I do it wrong so you can learn a new way to do it right.
Systems and Programming
This post is going to be as brief as I can. I hope it will be at least… There is a big change coming to my Docker images this week and it will disrupt some and help others, especially those who do ops with my images, or who have used my images. It’s not that ops was hard with my images, well… actually they were designed to try and make ops easier.
So, lets assume you’ve your own server and you’ve all the extra capacity in the world to host your site (not that you’ll really need much.) You’ve got Docker installed, you’ve got Docker-Compose installed and now you’re ready to go and start deploying your site with Docker.
Today at 10:30AM CST I was alerted to a possible ongoing situation with https://talk.jekyllrb.com, an outage that was quickly escalating out of control because of resources being consumed, CPU steal time, mixed with SideKiq, mixed with a volatile race in kswapd0.