Skip to content
Developer Edition runs on one host and does not require a product license file. Need multi-host deployment or performance testing at scale? Explore Enterprise Edition

Connect and run your first query

Connect to NebulaGraph Database Developer Edition with NebulaGraph Studio, load the movie sample dataset, and find a movie and its actors. Complete the four steps below in your browser.

Before you start

1. Sign in to NebulaGraph Studio

On the development machine running Docker, open http://127.0.0.1:7001 in your browser.

Access from another computer or use a custom port

Your browser reaches NebulaGraph Studio on port 7001; NebulaGraph Studio connects to NebulaGraph Database Developer Edition at graph-cluster:9669.

From another computer, open http://<development-machine-ip>:7001. Replace <development-machine-ip> with the IP address of the machine running Docker.

If you changed the 7001:7001 port mapping in compose.yaml, use the development machine port to the left of the colon in the browser address.

Confirm or enter these values on the login page, then select Connect:

Field Value
Address graph-cluster
Port 9669
Username root
Password Your current root password. The initial password in the supplied file is NebulaGraph01; if you have changed it, enter the new password.

After signing in, you can see SCHEMA and CONSOLE in the top navigation bar.

2. Load the movie sample dataset

A Schema groups graph types and graphs, much like a folder. This page uses the default Schema /default_schema, which the system creates automatically.

  1. Select the SCHEMA tab in the top navigation bar.
  2. Check that default_schema is selected at the top of the left sidebar.
  3. Select Use Demo Dataset in the center of the page.
  4. Wait for loading to finish. On success, the movie_type graph type details page opens.

Loading creates graph type movie_type and graph movie, then inserts movies, actors, and their relationships. movie_type defines the data structure; the queries below use graph movie.

After loading succeeds, Use Demo Dataset is no longer shown. On subsequent visits, continue with step 3 without reloading the dataset.

3. Find a movie

Select CONSOLE in the top navigation bar. Paste this query into the query editor and select the run button to find the movie with ID 3:

USE movie MATCH (m@Movie {id: 3}) RETURN m.id AS movie_id, m.name AS movie

Expected result:

movie_id movie
3 "Shadows in Paradise"

The query uses these elements:

  • USE movie specifies the graph for this query.
  • MATCH finds matching nodes. @Movie specifies the movie type, and {id: 3} specifies the movie ID. An ID distinguishes movies with the same name.
  • RETURN selects properties to display; AS names the result columns.

movie is the graph name; Movie is a node type. Names are case-sensitive.

Set a working graph for the session

The queries on this page specify the graph with USE movie, which applies only to the query that contains it. You can instead set a working graph for the session:

SESSION SET GRAPH movie

Then omit USE movie from the start of the query:

MATCH (m@Movie {id: 3}) RETURN m.id AS movie_id, m.name AS movie

Set the working graph again after reconnecting. See USE for details.

4. Find the movie's actors

In the same console, replace the movie query with the following statement and run it to find the movie's actors:

USE movie MATCH (a@Actor)-[:Act]->(m@Movie {id: 3})
RETURN a.id AS actor_id, a.name AS actor, m.name AS movie
ORDER BY actor_id

Expected result:

actor_id actor movie
4826 "Matti Pellonpää" "Shadows in Paradise"
4828 "Sakari Kuosmanen" "Shadows in Paradise"
5999 "Kati Outinen" "Shadows in Paradise"

(a@Actor)-[:Act]->(m@Movie) matches an actor's relationship to a movie they act in. The arrow points from actor to movie. ORDER BY actor_id sorts by actor ID.

Exit and resume your work

In NebulaGraph Studio, select Logout from the user menu in the upper right corner. In ngql, run :exit. After you exit the client, NebulaGraph Database Developer Edition continues running and retains its data. To stop the services, see Stop and restart the services.

Next time, sign in to NebulaGraph Studio and continue from step 3 to run the queries. You do not need to reload movie.

Other connection methods

These clients access the same NebulaGraph Database Developer Edition instance. After loading the data once, you can query it with any client.

Connect and load sample data with ngql

  1. In the development machine's terminal, open the directory containing compose.yaml and run:

    docker compose exec ngql ngql --host graph-cluster --port 9669 --user root

    The first ngql is the Compose service name; the second is the client command inside the container. Keep graph-cluster:9669, the query address on the container network.

  2. At the Password prompt, enter the current root password and press Enter.

  3. If movie has not been loaded in /default_schema, run this command. Skip this step if loading has already succeeded.

    :play movie

    The client displays Playing dataset: movie... at the start and Play dataset: movie done. on success. If loading fails, inspect the created objects and errors as described in Data loading.

Then run the examples from Find a movie and Find the movie's actors in the ngql interactive console.

For a multiline query, place ''' on its own line before and after the statement. Enter the closing line and press Enter to submit:

'''
USE movie MATCH (a@Actor)-[:Act]->(m@Movie {id: 3})
RETURN a.id AS actor_id, a.name AS actor, m.name AS movie
ORDER BY actor_id
'''

:play movie is an ngql client command. It cannot run in the query editors in NebulaGraph Studio or the built-in web console.

Connect with the built-in web console

NebulaGraph Database Console is a query interface built into the Graph service. It has a separate address from NebulaGraph Studio:

Browser location Address
On the same development machine as Docker http://127.0.0.1:19669/ui/console
On another computer http://<development-machine-ip>:19669/ui/console

Replace <development-machine-ip> with the IP address of the machine running Docker. The supplied configuration uses HTTP. If you changed the 19669:19669 mapping, use the development machine port to the left of the colon.

Enter username root and the account's current password, then select Connect. Run the statements from step 3 and step 4 one at a time in the editor, selecting the run button for each. Paste multiline statements directly.

If the data is not loaded yet, load it through NebulaGraph Studio or ngql first, then return here to query it.

Troubleshooting

Check Studio access and sign-in

Use the actual password when signing in. Each $$ in compose.yaml represents a single $ in the actual password.

If the values are correct, check service status to confirm that graphd is running. In your development machine's terminal, in the deployment directory, view the Graph service logs and check the reported errors:

docker compose logs --tail 100 graphd
Check sample data loading

If loading fails, check the errors and created objects as described in Data loading. You do not need to reload movie after it has loaded successfully. Repeated loads can cause name conflicts.

Verify the query connection

To test the connection separately, select CONSOLE in the NebulaGraph Studio top navigation bar. Paste this statement into the input area and select the run button:

RETURN 1 AS result

It should return a column named result with the value 1. This statement does not depend on sample graph data, so you can use it to check the connection and basic query execution.

Check query results

If a query fails or returns no rows, check the capitalization of graph name movie and node type Movie. Then run these statements one at a time in the same console:

SHOW CURRENT_SCHEMA

The path in the result should be /default_schema. If it differs, run SESSION SET SCHEMA /default_schema to switch back to the default Schema.

SHOW GRAPH TYPES

The result should list graph type movie_type.

SHOW GRAPHS

The result should list graph movie, with graph type movie_type and Schema /default_schema.

If loading was interrupted, the graph's existence does not confirm that its data is complete. Check the errors and existing objects as described in Data loading before continuing. Repeated loads can cause name conflicts.

What's next