Skip to content

Latest commit

 

History

History
75 lines (43 loc) · 3.81 KB

neptune-manual.md

File metadata and controls

75 lines (43 loc) · 3.81 KB
Raw

Amazon Neptune and graph-app-kit manual setup

For quick launch and an introduction, see the main graph-app-kit Neptune docs.

By using graph-app-kit with Amazon Neptune, you can visually explore graph database data and share point-and-click dashboard tools. This guides walks through manually launching Neptune, graph-app-kit, and connecting them. Alternatively, the CloudFormation templates enable quick launching preconfigured versions of both.

1. Manually setup Amazon Neptune

Ensure your Amazon Neptune database instance can be connected to by your graph-app-kit instance:

  • You must have or create an Amazon Neptune cluster. See the official Getting Started with Neptune docs.

  • Amazon Neptune clusters are hosted in a private VPC so the server hosting graph-app-kit must be granted access to the VPC. We strongly recommend hosting the graph-app-kit instance in a public subnet within the Neptune database's VPC.

  • If using IAM authorization on your Amazon Neptune cluster the sample code provided will need to be updated to support using SigV4 signing of requests. We recommend using this tool to simplify the process.

2. Manually launch and configure graph-app-kit for Amazon Neptune

Create an AWS EC2 graph-app-kit instance using the usual graph-app-kit first launch step, with the following launch settings:

  1. Set Network to the VPC ID value ("vpc-...") from 1. Setup Amazon Neptune (unless performing an alternative like manual VPC peering)
  2. Set Subnet to the PublicSubnet1 subnet ID value ("subnet-...") from 1. Setup Amazon Neptune
    • Auto-assign Public IP should default to Use subnet setting (Enable)

Continue through the graph-app-kit steps to download and build.

SSH into your graph-app-kit instance and set the following required environment variable configurations in your src/docker/.env file:

NEPTUNE_READER_PROTOCOL=wss
NEPTUNE_READER_HOST=<Insert Neptune DBClusterReadEndpoint like abc.xyz.mno.neptune.amazonaws.com>
NEPTUNE_READER_PORT=8182

For additional Neptune template options, see src/envs/neptune.env.

Reset and restart your graph-app-kit container:

cd src/docker
sudo docker-compose down -v
sudo docker-compose up -d

Watch logs with sudo docker-compose logs -f -t --tail=1

Access your Streamlit instance at http://the.public.ip.address:8501

4. Graph!

  • Go to your Streamlit homepage using the link from the launch section you followed
  • Select GREMLIN: SIMPLE SAMPLE from the dropdown to load a random sample of nodes from whatever Neptune database is connected
  • Continue to the instructions for creating custom views and adding common extensions like TLS, public/private dashboards, and more

Advanced Neptune Configuration

Run in an AWS VPC outside of the Neptune VPC

Follow AWS and Neptune VPC Peering instructions for mapping subnets across the VPCs.

Run outside of AWS and local developement

You can run Streamlit from outside of AWS. The challenge is that Amazon Neptune runs in a private VPC, meaning it exposes no internet-accessible API endpoints.

Options to connect to Amazon Neptune from a local computer include:

  • Setup an SSH tunnel for your internet connections

  • Configure a load balancer to expose the Neptune endpoint

  • Choose another way to configure secure access to a private VPC in AWS. Check the Getting Started with Neptune page for the current best recommended practices.