Persistence Graph Node Setup

The graph node setup is running on a digitalocean managed kubernetes cluster and consists of

• IPFS Node
• Firehose Node
• Graph Node

How to run

Prerequisistes

• Kubernetes cluster and kubectl - How to install
• Postgres
  • Digital Ocean Managed DB
  • Running Locally? Download

Note
Install kind to run local kubernetes cluster.
This setup is using Digital Ocean's do-block-storage storage class for PVC (Persistent Volume Claim). Change it if you have different PVC storage class for your cluster. (standard for local kind cluster)

graph-node uses a few Postgres extensions. If the Postgres user with which you run graph-node is a superuser, graph-node will enable these extensions when it initalizes the database. If the Postgres user is not a superuser, you will need to create the extensions manually since only superusers are allowed to do that. To create them you need to connect as a superuser, which in many installations is the postgres user:

    psql -q -X -U <SUPERUSER> graph-node <<EOF
create extension pg_trgm;
create extension pg_stat_statements;
create extension btree_gist;
create extension postgres_fdw;
grant usage on foreign data wrapper postgres_fdw to <USERNAME>;
EOF

Update Config Map

There are three setups: (under networks folder)

• devnet
• testnet
• mainnet

Each setup has it's configmap.yaml which consists of three ConfigMaps:

1. persistencecore
2. firehose-node-config
3. graph-node-config

Note For devnet the configs can be used as is.

You need to update the

• snapshot_restore_url - make sure the snapshot is in .tar.gz format and created with v6.1.0 persistence binary.
• seeds
• first_streamable_block - from what height to start streaming the blocks (should be greater then the snapshot height)
• force_init - use this only if you want to delete everything and start again. (Make sure to set it to false once done)

Run

Use this command to apply the configs to your cluster

kubectl apply -k networks/devnet # change to the network you are using 

How to upgrade

Make sure that the persistenceCore has the fh tag for the upgrade version.
And it has the docker image published for the same tag.
Ex. For v6.1.0 it is v6.1.0-fh

Change the persistencecore image in graph-node/firehose-node.yaml file.
And run kubectl apply.

How to deploy a subgraph

NOTE: Ensure you have right kubeconfig for the testnet/mainnet setup in Digital Ocean. Also note that this is temporary setup. We plan to implement Auth based k8s setup as possible.

Steps

1. Port forward below ports your local machine(same or any unused port):-
   a) IPFS node: port 5001
   b) Graph node: port 8020
   example cmd: kubectl port-forward pod-name 5001:5001 --namespace graph-core-1)

2. To deploy a subgraph with its precise name, put & execute below steps in your package.json:-

  "codegen": "graph codegen",
  "build": "graph build",
  "create-local": "graph create --node http://localhost:8020/ <subgraph-name>",
  "deploy-local": "graph deploy --node http://localhost:8020/ --ipfs http://localhost:5001 <subgraph-name>",

3. Check the deployed subgraphs at respective GraphiQL endpoints:-
   GraphQL endpoint for testnet: https://graph.testnet.persistence.one
   GraphQL endpoint for mainnet: https://graph.mainnet.persistence.one