#core-infrastructure #meta #software #selfhosting #networks #LLMs #my-world #outside #house #wood #fire #organisation #occupation

selfhosting

selfhosting covers most of my experiments tinkering with other people's hardware & software and homelab type stuff.

Current focus of my homelab is getting it set up again for hosting internet facing content without cloudflare.

If you're reading this... I did it!


All 15 posts tagged selfhosting:

Wireguard

Wireguard is a simple modern VPN protocol.

Compared to previous solutions, Wireguard was notable in being exceptionally small and simple. An elegant satisfying system :)

Linus Torvalds notably described it as a work of art compared to IPsec and OpenVPN. At the time, Wireguard was just 4000 lines of code compared to IPsec and OpenVPN which were hundreds of thousands of lines, on a good day.

As a result, it got into the kernel pretty quickly and is widely supported.

I use it to bootstrap a connection from internal headscale server out to external proxy without having to open ports.

remember: wg show` and e.g. `wg-quick down wg0


Filled a disk, and then kept getting error about acquiring cfs lock when trying to do anything, incl modifying VMs, the storage.cfg files, and running pvesr status

I noticed systemctl status pve-clus...


Open-webui

open-webui, formerly ollama-webui, is a Svelte application that serves as a ChatGPT-like interface for ollama which serves other other local and remot...


quartz is a static site generator. I started using it originally as it was focused on publishing Obsidian Markdown notes, and used Hugo as a backend, which was a...


caddy

I use caddy

Hit an issue where a public reverse proxy caddy config was giving this error:

Error: dial tcp x.x.x.x:xxx: connect: no route to host

That message suggests to me a routing issue, but I c...


twinny

Twinny is a vscode plugin for copilot-like functionality with local LLMs, such as code completion, chat and test generation.. In contrast to most plugins like this (cody, codegpt etc.) they don't see...


Slack Nebula

Slack Nebula is a very satisfying system. It's an overlay network written in go at Slack, and has been used heavily across their servers and infrastructure. Once setup, all devices can have an IP in t...


ik_llama

ik_llama.cpp is a fork of llama.cpp, and llama.cpp was originally what ollama was built on. (Ollama has since rewritten the bits they depended on.)

...


Ollama

Ollama is a tool for running local LLMs. It exposes an interface to download and run different published LLMs from it's library (which seems largely huggingface backed.)

How do I use it?

Curren...


tailscale/headscale

Tailscale is an overlay Mesh network. It relies on a central login server which then brokers wireguard connections behind the scenes. The tailscale also offers DNS, ACLs and a whole bunch of other bit...


SSH port forwarding

SSH port forwarding can be used as a simple hacky way to get traffic from a server on the internet without having to open ports on my home internet connection. It uses ssh as well so it's normally alw...


In 2023, I spent a lot of time playing with LLMs. Over the course of the year I got more confident in the idea that anything worth doing can be done locally. LLMs are so cool and fun to play around wi...


Here's how I want to deploy my stuff:

Provision hosts and hardware with Terraform and clicks

Provisioning of hosts should happen ideally via Terraform on unreliable hardware (e.g. cloud or pi or ...


Docker containers on overlay networks

Sometimes I want to join docker services onto overlay networks. If I want the service to be accessible over an overlay network, there are at least 2 options:

  1. configure the docker host on the overl...


selfhosting

selfhosting covers most of my experiments tinkering with other people's hardware & software and homelab type stuff.

Current focus of my homelab is getting it set up again for hosting internet facing content without cloudflare.

If you're reading this... I did it!