Copying Databases

About this module

Next, you will learn the ways that you can copy a database.

At the end of this module, you will be able to:

Copy a Neo4j database using the copy command.
Create a dump file for a database.
Load a dump file to create a database.

Copying a Neo4j database

You or one of your application developers may want to copy a Neo4j database. Perhaps the developers are at a juncture in their development where they want to save what they have worked on thus far. Or perhaps they want to give a copy of the database to another developer to work with on a different system.

The structure of a Neo4j database is proprietary and could change from one release to another. You must never copy the database files from one location in the filesystem/network to another location.

There are two ways that you can copy a database:

Create a copy within the same Neo4j instance.
Create a copy (dump file) that you will give to another developer to load into their Neo4j instance running on a different system.

Before you copy a database it must be stopped.

Later in this training, you will learn how to back up a database while it is online.

Copying a database within the same Neo4j instance

To copy a database within the same Neo4j instance, the source database must be stopped. Additionally, the target database that you will be creating must not exist.

You use the neo4j-admin tool to copy a database within your Neo4j instance. Here are the steps:

Ensure the Neo4j instance is started.
In cypher-shell stop the source database you want to copy:
```
:USE system
STOP DATABASE <sourceDB>;
```

In a different terminal window, enter this as the user, neo4j:

/usr/bin/neo4j-admin copy --from-database=<sourceDB> --to-database=<targetDB>

In cypher-shell again:
```
CREATE DATABASE <targetDB>;
```

There are other options you can use to customize what is copied. See the Neo4j Operations Manual for details.

Creating a dump file from a database

To create a dump file of a database, you must:

Ensure the Neo4j instance is started.
In cypher-shell, stop the database you want to dump:
```
:USE system
STOP DATABASE <sourceDB>;
```
Ensure that the directory where you will dump the database exists.
As the neo4j users, you can use the dump command of the neo4j-admin tool to create the dump file as follows:
```
/usr/bin/neo4j-admin dump --database=<sourceDB> --to=<absolute path to dump file>
```

Note that the user, neo4j, must have write access to the directory.

This will create a dump file that can now be transferred between systems to create a new database on a different system.

Creating a database from a dump file

Then, if you want to create a database from any dump file (offline backup) to use for a Neo4j instance, you must:

Use the load command of the neo4j-admin tool as the neo4j user, to create the database from the dump file as follows:
```
/usr/bin/neo4j-admin load --database=<targetDB> --from=<absolute path to dump file>
```
In cypher-shell create the database:
```
:USE system
CREATE DATABASE <targetDB>;
```

Before you begin

Make sure you have a terminal window open to Docker Neo4j instance for this course.
Ensure that the Docker Neo4j instance is started.

Exercise steps:

Part 1: First you will create a movies database that will be used in this exercise for copying.

In cypher-shell create a database named movies.
Exit out of cypher-shell.

Invoke cypher-shell using movies.cypher as input and specify movies as the database.

On OS X or Linux:

cat ~/docker-neo4j/files/movies.cypher | docker exec --interactive neo4j bin/cypher-shell --database movies -u neo4j -p <passwordYouSpecified>

On Windows:

type files\movies.cypher | docker exec --interactive neo4j bin/cypher-shell --database movies -u neo4j -p <passwordYouSpecified>

In cypher-shell confirm that the database was loaded. It will contain 171 nodes:
```
MATCH (n) RETURN count(n);
```

Part 2: You will make a copy of the movies database, movies2 for local use in the Neo4j instance.

In cypher-shell stop the movies database.
Exit cypher-shell.

Copy the movies database to the movies2 database using the neo4j-admin tool:

[sudo] docker exec --interactive neo4j bin/neo4j-admin copy --from-database=movies --to-database=movies2

In cypher-shell create movies2.
Confirm that this movies2 database has 171 nodes.
```
MATCH (n) RETURN count(n);
```

Part 3: You will dump the movies database.

Use the dump command of the neo4j-admin tool to create the dump file as follows:
```
[sudo] docker exec --interactive neo4j bin/neo4j-admin dump --database=movies --to=data/movies.dump
```
This writes the movies.dump file to the $HOME/docker-neo4j/data directory which is available to the Docker Neo4j instance.

Confirm that the movies.dump file was created.

Part 4: You use the dump file to create a database.

Use the load command of the neo4j-admin tool to create the database, movies3 from the dump file as follows:

[sudo] docker exec --interactive neo4j bin/neo4j-admin load --database=movies3 --from=data/movies.dump

In cypher-shell create movies3.
Confirm that this movies3 database has 171 nodes.

Exercise summary

You have now gained experience copying a database within the Neo4j instance and also creating a dump file that can be used to create a database on a different system.

Check your understanding

Question 1

You want to make a copy of an existing database that is online in your Neo4j instance. What must you do before you copy it?

Select the correct answer.

Export the schema of the database.
Export the nodes and relationships in the database.
Lock the database.
Stop the database.

Question 2

What tools do you use the create a copy of a database within the same Neo4j instance?

Select the correct answers.

cypher-shell
neo4j copy
neo4j-admin copy
neo4j-admin dump

Question 3

What tools do you use the create a copy of a database that will be used on a different system?