Running Gitlab simultaneously on two domains
The problem So you have a Gitlab instance running and you are thinking about migrating it to a new domain, without breaking existing functionality (such as the registry, links in a broad, hard-to-batch-edit documentation etc.) What you will probably find after a bit of searching is this six year old feature request on the Gitlab issue tracker, from which you will quickly understand that this is not supported. So, what to do?...
Tailscale - the magic mesh overlay network
Preface For some time, I had a pretty nice networking setup at home. The setup consisted of: pfSense on an older PC Engines APU2 board A semi-managed (VLAN-capable) Netgear GS108Ev3 8-port switch 2x TP-Link EAP 245 v3 access points, with the Omada controller software running on my NAS in Docker To reach home from anywhere, I used Wireguard through a jump-box running in Hetzner Cloud. This would let me get through any kind of (in my case, CG) NAT, as my pfSense router would connect to the VPS, and all of my devices would then also connect to that same VPS....
Open-Source Summit EU 2022 - what have you missed?
Introduction This year, I had the chance to visit the Open-Source Summit EU in Dublin and I decided to write a short post highlighting the things that were most interesting to me and hopefully to you, too. So, what did you miss? Tuesday (13/09/2022) Confidential Computing and Privacy-Enhancing Technologies - Mike Bursell, Profian We sort of know how to isolate workloads running on a single host from one another, using VMs, containers, whatever…...
Dealing with a Google Kubernetes Engine Cluster Outage
This post was originally posted to the heurekadevs.cz blog. Intro As the infrastructure team at Heureka, we are responsible for providing certain infrastructure components as a service (Infrastructure as a Service, or IaaS) to our developers. One such service is Kubernetes, for which we use the Google Kubernetes Engine, which serves as the core for many of our services, both in our “legacy” on-prem and in the cloud. Everything was running smoothly for a few months, until one day (luckily not while in full production yet); our pre-production (staging) cluster went down....
Safe rm -rf *
Every once in a while, you have to delete all files and folders recursively. Maybe you are doing this to clean up and you put a lot of time and thought to each command. Maybe you just need a quick clean up of your starting project to start over while learning. But maybe you are traversing directories back and forth, and need to wipe a lot of them at the same time, quickly....