Agola - Install on Openshift/Minishift 3.11 (4 dummies)


  1. You have access to a running OpenShift/Minishift instance

  2. You have oc tools installed locally and can connect to your OpenShift instance

Login as developer

oc login -u developer

oc new-project agola

oc new-app agola

At this point login to the OpenShift console as developer and enter into the “agola” project

I modified the file

to create an openshift compliant deployment config file.
Download from here:

Step by Step add portion of deployment config into your project

Select from “Add to Project” menu, Import YAML/JSON

Paste into dialog box the following yaml code to create a PersistentVolumeClaim and then click “Create” button

apiVersion: "v1"
kind: "PersistentVolumeClaim"
  name: "agola-vol"
    - "ReadWriteOnce"
      storage: "1Gi"

As for the above code copy and paste the following:


apiVersion: v1
kind: Service
  name: agola
    - port: 8000
      name: api
      nodePort: 30002
    app: agola
  type: NodePort

and then

apiVersion: v1
kind: Service
  name: agola-internal
    - port: 8000
      name: api
    - port: 4000
      name: runservice
    - port: 4002
      name: configstore
    - port: 4003
      name: gitserver
    app: agola
  clusterIP: None

Go to Applications → Services panel

Click on agola-internal link, copy Hostname value agola-internal.agola.svc.cluster.local

and use it to replace the urls where present agola-internal into the deployment config file.

Add RBAC Authorization

Logout, and login again as user admin with password admin, enter to the agola project
and select Import Yaml /Json as into below figure

Into the following dialog window past the content of the file one section at a time, pay attention to replace all occurrences of “namespace: default” with “namespace: agola”

Logout, and login as developer.

Select agola project, enter Application→Services section, select agola service

In the Service agola section select “create route” as in the image below.

In the following page select Create without editing anything

At this point you have the url of your Agola as in the below figure, but we need to recreate the pod to apply all changes made.

Go to Application - Pods section,

select the pending pod, by the Actions button click Delete and wait until the system recreate the pod.

At the end by a link like this

you can load Agola and follow the instruction in the official site to use it:


@rafpor75 Thanks for your howto!

I’d like to ask you if we can highlight the differences between a standard k8s agola deployment and a openshift one (i.e. the route resource that doesn’t exists in standard k8s) and then put them
in the k8s examples in the agola repository (creating some pull requests) to cover both kind of deployments. One possiblity will be to add more comments to highlight the openshift differences and also define additional openshift resources like routes commented out. So we’ll avoid having two diverging examples but just a single more commented one.

In the end users should be able to “automate” the deploy of agola also with openshift without doing manual steps using the ui.


Hi Simone, to automate the distribution in openshift, I should have to create an openshift template as deployment config, so the best thing could be to create a well commented deployment file (template) to add as example into k8s examples section, as you have already suggested. As soon as the file is ready, I will perform a pull request.