Here’s a summary of the lab I’m slowly building at home to model those I use in production on the job. I still feel it is well worth the cost and effort to do this at home “on prem” over using a cloud provider because the entire point is being able to be productive on-prem or in the cloud. Using only a cloud service provider robs me of the critical learning opportunities setting something up that is done “for me” by those providers (like Kubernetes Ingress, CNI, IAM, DNS, etc.)
Keep in mind the goals of this lab are not to keep things the same, but to have a sample of the most likely tech to encounter in the enterprise, (for example, both RedHat and Ubuntu servers).
Here are the stand-alone components:
Kubernetes with HA will have its own batteries-not-included to setup as well (which is why you want a home lab and not some cloud provider that does this all for you robbing you of the learning opportunity):
A year ago I couldn’t even tell you what all that architecture was about and why it was needed. Unfortunately, I can tell you why it is needed now, but still struggle to keep my management skills of all of it up to snuff.
I’ll be doing most everything from hardware and “netbooting” but I will do a Proxmox collection of VMs for some quick cloud applications as well, even though most of my work will be involving in-cluster Kubernetes application builds.
#learning #k8s #boost