Deployment

You will need to have Helm and kubectl installed locally and running kubernetes cluster.

provided script:

./deploy.sh -n golem

will deploy Golem with Redis, PostgreSQL and Nginx ingress to kubernetes namespace golem. Kubernetes Persistent Volumes are used to store Golem's data.

notes:

• by default, ingress is exposed under localhost:80, if you want to test golem-services in kubernetes locally (docker with kubernetes), and try to run commands with golem-cli, you may need to change golem-cli configuration to use that URL.
• by default, worker API Gateway endpoint is exposed under port 9006

Deployment in steps

create namespace

kubectl create namespace golem

PostgreSQL

https://artifacthub.io/packages/helm/bitnami/postgresql

install

helm upgrade --install -n golem golem-postgres oci://registry-1.docker.io/bitnamicharts/postgresql --set auth.database=golem_db --set auth.username=golem_user

delete

helm delete -n golem golem-postgres

get password (if you need it)

export GOLEM_POSTGRES_PASSWORD=$(kubectl get secret --namespace golem golem-postgres-postgresql -o jsonpath="{.data.password}" | base64 -d)

Redis

https://artifacthub.io/packages/helm/bitnami/redis

install

helm upgrade --install -n golem golem-redis oci://registry-1.docker.io/bitnamicharts/redis --set auth.enabled=true

delete

helm delete -n golem golem-redis

get password (if you need it)

export REDIS_PASSWORD=$(kubectl get secret --namespace golem golem-redis -o jsonpath="{.data.redis-password}" | base64 -d)

kubectl -n golem get service

service/golem-postgres-postgresql
service/golem-redis-master

ngnix ingress

install

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

you can watch the status by running

kubectl get service --namespace ingress-nginx ingress-nginx-controller --output wide --watch

Golem services

install

helm upgrade --install golem-default golem-chart -n golem

show kube files

helm template golem-chart

delete

helm delete -n golem golem-default

shell to the running pod/container

kubectl exec --stdin --tty -n golem  <pod> -- /bin/bash

scale worker executor deployment

kubectl -n golem scale deployment.apps/deployment-worker-executor-default --replicas=3