Halin Neo4j Monitoring
Halin is a Graph App for monitoring Neo4j.
- Performance monitoring (system load, memory usage)
- Checks your Neo4j configuration and finds problems, makes suggestions on how to improve.
- Works with both Neo4j Enterprise and Neo4j Community, clusters and stand-alone.
- User & Role Management
Deploying Halin on Your Own (e.g. behind a company firewall)
yarn installto pull dependencies
- Edit package.json locally, and change the “homepage” field to the URL where Halin
will be accessible when deployed. For example, if you’re using an internal company
https://fileserver.mycompany.comyou might set this to
npm run buildto build the finished HTML/CSS artifacts; this will produce a
distfolder containing everything needed to be deployed.
- Copy everything in the
distfolder to your file sharing server; in the example above that would be a
halindirectory hosted by
Halin’s build chain already does all of these steps; the public version of Halin is hosted
in an AWS S3 bucket. If you’d like to see a scripted version of how that was done,
.circleci/config.yml file, and look for the “Deploy to S3” step.
Running Halin in Development Mode
- Clone the repo
- Navigate to
Running in Docker
docker run -d -p 127.0.0.1:3000:3000 --rm -t mdavidallen/halin:latest
Open a browser to http://localhost:3000/
If you’d like to build the Halin docker container from source:
docker build -t halin:latest -f Dockerfile .
Running as a GraphApp
- Inside of Neo4j Desktop, go to application settings, scroll all the way to the bottom, enable development mode
- For development app entry point, use http://localhost:3000/ (that’s where yarn start will serve the app)
- For development app root path, enter /absolute/path/to/halin where you cloned the repo
- Finally, inside of desktop you’ll see a special tile labeled “Development App 9.9.9”. This will point to your running copy of Halin
- Does it support Neo4j Enterprise or Community?
Both. Some features must be disabled for community (such as user management) because community does not support them. For a comparison of features between Neo4j Enterprise and Community, see this link
- Does Halin capture any data about my database?
Halin runs purely in your browser and does not share details of your database outside of your machine. Even when you generate a diagnostic package, this is not shared. Halin does use a JS error reporting library called Sentry to help catch errors and improve the software, but nothing about your configuration is sent back to me. You may optionally tick a box in the diagnostics pane to “opt-in” share your diagnostic package, which helps us improve Halin & Neo4j.
- What versions of Neo4j does Halin support?
Supported: 3.4.* and 3.5.*; causal cluster and stand-alone, enterprise and community.
Unsupported: Neo4j < 3.4, embedded deployments, and HA deployments.
- How does Halin work?
Halin uses pure cypher/bolt connections to nodes in your cluster, and uses Neo4j’s existing management interfaces and queries, including things like JMX, to obtain all statistics. This approach will work on any Neo4j instance with zero modification or configuration.
If you’d like to see what kinds of queries Halin is running, click on the gear icon at the very bottom and you’ll get a pop-up showing all of the queries and their response times.
- Do I have to run Halin as an admin user?
No, but some types of data (for example your configuration) are not available unless you are an administrator. If you run halin with a low privileged user, some features may not be available or work as expected. An admin user is highly recommended.
- Why doesn’t (this feature) appear in Halin? Why can’t I use User administration?
Depending on how your database is configured, some options are not available. For example, active running query monitoring requires procedures that are only available in Neo4j Enterprise, so if you’re running Community this feature will be disabled. This is not due to a limitation of Halin, simply which functionality the Neo4j instance exposes.
Additionally, there are some special circumstances. For example if you use an LDAP authorization provider, Halin will not permit adding users because Neo4j itself is not the authority on the user set.
- Why is it called Halin?
- How much load does Halin put on my cluster?
Halin polls your Neo4j cluster by sending a variety of queries concurrently to each member of the cluster. Generally, these queries are answered very quickly by Neo4j, requiring a few milliseconds per query. The total load that Halin will place on the average small database, for example one running in Neo4j Desktop, will generally be about 1% of its load, with larger clusters experiencing a small fraction of a percent of the concurrent queries Neo4j can handle, and CPU. Further, these queries are strictly read-only (unless you are creating users/roles) and generally require little heap. All in all, the load Halin places should be negligible.
Installing Halin in Neo4j Desktop
Neo4j Desktop minimum version 1.1.10 is required.
- Open Neo4j Desktop
- Click on the “Graph Applications” tile at left
- Scroll to the bottom where it says “Install Graph Application”
- Enter the URL
https://neo.jfrog.io/neo/api/npm/npm/halininto the box
- Click install
That’s it. As Neo4j Desktop is installed, and as new versions of Halin are available, you’ll catch updates.
Serve in Neo4j Desktop env
To serve it as a graph application, read the docs for that in the Neo4j Desktop development pages.
This is a short overview of how Halin is put together, with hopes that it’s useful for those hacking on it or wanting to extend it.
The entirety of the application is broken into two trees:
- API, where all of the Neo4j, network, and datastructure parts reside, independent from Halin’s React-based UI, and
- Components, where all react-based content lives.
Core API Parts
The HalinContext class is a global and gets attached to the window object. It always has an array of ClusterMember instances. Halin treats single-node databases as a cluster with only one member. The HalinContext also has a ClusterManager object it can create, which handles cluster- wide operations, such as mapping queries across a cluster (user management).
Both the HalinContext object and the ClusterMembers that it has have the concept of feature probes; when Halin starts up it gathers basic information about which sorts of features are supported by the database, which version, enterprise vs. community, and so on. In this way, the UI can be adapted to a different layout depending on what the DB exposes and its version.
Halin exposes a “Query Library” which centralizes all of the queries Halin runs, and the display columns used in the UI corresponding to the data that comes back from the server.
A Datafeed is a polling data stream that requires a query, a set of display columns, and a rate. The datafeed can notify the caller when new data is available, and can handle a variety of error scenarios. Most visual components in Halin are backed by a Datafeed object.
Neo4j driver management is done centrally. Components are discouraged from creating drivers or even using them. By using the ClusterMember object to run queries, we get to centrally manage all of that and also track performance and errors. Additionally, Halin uses “session pooling” on top of the standard Neo4j driver. This was introduced because session creation/destruction requires extra roundtrips in the bolt protocol, and session reuse is desirable for improving latency to/from Neo4j.
Sentry is used throughout for error detection and reporting, and telemetry.
Higher order components (HOCs) are provided which allow other components to express their requirements. For example, in order to display metrics, those metrics have to be enabled server side. In order to administer users, you have to be using enterprise, and so on.
Each of the components then expresses requirements by wrapping themselves in a higher order component, or none, if they work anywhere. This allows the HOC to check the various feature probes in the API and selectively display components.
Key React Components
Most of the actual display components are quite straightforward; the most complex bits of Halin are in dealing with how different the product surface for Neo4j is depending on the version and available features that are exposed.
The key react components to have a look at are CypherDataTable, CypherTimeseries, and ClusterTimeseries.