Urgent.News

What's breaking now, across thousands of outlets.

Tech

Writing a Guix service from scratch, as a beginner

This article serves as a guide for beginners to create a custom service for the Caddy reverse proxy in Guix. The author, who is migrating from NixOS to Guix, has already set up a mail server using exim and is now working on migrating their VPS to Guix. They found that exim is the only MTA available in the official Guix channel and decided to explore custom services to better understand and utilize the Guix system.

For the proxy server, the author prefers using Caddy but it's not available as a service on Guix. This motivated the author to write this article as a step-by-step guide for newcomers to create their own service. The article focuses on building a simple and bare-bones service that does the job without being elegant or ready for distribution.

To accomplish this, the author defines a service type as a blueprint for the service, similar to a class in object-oriented programming. They mention that the service type requires a name and a list of extensions, with optional fields such as compose, extend, default-value, description, and location. The article then proceeds to create the simplest service possible, which does nothing, in order to familiarize itself with the concept of Guix services.

Written by urgent.news from Lobsters's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at aloysberger.com →

More in Tech

Week 13 of #100DaysOfCode: Completing Spring Microservices and Starting JUnit 5

Introduction This week was another challenging but rewarding part of my #100DaysOfCode journey. I started the week by learning about Docker and containerized Microservices , exploring how containers…

  • Completed Building Spring with Microservices course with 85% score
  • Learned Docker basics, containerized Microservices, and JUnit 5
  • Explored autoscaling, container orchestration, and Microservices challenges

My Extraction Score Was 0.08 and the Model Was Innocent: Rebuilding the Ruler

Update — v0.3.1 released. CauterRule is now live on GitHub and PyPI . It turns repeated agent failures into permanent standing rules — extract, replay-test, promote.

  • CauterRule tool evaluates cloud models across 40 corpora and 4,742 trajectory-runs
  • Extraction-accuracy metric initially 0.08, later adjusted to focus on trigger portion
  • Adjusted metric shows agreement scores from 0.74 to 0.93, accurate performance representation

More from Sunday 13 September →