More translations
All checks were successful
Trigger build for deploying blog updates / build (push) Successful in 0s

This commit is contained in:
2026-06-10 00:54:57 +02:00
parent 55678ff2fd
commit 9bb823b87e
10 changed files with 231 additions and 27 deletions

View File

@@ -1,6 +1,6 @@
---
title: First Post
description: Testint test testong
description: Testing testing
date: 2026-03-18
slug: test
---
@@ -12,28 +12,25 @@ I think it works for now!
## So...
Welcome to aranroig.com! I finally build it, it feels like a very important milestone. At the moment I'm able to write simple markdown documents
and displaying it as plain html with custom classes and a lot of other things. So, publishing new blog posts will be very easy, its just doing
a commit and pushing to master, and all the CI/CD pipeline will take about the rest.
Welcome to aranroig.com! I've finally built it, and it feels like a very important milestone. At the moment, I'm able to write simple Markdown documents and display them as plain HTML with custom classes and a lot of other features. So, publishing new blog posts will be very easy; it's just a matter of making a commit and pushing to master, and the entire CI/CD pipeline will take care of the rest.
The architecture of the website is very simple but also a bit complicated.
First of all, I am tunneling the connexions from a VPS to my homelab, since I am still trapped behind a CG-NAT. All traffic is redirected from
the VPS to my home using WireGuard. It is really easy to configure, it was just two different config files.
First of all, I am tunneling connections from a VPS to my homelab, since I am still trapped behind a CG-NAT. All traffic is redirected from the VPS to my home using WireGuard. It is really easy to configure; it was just two different config files.
At my homelab I have installed a [Proxmox Virtual Environment](https://www.proxmox.com). Proxmox lets you use a computer as a server, where you can create VMs and also LXC containers, which are separate instances of containers but they all share the same kernel of the host machine. Docker is built above LXC containers. These containers are also very efficient, since they share the same kernel. But if one kernel panics, so does the other ones, since they share the same kernel.
At my homelab, I have installed a [Proxmox Virtual Environment](https://www.proxmox.com). Proxmox lets you use a computer as a server, where you can create VMs and also LXC containers. These are separate container instances that all share the same kernel as the host machine. Docker is built on top of LXC containers. These containers are also very efficient since they share the same kernel. However, if one kernel panics, so do the others, since they share the same kernel.
So, the incomming connexions go to a LXC container which has nginx and within a internal network these requests are distributed to different places, such as this website, my personal [gitea server](https://git.aranroig.com), an instance of [Navidrome](https://www.navidrome.org/), and another one of [Immich](https://immich.app/), just to name a few.
So, incoming connections go to an LXC container running nginx, and within an internal network, these requests are distributed to different services, such as this website, my personal [Gitea server](https://git.aranroig.com), an instance of [Navidrome](https://www.navidrome.org/), and another one of [Immich](https://immich.app/), just to name a few.
I have also another LXC container that has a gitea worker in it. I have configured my repository so that every push in master it builds a docker image of the frontend and the backend of this website and then publishes them inside the gitea server, and another LXC container where everything is hosted pulls back the images and deploys them. This setup could be even get more crazier but for me this is already enough.
I also have another LXC container that runs a Gitea worker. I've configured my repository so that every push to master builds a Docker image of the frontend and backend of this website, then publishes them inside the Gitea server. Another LXC container, where everything is hosted, pulls back the images and deploys them. This setup could get even crazier, but for me, this is already enough.
I have no idea of what I can place here let me put an image
I have no idea what else to put here, so let me just add an image:
![Image](/blogs/pukeko.png)
Oh yeah of course it works!
Oh yeah, of course it works!
> Identated
> Indented
- Some
- List
@@ -47,4 +44,4 @@ Checklists are displayed wrong!
## Anyways
I should style in the future this a little bit. For now this just works as expected.
I should style this up a little bit in the future. For now, it just works as expected.

View File

@@ -5,19 +5,14 @@ date: 2026-04-29
slug: upgrade
---
Today I have upgraded the cluster from 16GB of ram to 48GB of ram. And also I added a new 8TB HDD drive.
Today, I upgraded the cluster's RAM from 16GB to 48GB. I also added a new 8TB HDD.
Now I have an small problem all the media and random files inside the original HDD were inside a proxmox virtual qcow disk which
is just a disk file that all the other containers from the proxmox instance had mounted. Now with this new disk
I made a ZFS filesystem.
I do have a small issue: all the media and miscellaneous files on the original HDD were stored in a Proxmox virtual .qcow disk image, which is essentially just a file that all the other containers on the Proxmox host were mounted to. With this new disk, I created a ZFS filesystem.
The ZFS filesystem is great for servers. It has a lot of features like having on-write compression and decompression
and also it is very easy to add new disks to a "zpool" of disks.
ZFS is excellent for servers. It offers many useful features, such as on-the-fly compression and decompression, and it's very straightforward to add new drives to a ZFS pool.
Right now I'm copying all the files inside the old virtual .qcow disk file onto the new ZFS filesystem that I have made
on the new disk. When this finishes, I will wipe the old disk and add it to the ZFS pool.
Currently, I'm copying all the files from the old virtual .qcow disk image onto the new ZFS filesystem I created on the new drive. Once that's done, I'll wipe the old disk and add it to the ZFS pool.
For now it will be on RAID0. These disks are relatively new, so maybe in the far future (in about 3 years) I will consider
maybe making a RAID5. I think it should be easy, right?
For now, it'll be configured as RAID0. Since these disks are relatively new, I might consider switching to RAID5 in the distant future (around 3 years from now). I think that should be straightforward, right?
Anyways see ya
Anyway, see ya!