Skip to content

Performance-aware DevOps blueprint environment based on RedHat's fabric8

License

Notifications You must be signed in to change notification settings

jonasheinisch/blueprint-fabric8

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

48 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

blueprint-fabric8

Performance-aware devops blueprint environment based on RedHat's fabric8

This repository contains an ansible playbook and an example host-file to setup an OpenShift cluster (this is based on https://github.com/openshift/openshift-ansible and deploy fabric8 on top of that.

The aim of this playbook is to setup a cluster with multiple nodes. If you want to setup a local single-node cluster as a testing environment, you should consider using fabric8 with minishift (cf. http://fabric8.io/guide/getStarted/minishift.html).

Requirements

Configuring the Installation

To configure the installation have a look at the oc.hosts file. To get further help, have a look at https://docs.openshift.org/latest/install_config/install/advanced_install.html.

Running the Ansible Playbook

To execute the playbook you first need to clone the openshift-ansible repo as a submodule:

$ git submodule add [email protected]:openshift/openshift-ansible.git openshift-ansible

The playbook will setup Docker, an OpenShift-Cluster and will run the fabric8-Installer gofabric8 on the master.

$ ansible-playbook oc-setup.yml -i oc.hosts

fabric8 user

To use fabric8 you can now log into the master node via SSH and setup a user:

$ sudo htpasswd /etc/origin/master/htpasswd <username>

To add cluster-admin rights to this user:

$ oadm policy add-cluster-role-to-user cluster-admin <username>

use fabric8

Now you can use fabric8 by simply accessing http://fabric8.oc.master.example.org (or whatever your domain is). Login with the user just created.

Troubleshooting

Ansible reports UNREACHABLE

In our testing environment sometimes during playing the playbook there has been an error reporting an SSH-error resulting in an abort with the failure UNREACHABLE. There seems to be no specific reason (probably just SSH temporarily unavailable) and you can just rerun the playbook from the beginning.

fabric8 is not available

Ensure the router and registry pods are running:

$ oc get pods

If they are not running (pending) you can check the reasons for that by entering:

$ oc describe pod <podname>

If your master node is not marked as schedulable, this could be a reason. Schedulable means that a pod can be deployed on this node. In this case run:

$ oadm manage-node master.example.com --schedulable=false

##Further information

These are some helpful links where you can get help, if you are in trouble with the installation:

About

Performance-aware DevOps blueprint environment based on RedHat's fabric8

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published