runner-manual-install

Installing the Codefresh runner manually

What you need

1. A computer with Linux that has bash/openssl/base64 installed
2. Codefresh CLI installed (v0.30.0+)
3. kubectl installed and with access to the Kubernetes cluster that will host the Codefresh runner

Step 1 - Prepare your cluster

First create a token that the runner will use for authentication (use any name you want)

codefresh create token behind-the-firewall

Note down the value. And export it as a variable

export API_TOKEN=5d4005badbb93850c1db7dsg345fggdfgcd7f71c6a246666182d

Create a namespace in your cluster that will hold the Codefresh runner and give it the privilages you want. The easiest way is to use the default namespace and give admin access to Codefresh.

kubectl create clusterrolebinding default-admin --clusterrole cluster-admin --serviceaccount=default:default

export NAMESPACE=default

Add your cluster to Codefresh. Here we use the name my-cluster for the integration.

codefresh create cluster my-cluster --namespace default --serviceaccount default --behind-firewall --kube-context my-gke-cluster-project-name

Export your cluster as a variable

export CLUSTER_NAME=my-cluster

Step 2 - Select Codefresh runner version

Export the variables used in the manifests. The manifests can be found in the assets folder.

export APP_NAME=venona
export NAMESPACE=default
export CLUSTER_NAME=my-cluster
export API_TOKEN=5d4005badbb93850c1db7dsg345fggdfgcd7f71c6a246666182d
export API_HOST=https://g.codefresh.io
export VOLUME_IMAGE_NAME=codefresh/dind-volume-provisioner
export VOLUME_IMAGE_TAG=v13
export VERSION=1.0.0
export MODE=InCluster
export IMAGE_NAME=codefresh/venona
export IMAGE_TAG=0.24.0

Step 3 - Create certificates, sign them and apply manifests

Run the following

chmod +x ./template.sh
chmod +x ./install-runner.sh
./install-runner.sh

After installation is complete you will see your new runtime environment in Codefresh and can use it in your pipelines. Please wait until all pods have finished their startup.

You can use the command kubectl get pods to see the progress.

The rendered manifests as they were applied are in ./tmp/codefresh/manifests.