#devops
16 posts
2026 - May
1 post2026 - March
1 post2026 - February
1 post2025 - September
1 post2018 - April
1 post2018 - January
1 post2017 - June
1 post2016 - December
1 post2016 - August
2 posts2016 - March
1 post2015 - December
1 post2009 - November
2 posts2008 - May
1 post2007 - August
1 post2026 - May
1 postOpen Source Best Practices with LLMs - The Bare Minimum
Open source projects built with LLMs are only ready with simple installation, reliable tests and CI, and documentation focused on the problem. Standardizing releases and deployments makes automation predictable.
2026 - March
1 postMigrating my Home Server with Claude Code | openSUSE MicroOS
I migrated 49 containers from Ubuntu to openSUSE MicroOS with Claude Code’s help, reorganizing Docker, NFS, ROCm, backups, and SELinux. Automation sped things up, but architecture and validation stayed human.
2026 - February
1 postSQLite + Kamal: Rails Deploy Without Drama | Behind The M.Akita Chronicles
How Rails 8 with SQLite and Kamal lets you run a full production app on a $12/month VPS with zero external services.
2025 - September
1 postInstalling Grafana on My Home Server
How I set up Grafana, Prometheus, and cAdvisor with Docker Compose to monitor my Intel NUC home server.
2018 - April
1 post[SmallTips] Running GitLab CI Runner Locally
GitLab CI Runner uses `.gitlab-ci.yml` to reproduce failing server jobs locally through Docker, speeding up diagnosis. The tradeoff is starting over without cache or artifacts.
2018 - January
1 postMy Notes about a Production-grade Ruby on Rails Deployment on Google Cloud Kubernetes Engine
After taking Google Cloud Kubernetes Engine from zero to production, the author gathers notes on Rails, Docker, Cloud SQL, secrets, NGINX Ingress, and rolling updates. The platform works, but requires infrastructure experience.
2017 - June
1 postRails 5.1 Deployment Checklist for Heroku
This checklist prepares a small Rails 5.1 app for production on Heroku, covering PostgreSQL, Procfiles, Webpack, a CDN with CORS, Memcached, and SSL, without claiming to fit every scenario.
2016 - December
1 postElixir Phoenix App deployed into a Load Balanced DigitalOcean setup
The author builds a two-node Phoenix cluster on DigitalOcean with HAProxy and PostgreSQL, and gets distributed WebSockets running with Erlang. It works, but edeliver is slow and login still fails intermittently.
2016 - August
2 postsMoving away from Slack into Rocket.chat. Good Enough.
I chose Rocket.chat to replace Slack and built a production setup with MongoDB in a replica set, multiple Node.js instances, NGINX, SSL, and monitoring. It works, but it’s not for beginners.
Moving to GitLab! Yes, it's worth it!
While migrating hundreds of repositories from GitHub and Bitbucket to GitLab CE, I preserved important history, sped up builds with GitLab Runner, and cut costs by up to 11 times.
2016 - March
1 postIs your Rails app ready for Production?
A Rails production checklist covers deployment, web servers, monitoring, dependencies, databases, queues, CDNs, and access. The final recommendation is to optimize with metrics before rewriting everything.
2015 - December
1 postEx Pusher Lite - Part 2 - First Working Core!
Part 2 turns ExPusherLite into a working Phoenix core with apps, keys and secrets, authentication, channels, broadcasts, and Heroku deployment. Tests were quick, but security remains fragile.
2009 - November
2 postsChatting with Adam Jacob
Adam Jacob and Jesse Robbins explain how Chef began, defend its Ruby-based flexibility, and show how the tool balances automation, scalability, and agility in Web Ops.
Chatting with Luke Kanies
Luke Kanies explains why he created Puppet, how its abstraction of resources and dependencies simplifies heterogeneous configurations, and why ITIL can become a checklist without solving problems.
2008 - May
1 postChatting with Hongli Lai and Ninh Bui (Phusion)
Hongli Lai and Ninh Bui explain how Phusion Passenger simplifies Rails deployment, shuts down idle processes, and matches or outperforms Mongrel clusters. Ruby Enterprise Edition saved 33% of memory on average.
2007 - August
1 postChatting with Jamis Buck
Jamis Buck tells how he moved from Java to Ruby, joined 37signals, and created Capistrano to run deployments across multiple servers. He also explains the name change and the limit for Windows destinations.