Skip to content
Joe Ferr
Joe Ferr

resume/portfolio

  • About me
  • RESUME/CV
  • Blog
Joe Ferr

resume/portfolio

Kubernetes

I started with Kubernetes around 2016. The company I was working for was looking to convert from a weblogic and tomcat centered monolith to using containers (docker at that time)

we knew that we wanted some sort of service discovery but it wasn’t clear what we needed since Kubernetes was far from mainstream at that point

we started looking at and POC’ing the netflix toolset (eureka etc) and we liked it but it was clearly not the complete answer to our problems and it didn’t seem like it was going to be the “winner” in this technology space

I was part of a cross-team architecture group that gather requirements, researched possible toolsets, did POC’s and finally chose and rolled out Kubernetes

We jumped into Kubernetes full speed ahead. I typically learn things best by using a tool or platforms API and writing code (typically python these days) so I started there

Very quickly I had written a kubernetes dashboard for visualization and with links/tools specific to our company tools (splunk, appdynamics at first, grype security scans first, then trivy, instana etc) and a set of rest endpoints to be used by our deployment tooling (which I also wrote…this was before so many ci/cd tools existed that supported kubernetes…but even to this day it does some things that you don’t get with any ci/cd tools or gitops solutions that I know of)

we used ansible for IAC for the on premise provisioning and used coreos, then flatcar as a small minimal linux distro with an atomic update for security pu rposes. We started with kubespray and I wrote and maintained 95% of this code. Later on we used some managed cloud kubernetes…we used Oracle’s cloud for that and I setup and managed the infrastructure as code (IAC) for that using Terraform. Generally we were on premise but our services at times used AWS resources

This was before vendors had built in pipelines (e.g. AZDO, github actions etc) so I wrote automation to manage our SDLC. It was shell and python (dockerized) managed via Jenkins and hits rest endpoints on the kubernetes dashboard that I wrote in order to get various statuses and information. I updated it a bit over the years but it’s still in use and it actually has more functionality that the “out of the box” functionality of many popular toolsets

©2025 Joe Ferr | WordPress Theme by SuperbThemes