metadata

README

GoData

Simple and fast server that serves the NFTs written in Go.

Usage

go run ./... --port [port]

Enpoints:

• GET /:tokenId to get a metadata entry
• POST /:tokenId to add an entry (requires Authorization: [METADATA_API_KEY] header)
• GET / for healthchecks, returns empty 200 OK

Under the hood

Server uses Gin (gin-gonic) framework for serving metadata stored in Redis database as JSON. Deployment is onto a cluster provisioned from Linode. The format complies to the opensea.io metadata standards and employs NGINX ingress with TLS certificates from lets-encrypt through cert-manager.io. The API packed into a docker image of piotrostr/metadata is deployed aside a Redis image.

Deployment

1. The configuration file config.json is required for production deployments.

   field	value
   name	Collection name
   description	Description
   external_url	GS/S3/.. URL for placholder image
   placeholder_image	IPFS url for placeholder image

2. Redis Database is also required, without it the data endpoints return nil.

Note: The steps are user-specific since there is a number of variables like certificate issuer email etc, the steps are more of guidelines rather than walkthrough-tutorial or local setup.

Provision the cluster

cd terraform/linode && terraform apply
terraform output kubeconfig | jq -r '@base64d' > ~/.kube/lke.yaml
export KUBECONFIG=~/.kube/lke.yaml
cd -

Deploy to the cluster

1. Install ingress

   helm upgrade --install ingress-nginx ingress-nginx \
     --repo https://kubernetes.github.io/ingress-nginx \
     --namespace ingress-nginx --create-namespace
   

2. To get the IPv4 of the Ingress

   kubectl get services \
     --namespace ingress-nginx \
     -o wide \
     -w \
     ingress-nginx-controller
   

3. Add A-Record to the domain from the Ingress manifest (in this case metadata.smplvserse.xyz) pointing to the IPv4 of the Ingress

4. Install cert-manager

   kubectl apply -f https://github.com/cert-manager/cert-manager/releases/download/v1.8.0/cert-manager.yaml
   

5. Add secret (required env var for the metadata api)

   kubectl create secret generic metadata-api-key --from-literal METADATA_API_KEY=[secret]
   

6. Apply the configuration

   skaffold apply manifest.yaml
   

   I prefer skaffold to kubectl for applying deployments as it waits for them to stabilise and exits with error code 1 in case any container fails.