Skip to content

Commit

Permalink
Publish: first blog outage
Browse files Browse the repository at this point in the history
  • Loading branch information
brandtkeller committed Dec 21, 2023
1 parent ef8aaf3 commit 4b8ecbd
Show file tree
Hide file tree
Showing 3 changed files with 47 additions and 3 deletions.
8 changes: 5 additions & 3 deletions content/posts/blogging/be-kind-choose-violence.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
---
title: "Be Kind - Choose Violence"
date: 2023-12-18T11:30:34-07:00
date: 2023-12-18T05:30:00-07:00
tags: ["Blogging", "Experiment"]
series: "blogging"
draft: false
Expand Down Expand Up @@ -36,7 +36,7 @@ Often we become beholden to practices that do nothing but slow progress and inno

Repeat - and when you hit those blockers - IE code review or some other time commitment - find out how to next best pivot and continue down a new thread.

Code stuck in review? Ping - be direct. No other options available? Execute on that next task with what you know now. Execute on next highest priority for your down. Know what it is you want to accomplish each day and execute.
Code stuck in review? Ping - be direct. No other options available? Execute on that next task with what you know now. Execute on next highest priority. Know what it is you want to accomplish each day and execute.

Finally instill this proactive behavior in others. Surround yourself with those who are Mission-Focused and you'll find that the progress you can have in a day and the ways you can challenge each-other are mind-boggling.

Expand All @@ -45,7 +45,9 @@ Finally instill this proactive behavior in others. Surround yourself with those
> ‘Be polite, be professional, but have a plan to kill everybody you meet.’
- General Mattis

Treat others with respect - Assume positive intent - Execute/Execute/Execute
Treat others with respect - Remove your ego - Assume positive intent - Execute/Execute/Execute

It's not one or the other - rather it is a pursuit of making myself better each day by engaging meaningfully with others and executing on the things I want to accomplish.

## Nerd Notes - New domain whodis

Expand Down
42 changes: 42 additions & 0 deletions content/posts/homelab/outage-report-23.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,42 @@
---
title: "First Blog Outage - Post Mortem"
date: 2023-12-21T05:30:00-07:00
tags: ["Homelab", "Experiment"]
series: "homelab"
draft: false
---

![](/images/coauthor.png)
(Picture of my co-author and I writing this post)

First off - this is meant to be a very light-hearted post. Something that I hope to impress on others is that Homelabs are a great opportunity to learn and grow your skills in new and relevant ways - but they come at the cost of being a labor of love. Something you maintain entirely in your free time and is best-effort to maintain and keep healthy.

## The problem

Shout out to Gavin for tossing me a message that something was wrong - I have no alerting on this setup and that should be a goal for future Brandt.

Nonetheless - attempting to hit my domain left me with a gateway error.

## Troubleshoot

The error was obvious very quickly - as noted in the original blog post - the infrastructure that supports this blog is 2x Raspberry Pi 4's running K3s in a Highly Available 2x Server w/ External Datastore configuration.

I switched my kubecontext to that cluster and was getting a connection error over Kubectl - so the problem looked to be with the cluster. SSH'd into one of the nodes and looked at `journalctl` - error was related to the external datastore/k3s having a problem with the certificates.

## Certificates

I run my "production" homelab cluster on separate infrastructure and use the crunchydata pg-operator to deploy postgres clusters - this is what I use for the blog k3s cluster datastore. I currently have it configured with cert-manager and a self-signed issuer.

Problem here is that I didn't modify some of my other letsencrypt certificate templates and the expiration was set for 90 days - well folks - we've been blogging for about that long now.

So Cert-manager rolled my self-signed certificates which the postgres cluster was updated to use and then all hell broke loose.

## Recovery

Recovery was actually pretty easy - bring the new ca/cert/key over to each node and restart the k3s service.... DONE!

Everything came back online and healthy!

## Next Steps

Still a work in progress - winter vacation has a limited amount of coding time available - Long term plan may be to re-work my postgres self-signed certificates to be more static - otherwise move my blog to be orchestrated on the "production" homelab cluster.
Binary file added static/images/coauthor.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

0 comments on commit 4b8ecbd

Please sign in to comment.