Database Reference
In-Depth Information
.loadPropertiesFromFile(machineAFilename)
.newGraphDatabase();
Server mode
Single server instances that are not currently running against the Enterprise edition will
first need to be upgraded to the Enterprise edition. Then it's simply a case of configuring
the server to be part of an appropriate cluster.
11.2.2. Setting up a Neo4j cluster
Let's go through the process of setting up a three-machine, server-based cluster. Neo4j HA
works just as well for embedded instances. As this is primarily going to be used for demo
purposes, we'll run all three instances on the same machine; but in production you'd more
than likely run different instances on different machines.
Note
These instructions are for *nix environments (the actual examples were done on OS X
10.9.2), but an equivalent process can be followed for a Windows setup. We ran this HA
setup against Neo4j version 2.0.2.
Initial setup
Follow these steps to perform the initial setup:
1
. Create a base parent directory, such as ~/n4jia/clusterexample.
2
. Download the Enterprise version of Neo4j and unpack the zip file (for Windows)
or tar file (for Unix) into this folder. On the Mac, this would look like the following:
~/n4jia/clusterexample/neo4j-enterprise-2.0.2
3
. Rename this folder machine01, and duplicate the directory, along with all of
its content, into directories machine02 and machine03. Your resulting file structure
should look like this:
