Skip to content

Latest commit

 

History

History
63 lines (42 loc) · 1.58 KB

README.md

File metadata and controls

63 lines (42 loc) · 1.58 KB
Raw

Docker-Live

A live lab environment for PISP Demos. Deployed using K8s + Helm

Prerequisites

  • helm v3 or later
  • kubectl and access to a running cluster
  • kubens
  • a kubeconfig file (e.g. ~/.kube/config)
  • a kubectl namespace, for this example, we will use pisp-lab

Installing

cd ./docker-live

# create and switch to namespace
kubectl create namespace pisp-lab
kubens pisp-lab

# install the switch
make install-switch

# install the participants
# Currently dfspa, dfspb, pispa
make install-participants

# wait for pods to be up and running
kubectl get po

# find the IP address of the load balancer and set ELB_URL variable
make get-elb

# Will print something like the following
# ****.elb.amazonaws.com

# Export
export ELB_URL=****.elb.amazonaws.com

# check the health of the services
make health-check

# seed the database
npm run reseed:docker-live

# run a test transaction
make transfer-p2p

Kubecl .yaml files + Helm Charts

For this lab, we use a simplified version of the Mojaloop Helm charts. This includes:

  • single instance MySQL
  • simplified Kafka with readymade public/kafka chart

Additionally, the install is broken into separate pieces:

  1. Prerequisites: mysql, kafka, with a combination of helm and kafka (see .install-base in the ./Makefile)
  2. Switch: The application expressed as helm charts in ./charts-switch
  3. Participants: The DFSP and PISP participants. Defined in helm charts at ./charts-participant, and customized using values.yml files for either a PISP or DFSP