¶

This topic describes how to deploy Studio locally by Docker, RPM, and tar package.

RPM-based Studio¶

Prerequisites¶

Before you deploy RPM-based Studio, you must confirm that:

• The Nebula Graph services are deployed and started. For more information, see Nebula Graph Database Manual.

• If your Linux distribution is CentOS, install lsof and Node.js of versions above v10.16.0+.

  Note

  node and npm should be installed in /usr/bin/ directory. Avoid the situation that the node command cannot be found during RPM installation. For example, the default directory of nodejs12 is in /opt/rh/rh-nodejs12, you can use following commands to build soft link:

  $ sudo ln -s /opt/rh/rh-nodejs12/root/usr/bin/node /usr/bin/node
  $ sudo ln -s /opt/rh/rh-nodejs12/root/usr/bin/npm /usr/bin/npm
  

• Before the installation starts, the following ports are not occupied.

  Port	Description
  7001	Web service provided by Studio.
  8080	HTTP service provided by Nebula HTTP Gateway.

Install¶

1. Select and download the RPM package according to your needs. It is recommended to select the latest version. Common links are as follows:

   Installation package	Checksum	Nebula version
   nebula-graph-studio-3.1.0-1.x86_64.rpm	nebula-graph-studio-3.1.0-1.x86_64.rpm.sha256	v2.6.1

2. Use sudo rpm -i <rpm> to install RPM package.

   For example, install Studio 3.1.0, use the following command:

   sudo rpm -i nebula-graph-studio-3.1.0-1.x86_64.rpm
   

   When the screen returns the following message, it means that the PRM-based Studio has been successfully started.

   egg started on http://0.0.0.0:7001
   nohup: Add the output to "nohup.out"
   

3. When Docker-based Studio is started, use http://ip address:7001 to get access to Studio.

   Note

   Run ifconfig or ipconfig to get the IP address of the machine where Docker-based Studio is running. On the machine running Docker-based Studio, you can use http://localhost:7001 to get access to Studio.

   If you can see the Config Server page on the browser, Docker-based Studio is started successfully.

   

Uninstall¶

Users can uninstall Studio using the following command:

sudo rpm -e nebula-graph-studio-3.1.0-1.x86_64

Exception handling¶

If the automatic start fails during the installation process or you want to manually start or stop the service, use the following command:

• Start the service manually

  bash /usr/local/nebula-graph-studio/scripts/rpm/start.sh
  

• Stop the service manually

  bash /usr/local/nebula-graph-studio/scripts/rpm/stop.sh
  

If you encounter an error bind EADDRINUSE 0.0.0.0:7001 when starting the service, you can use the following command to check port 7001 usage.

lsof -i:7001

If the port is occupied and the process on that port cannot be terminated, you can use the following command to change Studio service port and restart the service.

 //Open the configuration file
 $ vi config/config.default.js

 //Change the port number
 ...
     config.cluster = {
         listen: {
             port: 7001, // Modify this port number and change it to any one currently available
             hostname: '0.0.0.0',
         },
     };
 ...

 //Restart npm
 $ npm run start

tar-based Studio¶

Prerequisites¶

Before you deploy tar-based Studio , you must do a check of these:

• The Nebula Graph services are deployed and started. For more information, see Nebula Graph Database Manual.

• The Linux distribution is CentOS, installed lsof and Node.js of version above v10.16.0+.

  Note

  node and npm should be installed in /usr/bin/ directory. Avoid the situation that the node command cannot be found during RPM installation. For example, the default directory of nodejs12 is in /opt/rh/rh-nodejs12, you can use following commands to build soft link:

  $ sudo ln -s /opt/rh/rh-nodejs12/root/usr/bin/node /usr/bin/node
  $ sudo ln -s /opt/rh/rh-nodejs12/root/usr/bin/npm /usr/bin/npm
  

• Before the installation starts, the following ports are not occupied.

  Port	Description
  7001	Web service provided by Studio
  8080	Nebula-http-gateway, Client's HTTP service

Install¶

1. Select and download the tar package according to your needs. It is recommended to select the latest version. Common links are as follows:

   Installation package	Studio version
   nebula-graph-studio-3.1.0-1.x86_64.tar.gz	3.1.0

2. Use tar -xvf to decompress the tar package.

   tar -xvf nebula-graph-studio-3.1.0-1.x86_64.tar.gz
   

Procedure¶

1. Deploy and start nebula-http-gateway.

   $ cd nebula-http-gateway
   $ nohup ./nebula-httpd &
   

2. Deploy and start nebula-graph-studio.

   $ cd nebula-graph-studio
   $ npm run start
   

   Caution

   Studio 2.6.1 version is not dependent on nebula-importer, so the installation and deployment procedure is different from Studio v3.0.0.

3. When tar-based Studio is started, use http://ip address:7001 to get access to Studio.

   Note

   Run ifconfig or ipconfig to get the IP address of the machine where Docker-based Studio is running. On the machine running Docker-based Studio, you can use http://localhost:7001 to get access to Studio.

   If you can see the Config Server page on the browser, Docker-based Studio is started successfully.

   

Stop Service¶

You can use kill pid to stop the service:

$ kill $(lsof -t -i :8080) # stop nebula-http-gateway
$ cd nebula-graph-studio
$ npm run stop # stop nebula-graph-studio

Docker-based Studio¶

Prerequisites¶

Before you deploy Docker-based Studio, you must do a check of these:

• The Nebula Graph services are deployed and started. For more information, see Nebula Graph Database Manual.

• On the machine where Studio will run, Docker Compose is installed and started. For more information, see Docker Compose Documentation.

• Before the installation starts, the following ports are not occupied.

  Port	Description
  7001	Web service provided by Studio
  8080	Nebula-http-gateway, Client's HTTP service

Procedure¶

To deploy and start Docker-based Studio, run the following commands. Here we use Nebula Graph v2.6.1 for demonstration:

1. Download the configuration files for the deployment.

   Installation package	Nebula Graph version
   nebula-graph-studio-v3.tar.gz	v2.6.1

2. Create the nebula-graph-studio-v3 directory and decompress the installation package to the directory.

   mkdir nebula-graph-studio-v3 && tar -zxvf nebula-graph-studio-v3.gz -C nebula-graph-studio-v3
   

3. Change to the nebula-graph-studio-v3 directory.

   cd nebula-graph-studio-v3
   

4. Pull the Docker image of Studio.

   docker-compose pull
   

5. Build and start Docker-based Studio. In this command, -d is to run the containers in the background.

   docker-compose up -d
   

   If these lines are returned, Docker-based Studio v3.x is deployed and started.

   Creating docker_client_1   ... done
   Creating docker_web_1      ... done
   Creating docker_nginx_1    ... done
   

6. When Docker-based Studio is started, use http://ip address:7001 to get access to Studio.

   Note

   Run ifconfig or ipconfig to get the IP address of the machine where Docker-based Studio is running. On the machine running Docker-based Studio, you can use http://localhost:7001 to get access to Studio.

   If you can see the Config Server page on the browser, Docker-based Studio is started successfully.

   

Next to do¶

On the Config Server page, connect Docker-based Studio to Nebula Graph. For more information, see Connect to Nebula Graph.