OpenShift is a comprehensive Kubernetes platform developed by Red Hat that provides a range of tools and services for building, deploying, and managing containerized applications.
For us, Mulesoft Developers and Architects, it is another important technology, because it's one of the supported platforms to host Runtime Fabric.
OpenShift is a Platform as a Service and can be deployed to multiple types of deployment.
In this post, we're going to have a look at the installation methods for OpenShift.
- Installer Provisioned
- User Provisioned
- Laptop/Desktop
Installer Provisioned
This is the method provided by infrastructure providers like:- AWS
- Azure
- GCP
- Red Hat OpenStack Platform (RHOSP)
- Red Hat Enterprise Virtualization (RHEV).
The installer deploys a bootstrap node and then the bootstrap node configures and deploys the master and worker nodes.
So, in theory, this is a very simple installation method, depending on how complex your infrastructure configuration must be.
User Provisioned
In this method, it is us, the user, who provisions the infrastructure for OpenShift. That means we need to provide all the nodes, the networking infrastructure, firewall/firewall rules, permissions, etc and then install OpenShift on top of that infrastructure.This method supports:
- AWS
- Azure
- GCP
- RHOSP
- VMware VSphere
- IBM Z
- IBM Power
- Bare metal
Laptop/Desktop
This is the easiest way to get an Openshift instance and start learning. It’s mainly built for testing. This method uses CodeReady containers.What are CodeReady Containers?
- It’s a Virtual Machine.
- It’s a minimal Openshift setup in a single VM
- It’s only for development and testing
The caveat of this type of deployment is that it’s ephemeral. All the storage is temporary. But this should not be a problem because what we want to do with it is just testing.
Another limitation is that when you start it up the operators are not enabled, just to speed up the VM startup. The last limitation is that there’s no upgrade path for this installation. If you need to upgrade to another version you basically need to delete the VM and install a new one with the newer version.
So, with all that said, this is NOT an option for Production environments