Getting Started #

Prerequisites #

• Docker installed. You can find a guide here.
• Docker-Compose installed. You can find the installation guide here. For Installing on ARM devices, use this guide.
• Kubectl installed. Official installation documentation.
• Kubernetes installed and cluster configured. Official installation guide under this link.

Seting Up the Project #

In this section, we go through the steps of setting up the project:

1. Clone the project from the GitLab repository.
2. Download the training data and genesis transaction from here.
3. Put the data folder (contains training data) into the peer folder.
4. Put the genesis.npy (example genesis transaction) file form the tangle folder into the data folder.

Building and Pushing Images #

To benefit from the newest changes made in the code, the images need to be built and pushed to the container registry. The images will be pushed to a private repository on DockerHub. Let’s go step by step through the process of building and publishing the images to the private registry:

1. Login to DockerHub with the docker login command. The username of the private registry is hpimpss2020 and the password can be obtained from the creators of this project.
2. Depending on the target architecture(s) for the network run the following command to build and push your images:

docker-compose build && docker-compose push

docker-compose -f docker-compose.yml -f docker-compose.arm.yml build && docker-compose -f docker-compose.yml -f docker-compose.arm.yml push

docker-compose -f docker-compose.yml -f docker-compose.arm.yml -f docker-compose.arm64.yml build && docker-compose -f docker-compose.yml -f docker-compose.arm.yml -f docker-compose.arm64.yml push

Running with Kubernetes #

Configuring peer deployment #

Different components of the project need to be deployed on the Kubernetes cluster. You can find all service, deployment, config, and role files in the k8s folder in the root of the project. You can configure and change each deployment file. This includes especially the peer-deployment.yaml file, in which you can set multiple environment variables to configure your peer accordingly. The environment variables are described in the table below:

Setting up image pull secret #

A Kubernetes cluster uses the secret of docker-registry type to authenticate with a container registry to pull a private image.

If you already ran docker login, you can copy that credential into Kubernetes:

kubectl create secret generic regcred \
    --from-file=.dockerconfigjson=<path/to/.docker/config.json> \
    --type=kubernetes.io/dockerconfigjson

You can find more about pulling images from a private registry in the official Kubernetes documentation.

Deploy components on cluster #

To deploy different components like B_ootstrap_, _Peer_, _Grafana_, _Prometheus_, and _Visualization_ on the cluster you can execute the following command from the root of the project:

kubectl apply -f k8s/

If you want to deploy on a specific architecture (e.g., arm64) change the image names in k8s/*-deployment.yaml

To scale the number of peers you can use this command:

kubectl scale --replicas=<number-of-replicas> -f k8s/peer-deployment.yaml

Accessing the Services on Kubernetes #

After successfully deploying all the components, you can access each component using its designated service. Each service uses the LoadBalancerIP and the port it is exposing. In other words, by accessing http://<Service-LoadBlancerIP>:<Service-Port> in your browser you will be able to reach each service’s HTTP server. In the table below you can see each service and the port number exposed by default:

Note that there is no need to reach every service from the outside of the cluster. Services for Peer and Bootstrap are used internally for services discovery and pass data between pods.