<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Operations on GeekyRyan</title><link>https://rnemeth90.github.io/tags/operations/</link><description>Recent content in Operations on GeekyRyan</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Fri, 27 Jun 2025 09:00:00 +0000</lastBuildDate><atom:link href="https://rnemeth90.github.io/tags/operations/index.xml" rel="self" type="application/rss+xml"/><item><title>Upgrading the Auxiliary Services That Keep a Kubernetes Platform Running</title><link>https://rnemeth90.github.io/projects/2025-06-27-kubernetes-auxiliary-service-upgrades/</link><pubDate>Fri, 27 Jun 2025 09:00:00 +0000</pubDate><guid>https://rnemeth90.github.io/projects/2025-06-27-kubernetes-auxiliary-service-upgrades/</guid><description>&lt;p&gt;Running a Kubernetes cluster means managing a whole ecosystem of supporting services - ingress controller, cert automation, cost visibility tooling, authentication proxy, metrics stack, autoscaling add-ons. They all need upgrades on their own cadence, and letting them rot is just as risky as letting Kubernetes itself fall behind.&lt;/p&gt;&#10;&lt;p&gt;This was a coordinated upgrade pass through that whole ecosystem - ingress, certs, autoscaling, auth proxy, the metrics and dashboarding stack. These services depend on each other (ingress and the auth proxy, metrics and autoscaler, etc.), so the upgrades had to be sequenced and tested carefully, not just done all at once.&lt;/p&gt;</description></item><item><title>A Grab Bag of Platform Environment Refinements</title><link>https://rnemeth90.github.io/projects/2025-01-27-platform-environment-refinements/</link><pubDate>Mon, 27 Jan 2025 09:00:00 +0000</pubDate><guid>https://rnemeth90.github.io/projects/2025-01-27-platform-environment-refinements/</guid><description>&lt;p&gt;Not every project fits into a neat story. Sometimes you&amp;rsquo;ve just got a pile of small things that need fixing in a given environment - configuration tweaks, resource layout improvements, general operational cleanup. We tracked this stuff under a single feature so it had visibility and could get prioritized alongside the bigger work.&lt;/p&gt;&#10;&lt;p&gt;The lesson here is simple: if you treat &amp;ldquo;the pile of small improvements&amp;rdquo; as a tracked item instead of tribal knowledge that only happens when someone has free time, it actually gets done and stays on the radar.&lt;/p&gt;</description></item><item><title>Annual TLS Certificate Renewal Across Customer-Facing Endpoints</title><link>https://rnemeth90.github.io/projects/2024-08-29-yearly-tls-certificate-renewal/</link><pubDate>Thu, 29 Aug 2024 09:00:00 +0000</pubDate><guid>https://rnemeth90.github.io/projects/2024-08-29-yearly-tls-certificate-renewal/</guid><description>&lt;p&gt;Every year: renew the TLS certificates before they expire. It&amp;rsquo;s not glamorous, but it&amp;rsquo;s one of the more embarrassing ways to take down a customer-facing service if you slip up on it.&lt;/p&gt;&#10;&lt;p&gt;This round meant hitting certificates in multiple places - NGINX endpoints, IIS services, and some Terraform configs where certs are hard-coded and need manual updates instead of just dropping in a new one. Each has its own deployment quirks, so the real discipline is keeping an accurate inventory of everywhere a cert lives, not just the obvious spots.&lt;/p&gt;</description></item></channel></rss>