Setting up Git to work over SSH gave me a bit of a headache. I wanted to host my own private Git service in my homelab, so I chose Forgejo. The installation itself was easy, but getting SSH to work properly required a bit more digging.

My Homelab Environment

• Ubuntu LXC on Proxmox
• Forgejo installed in the Linux container
• Pi-hole for DNS
• Nginx Proxy Manager for reverse proxy

Generating a New SSH Key Pair for Forgejo

First, I recommend creating a new pair of SSH keys dedicated to Forgejo. On macOS, you can generate them like this:

ssh-keygen -t ed25519 -a 100 -C "forgejo user email"
pbcopy < ~/.ssh/forgejo.pub // on macOS — or just copy the file content manually

Once you have the .pub file, go to Forgejo → Settings → SSH Keys, click Add, and paste the key.

Why SSH Was Still Asking for a Password

At this point, when I tried testing the connection with:

ssh -T git@git.filut.ovh

SSH still asked me for a password.
I initially tried modifying my ~/.ssh/config by adding a new Host section with all the required options, hoping it would help SSH pick the correct identity. Unfortunately, no luck.
After some research, I discovered the real issue: when connecting over the domain name while using Nginx Proxy Manager, SSH was trying to connect to the NPM host instead of my Forgejo container.

The Solution: Bypassing the Proxy for SSH Access

I found a blog post by William Killerud where he described bypassing Cloudflare for SSH. Turns out, using a similar approach solved my problem as well.
In the end my .ssh/config file looks like this:

Host git.filut.ovh
  HostName 192.168.0.86
  User git
  IdentityFile ~/.ssh/forgejo
  AddKeysToAgent yes
  UseKeychain yes

And here's the proof that everything is finally working 🎉

Until next time!

It's been ages since I last updated my LXCs, and one day I suddenly thought — today is the day.
I started with the most obvious choice: the entry point to all other services — Nginx Proxy Manager.

Running the Update

As usual, I ran:

apt update
apt upgrade

After a few minutes the upgrade finished, but I quickly realized that none of my self‑hosted services were reachable. That could only mean one thing — the reverse proxy was down.

Checking Docker

A quick look at docker ps showed that no containers were running.
I navigated to the directory containing my entire docker-compose setup and ran:

docker compose up

…only to see this friendly error:

Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: open sysctl net.ipv4.ip_unprivileged_port_start file: reopen fd 8: permission denied

What Caused the Problem?

Fortunately, the solution turned out to be simple — no need to roll back Docker or modify configs.

The issue was an outdated Proxmox host.

Fixing the Host

I logged into my PVE host and performed:

apt update && apt upgrade

In the end, my Proxmox instance jumped from v8.4.2 to v8.4.16, and just like that, Nginx Proxy Manager was running again.

In this article I will show how I set up a custom NAS server out of some old hardware and OpenMediaVault. This will be my base storage solution and I plan to use it with media player and as a general backup solution.

Hardware

Some time ago I bought used HP G3 desktop for 20 euros. The specs are quite low e.g. an old Intel i5, 8 gigs of RAM, nothing fancy. I equipped it with a set of WD Red Plus 4TB and 256Gb Nvme drive.

OpenMediaVault 7 Installation

Download the .iso file from official site, and burn it onto a USB stick. For macOs I would recommend using Etcher. During the installation make sure that the device is connected to the network. DHCP will set an IP address for the machine. We will change it later to make it static. Personally, I left each configuration as default, however feel free to customize as needed (language, timezone etc.).

After installation, the system will reboot. Make sure to remove the USB stick. After restart the IP address will be shown . In my case it was eno1: 192.168.0.190. When I entered the IP address in the browser, I was greeted with the login page. For first login, use admin for login and openmediavault as password.

Although I downloaded the latest version of OMV first thing I usually do is to update the entire system. Navigate to System -> Update management and update the system.

Changing OMV IP address

There is always a risk that DHCP given IP may change in the future. OpenMediaVault will be my primary data storage solution and many devices will be connected to it. That's why I want to have static IP here. We can easily modify its value by going to Network -> Interfaces menu, and change the address.

I gave the following configuration:

For DNS I used my PiHole instance, which is at 192.168.0.80. If you don't have your own DNS server, you can use for example Google which is 8.8.8.8. I don't use IPv6, so I left it disabled. And after applying changes, I entered the new IP address, and OMV worked as expected.

Create a BTRFS file system OpenMediaVault

I will focus on BTFRS file system type as for my use case and honestly majority of users it will be the best and easiest solution to start with. BTRFS characteristics:

• Provide snapshots
• Raid support on file system level

I started with drives not plugged in, so now it is good time to plug them in. New drives should be visible in Storage -> Disks pane.

Now navigate to Storage -> File Systems. Hit the + icon, and then BTRFS. On next screen choose RAID1, and select your disks.

Why use RAID1?

Having two disks RAID1 makes a live copy to second disk. It is called mirror. However it should not be considered as a backup. More like availability case when one drive fails then file system would be still accessible.

On the next screen I left the default values. Don't forget to commit the changes.

In RAID1, we should have maximum capacity the same as the smallest drive. I used 4TB drives, so I expect capacity around that number. OMV showed sum of two disks, but as it was reported, this is a bug that will correct itself once data is transferred.

Create SMB share in OpenMediaVault

A few thing we have to set to create a SMB share. Shares Go to Storage -> Shared Folder and hit + icon. Give a name to your share. I gave mine movies as I plan to later integrate it with Jellyfin (self-hosted app for storing and watching videos).

We can't use the share straight away. We need to enable the SMB service. To do so, navigate to Services -> SMB/CIFS and go to Setting. Tick Enabled, and save the changes.

Next go to Shares and create new share. I left default options on.

One last change is to grant user permission to the share.

Testing OpenMedaVault SMB share

Now that we have all set up properly, let's test the solution. I'm on MacOs so I will be using Finder to connect to shared folder. On Windows or Linux it shouldn't be hard to connect to it.

And yes, after providing user and password, our share is accessible.

Summary

In this article, I shared how I built a custom NAS server using an old HP desktop and OpenMediaVault 7. I walked through the hardware setup, installing OMV, configuring a static IP, and creating a BTRFS file system with RAID1 for redundancy. I also showed how to set up an SMB share so the NAS can be accessed from other devices on the network. It’s a simple, budget-friendly solution that works great for media storage and backups.

I moved to a new apartment and I got a new router. The TPLink C64 default DHCP settings are different from my previous one so my entire configuration was corrupted and I lost connection to the Proxmox WebUI. I will describe the steps I took to make it work again.

Goals

• Restore Proxmox Web GUI

Checking TPLink's DHCP IP ranges

To start off, let's first check what is the DHCP ip range. I want my Proxmox to be outside that range. Otherwise, I might risk the IP address changing in the feature. On the TPLink C64 you can view that information under Advanced -> Network -> DHCP server.

In my case, the default range is between 192.168.0.100 and 192.168.0.249.

How to configure IP address for Proxmox

Let's change the configuration. Since I could not access the Proxmox WebUI, I had to connect a screen to Proxmox host with an HDMI cable.

You should see the login screen. Log in as root.
Next, we need to edit interfaces file. Run nano /etc/network/interfaces
Below you can find my configuration. I set my address to 192.168.0.90/24.

auto lo
iface lo inet loopback

iface enp0s31f6 inet manual

auto vmbr0
iface vmbr0 inet static
        address 192.168.0.90/24 # change to desired ip
        gateway 192.168.0.1     # change gateway ip
        bridge-ports enp0s31f6
        bridge-stp off
        bridge-fd 0


source /etc/network/interfaces.d/*

Once configuration is changed, restart the networking service. To do that, enter:

systemctl restart networking

Now it is time to verify whether the changes actually worked. Enter the IP address in your browser:
https://192.168.0.90:8006

Summary

In this short tutorial you learned how to regain access to Proxmox WebUI after changing network environment. I covered how to check router's DHCP IP range, configure static IP address in Proxmox, and restarting the network service.

The heating season in Poland is ongoing. Automating room temperature can significantly reduce costs. Here, I will present a simple setup that utilizes just two devices.

What will You learn?

• Create a new Home Assistant automation
• Execute a specific action based on a trigger
• Monitor automation executions

Related posts

• How to install Home Assistant
• How to set up Zigbee2MQTT
• Avatto TRV6 installation and review

What devices to use?

I began building my home automation with basic devices, including the Aqara temperature sensor and the Avatto TRV6 thermostat. I recently purchased both on AliExpress at a fair price. Please check out the linked articles if you would like to learn more about them and their installation.

Setting up heat automation

Open Home Assistant homepage and navigate to Settings -> Automation.

Since this is a simple automation, I will create two procedures: one for turning on the heating and another for switching it off.

Let’s start with the first one. Click Create automation button, and then Create New Automation.

Add a trigger. In my case, it’s Aqara sensor, placed in the living area.

When the temperature stays below 21 degrees Celsius for at least 10 minutes, the automation will be triggered.

Next up, you can add an optional condition for example time is between. Personally I don’t want to heat the living room in the night, because I’m in bedroom. So here I’ve added another condition if current time is between 6am and 10pm.

Last step for this automation is to set what actually needs to be done, so turning on the heating. I chose the correct thermostat and set its mode to `heat`.

That’s it for now.  Click Save and you will be asked to give that automation a name.

If you prefer YAML code, I’m sharing it below:

alias: Living Room Radiator Daily
description: Heat when below 21degrees between 6 and 22
trigger:
  - type: temperature
    platform: device
    device_id: 9a40ade37c2b29a15e93c3a687521361
    entity_id: 7d8b003e895fcb140feb0535e63daf7b
    domain: sensor
    below: 21
    for:
      hours: 0
      minutes: 10
      seconds: 0
condition:
  - condition: time
    after: "06:00:00"
    before: "22:00:00"
action:
  - device_id: dbd3dc4af8c1a68b75a9b23532abdebf
    domain: climate
    entity_id: 3fa5058913a0b77ee1d623acedd43848
    type: set_hvac_mode
    hvac_mode: heat
mode: single

Turning off the heating

In the same manner as previously create a new automation.

I will also check the current room temperature, but this time, the condition is that the temperature must be above 21 degrees Celsius.

And the actual action to be executed:

Same as previously I’m sharing the YAML:

alias: Living Room Radiator Daily
description: Heat when below 21degrees between 6 and 22
trigger:
  - type: temperature
    platform: device
    device_id: 9a40ade37c2b29a15e93c3a687521361
    entity_id: 7d8b003e895fcb140feb0535e63daf7b
    domain: sensor
    below: 21
    for:
      hours: 0
      minutes: 10
      seconds: 0
condition:
  - condition: time
    after: "06:00:00"
    before: "22:00:00"
action:
  - device_id: dbd3dc4af8c1a68b75a9b23532abdebf
    domain: climate
    entity_id: 3fa5058913a0b77ee1d623acedd43848
    type: set_hvac_mode
    hvac_mode: heat
mode: single

And that’s it! Your simple heating management should be up and running. 

Monitoring automation execution

You can check whether your automation is working as expected using two methods.

The first, and in my opinion the most useful, is to view the Logbook. You can access it from main navigation panel on the left hand side.

In the screenshot, you can clearly see exactly when the heating was turned on and when it was turned off.

Another option, especially useful when debugging is to check Traces. Traces are accessible from automation menu. There you can check every step and parameters used.

If you found this article helpful, consider buying me a coffee. This helps me continue writing these articles for you!

Conclusion

In this article I showed you how to set up a simple automation in Home Assistant. With this automation, you will be able to manage the heating in your home. It serves as a baseline, so if you need something more tailored to your needs, please experiment and share your results!

When developing applications or running any sort of service in your home, sooner or later, you will need a relational database. PostgreSQL is one of the best choices available.

What will You learn?

In this tutorial, I will guide you through setting up PostgreSQL on an Ubuntu machine in my homelab.

PostgreSQL Installation on Ubuntu

I assume that you have already installed a fresh copy of Ubuntu, or you have one running. If not, check out Creating LXC Containers tutorial for a quick start (https://maciejfil.com/how-to-create-linux-containers-in-proxmox).

Let’s start by updating the repositories and system (especially if it is a fresh Ubuntu copy).

apt update && apt upgrade

Now we can install PostgreSQL. Run the following command:

apt install postgresql

Verify the installation by checking if the service is running.

systemctl status postgresql

Create new PostgreSQL database and table

As shown in the screenshot, it seems to be working. Let’s try to log into our freshly installed database. During installation, user postgres was created for us. So, let’s use it:

sudo -u postgres psql

At this point, please set up a password. Just type \password postgres and you will be asked to provide one.

Next, let’s list all database schemas in PostgreSQL with the \list command. By default, database postgres is created for us.

Let’s add a new database, call it test_db and create a table Users

CREATE DATABASE test_db;

With below command we will switch context to use only test_db.

\c test_db

And now let’s create a table with some example data.

CREATE TABLE Users ( id SERIAL PRIMARY KEY, username VARCHAR(50) NOT NULL, email VARCHAR(100) UNIQUE NOT NULL );

INSERT INTO Users (username, email) VALUES 
('john_doe', 'john.doe@example.com'),
('jane_smith', 'jane.smith@example.com'),
('mike_jones', 'mike.jones@example.com');

select * from users;

Connecting to PostgreSQL from client

Before we begin, let’s check what port PostgreSQL is exposing. In terminal type netstat -plnt. It is 5453. I was trying to connect to database from my laptop, but it failed.

By default, every new installation of PostgreSQL allows connections only locally — in my case, when I’m logged into the Ubuntu machine. There may be a time when you need to allow an external service to connect to the database. In order to do so, we need to perform couple of steps.

I’m always backing up every config file before I make any changes, so execute the following:

cd /etc/postgresql/16/main
cp postgresql.conf postgresql.confBACUP

Then open file in your favourite text editor. I’m using nano.

nano postgresql.conf

Find the listen_addresses line, uncomment it, and put * in the brackets.

Next up open pg_hba.conf and add the following line:

host all all 0.0.0.0/0 md5

This configuration line basically means that we will be able to connect to any database with any user from any IP address. You can adjust those parameters as needed.

Save all and exit. To make those changes take affect, we need to restart postgresql service.

systemctl restart postgresql
systemctl status postgresql

If everything was done correctly, then service should be active again.

As you can see, I was able to connect from Visual Studio Code and query the Users table.

Conclusion

In this tutorial, I’ve guided you through installing and performing the initial setup of a PostgreSQL database. Try it yourself and let me know if you had any issues installing it.

If you found this article helpful, consider buying me a coffee. This helps me continue writing these articles for you!

References:

https://www.postgresql.org/docs/16/index.html

The heating season in Warsaw has begun. I’ve heard that smart radiator thermostats can work miracles when it comes to lowering heating bills. I’ve spotted a good price for the Avatto TRV6 Tuya Zigbee thermostat and decided to give it a try. My friends who have similar solutions in their homes claim that after the first winter season, the savings they made paid off the entire investment!

What will you learn?

You will learn how to install the thermostat onto a radiator with a Danfoss-type valve. Then, I will guide you through connecting the device to Home Assistant via Zigbee2MQTT, and, at the end, we will set up an simple automation.

Avatto Thermostat TRV6 review

I bought my devices just a few days ago on AliExpress. The delivery took around 2 weeks to Warsaw, so quite fast. 

Each package contains a thermostat, valve adapters, and a user manual. Please note that no batteries are included. 

Overall, the materials it is made from feel a bit cheap, but I’m more interested in its capabilities than its looks. You get what you pay for, and for around 30$, that’s a bargain!

How to install Avatto thermostat?

Before installing it on the radiator, completely remove the white cover (there is a battery sign at the bottom). Install two fresh AA batteries. Personally I used IKEA rechargeable batteries (LADDA). Once installed, it should automatically turn on, and the dot screen should show LA sign. Give it at least 2 minutes to perform initial setup.

In the meantime, remove your old thermostat from the valve. My valve is of Danfoss type, so I had to install an adapter first. Make sure it sits tightly and is screwed in (but be gentle; overtightening may cause the plastic to break). Next, install the thermostat, and again, in need to sit thigh, but be careful when tightening. Now lower the temperature on the thermostat by turning the knob counterclockwise until you see 'OF' on the screen. That’s all for now!

Setup communication with HA

If you’re using Zigbee2MQTT, then the installation process is pretty straightforward. First of all, Permit Join in Home Assistant, and then while the thermostat is in OF position, press and hold circular button for couple of seconds, until the WiFi icon appears. Home Assistant should recognize and pair the device shortly. Check out this link for detailed instructions on pairing a new device. https://maciejfil.com/how-to-set-up-zigbee2mqtt-for-home-assistant/

Create Home Assistant automation

Alright, so I created a really simple automation. Go to Settings, and then Automation. There you can add a new automation. When I wake up and go to the living room, I like when it is a little bit warmer in the morning. So my automation every day at 6.30AM turns on the radiator for 20 minutes, and then it turns off. Below you can find the whole YAML configuration.

alias: Living Room Warmup DAILY
description: Turn on the heating every day at 6.30am
trigger:
  - platform: time_pattern
    hours: "6"
    minutes: "30"
condition: []
action:
  - device_id: dbd3dc4af8c1a68b75a9b23532abdebf
    domain: climate
    entity_id: 3fa5058913a0b77ee1d623acedd43848
    type: set_hvac_mode
    hvac_mode: heat
  - delay:
      hours: 0
      minutes: 20
      seconds: 0
      milliseconds: 0
  - device_id: dbd3dc4af8c1a68b75a9b23532abdebf
    domain: climate
    entity_id: 3fa5058913a0b77ee1d623acedd43848
    type: set_hvac_mode
    hvac_mode: "off"
mode: single

Feel free to experiment with your use cases! For example in my bedroom Home Assistant has a task to maintain 21 degrees between 6am and 10pm.

If you found this article helpful, consider buying me a coffee. This helps me continue writing these articles for you!

Conclusion

In this article, I showed you how to install and configure the thermostat with Home Assistant. With just a few automations you can easily maintain desired temperature in your home and therefore decrease the cost of heating.

In this article, I will be exposing Home Assistant installed in my homelab to the internet. Currently, I can only check the state of sensors installed in my home via WiFi. With Cloudflare Tunnel service, you can access your Home Assistant from anywhere, even if your ISP doesn’t provide a static IP address. And it’s a completely no-cost solution! Well… besides buying a domain.

Prerequisites

1. You need to have your own domain.
2. The domain should be added to Cloudflare (DNS pointing to CF).
3. A running Home Assistant instance (https://maciejfil.com/how-to-install-home-assistant-os-in-proxmox/)

Set up CloudFlare Tunnel for Home Assistant

Let’s start by logging into Cloudflare. On your homepage, navigate to Zero Trust, and then Networks/Tunnels. Click Add a tunnel.

On next page, choose Cloudflared and hit Next.

Choose a name for your tunnel. I’ve chosen HomeAssistant.

On the next page, you will have a chance to select your preferred way to install the Cloudflared connector. To make things easy, I’ve chosen Docker. Copy the entire command line. We will use the token value later in this tutorial (the value after —token).

On the next page, we need to configure the actual address that we want to use. For my setup, I will use ha.maciejfil.com. Under Service Type, choose HTTP, and set the URL to homeassistant.local:8123. Don’t worry about HTTP. That only means communication inside the HA virtual machine will not be encrypted, but from client to the machine itself, it will be fully encrypted. You are safe.

Now, let’s jump to Home Assistant. Go to Settings -> Add-ons -> Add-on Store.

Here, we need to add a new repository (three-dot menu). Paste the following link, click Add, and refresh the page: https://github.com/brenner-tobias/ha-addons.

Now, you should see Cloudflared. Click on it and Install.

Now, in the upper menu, go to Configuration. Click on the little Advanced switch and find Cloudflare Tunnel Token. Paste the token that we saved a few steps earlier.

Click Save. You will be prompted to restart the service, so agree to it.

The last step is to add additional configuration to the configuration.yaml file. If you haven’t installed File Editor, you can do that from the Add-On Store. It’s available by default.

Go to File Editor and from the Directory icon, find and open the configuration.yaml file.

In the file, copy and paste the following lines:

# Cloudflared conf
http:
  use_x_forwarded_for: true
  trusted_proxies: 172.30.33.0/24

Click the Save icon in the top right corner. That’s all. Now, we need to restart Home Assistant so that the new configuration can be applied.

Head over to Developer Tools and click Check Configuration. You should see a green message indicating that everything is okay. Hit the Restart button and confirm. After a few seconds, Home Assistant should be up and running again.

It’s time to test our solution. Type the address into your browser. In my case, it was ha.maciejfil.com. As you can see, I was able to access my Home Assistant instance through a custom address and from outside my local WiFi network!

Conclusion

In this tutorial, I’ve shown you how easily you can securely expose your Home Assistant to the internet. Now, you will be able to access your home data from anywhere in the world. Please remember to use strong passwords and enable 2FA!

If you found this article helpful, consider buying me a coffee. This helps me continue writing these articles for you!

References:
https://github.com/brenner-tobias/addon-cloudflared/wiki/How-tos

Finally, my Home Assistant is up and running. Now it’s time to connect my very first home sensor. Today, I will take you through the process of setting up Zigbee2MQTT and connecting Aqara temperature sensor.

Prerequisites

1. Running Home Assistant instance ( https://maciejfil.com/how-to-install-home-assistant-os-in-proxmox/ )

How to Install Zigbee2MQTT in Home Assistant?

To start with,, we need to install message broker - MQTT.  Go to your Home Assistant home page and open Add-on Store.

Next up, click Mosquito Broker and then Install.

Once we have installed it, a good practice is to have separate HA user that can access message queue (we don’t want to expose our secrets, right?). Go to People and add a new user.  Remember to tick Local access only.

Add Sonoff Zigbee Bridge to HA

Now we need to pair our bridge with Zigbee2MQTT. I will use Sonoff ZBDongle-E. Since my Home Assistant is a VM I need to add new device to it. Go to Proxmox and find the HA virtual machine. In Hardware tab, choose Add device.

To apply that change in the VM we need to restart the machine. You can do it either from Proxmox, or from HA terminal. I choose HA terminal method. Enter the VM and type host restart. After a few seconds, your HA instance should be up again.

Install Aqara temperature sensor.

The device should be visible now. Go to System -> Hardware and click All Hardware.

In most cases USB dongles are visible under ttyUSB0. Expand that and search for devlinks. Copy that value and save it somewhere.

Install Zigbee2MQTT add-on

Head over to Add-ons store menu. To be able to install Zigbee2MQTT we need to first add a custom repository. Click the three dot menu in top right corner then Repositories. Paste the following link https://github.com/zigbee2mqtt/hassio-zigbee2mqtt. Click Add. 

Now you should be able to see new apps in the store.

Click and install Zigbee2MQTT add-on. Make sure to enable all the options.

Configure Zigbee2MQTT

We need to perform some basic configuration to make our solution work. You will need the user credentials and devlink of the USB Bridge from previous steps.

Still being in the add-on menu, click Configuration tab. I prefer to edit entire YAML file and so I recommend to you as well. Click the three dot button and choose Edit in YAML. 

You can paste the configuration from here, but please change user credential and serial port.

data_path: /config/zigbee2mqtt
socat:
  enabled: false
  master: pty,raw,echo=0,link=/tmp/ttyZ2M,mode=777
  slave: tcp-listen:8485,keepalive,nodelay,reuseaddr,keepidle=1,keepintvl=1,keepcnt=5
  options: "-d -d"
  log: false
mqtt:
  server: mqtt://core-mosquitto:1883
  user: mqtt-service
  password: password
serial:
  port: >-
    /dev/serial/by-id/usb-Itead_Sonoff_Zigbee_3.0_USB_Dongle_Plus_V2_421c3f29cc12ef1185d16db8bf9df066-if00-port0
  adapter: ezsp

Leave the rest as default, and save the changes.  We can now start the add-on. In the log tab you should see that Zigbee2MQTT is up and running.

Pair Aqara device with Home Assistant

Only one last step is ahead of us! Go to Zigbee2MQTT in the sidebar. You should see a page containing connected devices. Obviously now it’s empty.

To start the pairing, click Permit join (ALL) option in the top bar. This is a state in which any device can connect to our zigbee network for limited time. At the same time, on your device press the pairing button. In my case - Aqara Temperature Sensor pairing button is on the top and I had to press it for 5 seconds until blue LED appeared. Once paired, your device should be visible in the list.

Now you can proceed with adding new devices, or just disable join. The very last step that I recommend you doing is to name your device with something more friendly. Just click the blue button on the right and rename it accordingly. In my case it is Bedroom.

And that’s it! You should now see the temperature values in your room.

Conclusion

In this tutorial I described the entire process of connecting new Aqara sensor to Home Assistant. Try it now yourself!

If you found this article helpful, consider buying me a coffee. This helps me continue writing these articles for you!

References:
https://github.com/zigbee2mqtt/hassio-zigbee2mqtt#installation

I have always wanted to have some kind of home automation at my place. Now that I have my homelab, it is the perfect time to build one.

In this article, I will show you how to install Home Assistant OS using a nifty little shell script that I came across some time ago. It's incredibly easy and fun!

Prerequisites:

1. Proxmox installed on your machine.
2. At least 40GiB of free HDD storage.

HAOS Installation process

Grab the command below and paste it into the Proxmox terminal.

bash -c "$(wget -qLO - https://github.com/tteck/Proxmox/raw/main/misc/post-pve-install.sh)"

It can't get much easier than this. You will be guided in this step-by-step installation process, but there will be some places that we have to make some input.

The first screen that will appear:

Hit enter.

On the next screen, select Yes, and on the following screen, choose Advanced.

On the version screen, please select the stable version - it is the most reliable.

HINT: to exit selection mode, press `TAB`.

Leave the default Machine Type.

On the next screen, you can choose your hostname. I chose `home-assistant`, but feel free to provide any name.

For CPU cores, my recommendation is 2. Two cores will be sufficient for most use cases.

RAM can be left default. I set mine at 4096MiB.

Agree to the last prompts, and we are done with the configuration.

And this is how the installation process looks like:

Home Assistant Initial Setup

Now that we have our new VM installed, let’s connect to our HA instance. Head to your newly created VM shell, and you should see a screen similar to this:

Take note of the IP address of the VM, then head to your browser and paste that address. Don’t forget to add port. In my case the address would be: `192.168.1.34:8123`. 

Once you open the website, Home Assistant will take a few moments to configure itself, and then you will see the welcome screen. Two options there - either complete new setup, or recreate instance from backup. Since this is a new installation, I'm choosing 'Create My Smart Home`

You will be asked to enter your credentials. Remember to use strong password!

Also, in order to present to you the latest weather forecast and also enable some features that use geolocation, select your home address.

And that’s it. Your Home Assistant is ready!

Conclusion

In this tutorial, I guided you through the entire process of installing Home Assistant OS. Now you are ready to create your own home automation.

If you found this article helpful, consider buying me a coffee. This helps me continue writing these articles for you!

References:
https://www.derekseaman.com/2023/10/home-assistant-proxmox-ve-8-0-quick-start-guide-2.html

When setting up your home automation system, one of the most popular devices to connect all your sensors is the Sonoff ZBDongle-E. It is commonly used with Zigbee2MQTT and Home Assistant.

The first step you should take after purchasing one is to update the firmware, or in other words, flash it. This ensures you have the latest security and stability updates installed.

In this short tutorial I will show you how to update the firmware.

Updating Sonoff ZBDongle-E

In my case, I used a MacBook Air, but you perform the update on any operating system. 

You should use browsers like Chrome or Microsoft Edge.

1. Head over to this Github page https://darkxst.github.io/silabs-firmware-builder/ . You will notice several options. 
2. Scroll down a bit, and you will see Connect button. 
3. After a few seconds, you will need to choose which firmware to install. I suggest selecting the default one - the latest. At the time of writing this article, the latest stable version is EZSP 7.3.1.0. 
4. Proceed with the update process. After a few minutes, you should see a result similar to the screen below:

For Windows users

You may be prompted to manually update device drivers. If so, follow the link provided there and download the latest drivers. Then, go to Device Manager, locate the Dongle, and Update the driver by right clicking on the Dongle. You will have to provide directory where you saved downloaded drivers. After successful update, remove and replug in the dongle. It should now be visible in Chrome/Edge.

Troubleshooting

During my research, quite a lot of users had problems connecting to the Sonoff Dongle-E. Each time they clicked Connect button, they had an error.

To avoid this:

1. Make sure that you are using the latest versions of Chrome or MS Edge.
2. Disable any kind of ad-blockers.
3. Have device drivers always up to date.

There is also another way of flashing the dongle, but I will cover that in later tutorial.

Conclusion

In this tutorial I shown you how to update your Sonoff ZBDongle-E device to the latest version. Now you should be ready to connect it to your home automation solution.

If you found this article helpful, consider buying me a coffee. This helps me continue writing these articles for you!

References:
https://darkxst.github.io/silabs-firmware-builder/

Since my homelab is expanding, I came up with the idea that maybe it is time to start documenting the solutions I am building so that I do not have to keep everything in my head. I’m terrible at drawing, so discovering Diagram as Code was a the perfect match.

What is Diagram as Code?

It’s a tool written in Python, that lets you easily create neat looking diagrams, especially good for prototyping architecture or any other IT-related work.

Why you need it?

The main advantage of that solution is that it allows you to track every little change - as everything is written with simple Python script. Then you just push it to your Git repository, and from that point on, you have the entire history. Besides, it has a large number of icons, so you do not have to look around for a specific SVG file on the internet.

How to install DaC?

Below instructions are made for macOs users, but I’m linking resources below for other operating systems, so that you could still benefit from this article.

First step is to install Python libraries. I’m using homebrew to manage installed packages.

brew install python3

Then we need to install Graphviz, which is a visualization software. What that package does is to produce a final diagram image as a png file.

brew install graphviz

We need to also install diagrams module.

First, create a folder where you will store your diagrams.

Then we need to set up a virtual environment so that we can install the diagrams module.

python3 -m venv ./env
source ./env/bin/activate

then run pip3 install diagrams

I would also suggest initiating a Git repository here to track changes using a simple git init.

You should be now ready to go!

Diagram as Code example

As I mentioned in the beginning, I wanted to document my local network architecture. This is a sample of the code I created.

You can copy and paste it to your file and see the outcome. 

Don’t forget to save your file and execute the script with `python3 diagram.py`.

from diagrams import Diagram, Cluster
from diagrams.generic.network import Router, Switch
from diagrams.onprem.proxmox import ProxmoxVE
from diagrams.onprem.compute import Server
from diagrams.onprem.database import Postgresql
from diagrams.onprem.groupware import Nextcloud
from diagrams.onprem.client import Client
from diagrams.generic.os import Ubuntu

with Diagram("Home network", show=True, direction="TB"):
    router = Router("router")
    switch = Switch("switch")
    tv = Client("tv")
    xbox = Client("xbox")
    
    with Cluster("Proxmox"):
        proxmox = ProxmoxVE("pve-1")
        proxmox - [Server("home-assistant"),
                  Postgresql("postgresql"),
                  Nextcloud("next-cloud"),
                  Ubuntu("ws-2")]

    router << proxmox
    router << switch
    switch << tv
    switch << xbox

After running in terminal python3 diagram.py a png file will be created in your directory. Mine diagram looks like that:

Conclusion

In this article, I briefly described how you can create any sort of architectural diagram, that is easy to maintain and track history. That was just an introduction. Feel free to explore the topic and adjust the options to your needs. You can find detailed documentation below.

References:

References:
https://diagrams.mingrammer.com/

In today's article, I will guide you through setting up a Linux container.

What Is a Linux Container (LXC)? 

In simple terms, each Linux container is an extremely lightweight virtualization solution. Each container acts as a full-fledged operating system - just like in a VM - but with one difference: it is shares the host’s kernel. 

You can run LXCs on every linux distribution. In its early stages, Docker (which I’m sure you are familiar with) utilized Linux containers.

#1 Download the Template

As the first step, let’s jump into Proxmox and head over to local storage. Then click CT Templates. In the upper part click Templates and a list of available templates will appear. Select desired template. For the purpose of this tutorial, I will be using Ubuntu 24.04. Hit Download and wait until it finishes.

#2 Create LXC

Let me now guide you through the process of creating Linux container. In Proxmox, the process is quite simple and can be done entirely through the GUI. In top right corner find Create CT, and click it. Window should pop up.

There, fill in required data: 

• CT ID - I recommend leaving as is, but you can change it,
• hostname - of your choice,
• password, confirm password,
• SSH public key - if you have, paste here contents of the .pub file

Leave Unprivileged container (has limited access to host resources, thus a bit more secure) and Nesting checkbox as is and hit Next.

In the next window choose the template you want to use. As mentioned earlier, I will be using Ubuntu.

Next up, we will set up Disk space. I suggest leaving it as is - 8GB, but feel free to extend as you like. Other parameters can be left as default.

In the next tab, we will be configuring CPU. Number of cores available depends on the machine that you are running. In my case I have 4 cores available, but I don’t want to use them all. Since this LXC is just for the purposes of this article, I’m choosing 1 core, and setting the CPU Limit to 1 as well. CPU limit is important to set, as trust me - you don’t want to let the container to adjust number of cores used by itself.

Memory tab is quite straight forward. I’m not planning to use that LXC heavily, I will set both Memory and Swap to 512MiB.

Next, the Networking tab. Here I’m using DHCP to acquire IPv4 address, but if you can setup it manually if you want so. Other than that, I leave all other parameters as default. 

In `DNS` section I also leave it default as I’m not running any local DNS server.

At the last tab there will be a summary of all our setup. Hit Finish, and wait a bit.

You should notice on the left side that the LXC has been created, but it is not running yet.

Click on the new container and press `Start` to run it.

#3 Verify the solution

While still in your container, navigate to Console.  Login as root. You should be welcomed with a standard message, just like in the screen below.

To learn, what IP your machine has, you can type ip a command.

Now you can easily SSH as root to your new Linux Container.

Conclusion

In this article I showed you how to create Linux Containers in Proxmox GUI. You can see that the process takes only a few minutes. Try it out in your own Proxmox instance!

References:

https://www.redhat.com/en/topics/containers/whats-a-linux-container
https://www.baeldung.com/linux/privileged-container-capabilities
https://www.cyberciti.biz/tips/linux-swap-space.html

I was looking for a solution to back up my MySQL database in a automatic manner. I found that the easiest solution would be to utilize mysqldump which would be run by a cron job. I had a few struggles making that work, so today in this short article, I will guide you on how to set it up in an Ubuntu environment.

#1 install mysqldump

You should install the latest version of mysqldump. To check if you have one, paste into terminal mysqldump --version . If not, install it using the following: apt-get install mysqlclient . That command will install all the necessary packages. Confirm the installation with mysqldump --version command.

#2 perform a trial backup

Before we continue, prepare connection data to your database. We will need database user, password, host (the address) and database name, that you would like to backup.

Copy and paste the command. Make sure to fill in your connection details:

mysqldump -u DB_USERNAME -h DB_HOST -p DB_NAME > db_dump.sql 

When you hit enter, you will be asked to provide password. Enter it, and then backup will start. Depending on how big is your database, the process may take a while.

If succeeded (I bet it went okay!) you should be able to see an sql file in your directory. But now, how can we automate database backup?

#3 create a shell script

You probably noticed, that in previous step we had to manually fill in database password. That complicate things, but there is an easy solution. Create a file called .my.cnf in your home directory, and then open it in your text editor of choice. I'm using nano .

touch .my.cnf
nano .my.cnf

Paste in the following:

[mysqldump]
password=DB_PASSWORD

Save and close the file.

Next up, change the CHMOD of that file to 600 chmod 600 .my.cnf

You can now test if the solution works with the command:

mysqldump -u DB_USERNAME -h DB_HOST DB_NAME > db_dump.sql 

Notice that I've removed the -p parameter.

Okay, now lets create a shell script. Create a file for example touch /usr/local/bin/backup_database.sh . Straight away, make the file executable, so chmod +x /usr/local/bin/backup_database.sh . Open it and paste the following bash commands.

#!/bin/bash

DB_USER="db_user"
DB_HOST="db_host"
DB_NAME="db_name"
# Backup dir - full path e.g
BACKUP_DIR="home/ubuntu/backups/database"
DATE=$(date +%Y%m%d%H%M)

# Creates backup dir if not exists
mkdir -p ${BACKUP_DIR}

# specify output file names with dir
DUMP_FILE="${BACKUP_DIR}/${DB_NAME}_backup_${DATE}.sql"
TAR_FILE="${BACKUP_DIR}/${DB_NAME}_backup_${DATE}.tar.gz"

# execute backup
mysqldump -h ${DB_HOST} -u ${DB_USER} ${DB_NAME} > ${DUMP_FILE}

# checking if success
if [ $? -eq 0 ]; then
  echo "Database backup successful, compressing the file..."

  tar -czf ${TAR_FILE} -C ${BACKUP_DIR} $(basename ${DUMP_FILE})

  # delete non-compressed file
  rm ${DUMP_FILE}

  echo "Backup completed: ${TAR_FILE}"
else
  echo "Backup failed!"
fi

I briefly explained each line in the comments. Now let's execute the script!

. /usr/local/bin/backup-database.sh and you should see that backup was successful.

#4 schedule a cron job

Of course, we don't want to run that script manually. We would prefer to schedule it somewhere for the middle of the night 😄. Here is a quick cron tutorial for you.

Type crontab -e and hit enter - your text editor will open. Add to the very end of file the following:

10 0 * * * /usr/local/bin/backup-database.sh

Save and close the text editor. And that's it! You've just scheduled an automatic job to backup your MySQL database everyday at 0.10am!

Conclusion

In this article, I showed you how to set up a fully automatic MySQL database backup in a Linux environment in four steps. Try it now, and remember to backup your precious data!

It turned out, that while creating a VM template in Proxmox 10G of storage is not enough if you want Ubuntu server and some apps running on Docker. I learned this the hard way 😄.

I found extending disk space in the VM quite challenging, here's how to do it.

#1 extend VM disk space in Proxmox

It can be easily done in Proxmox GUI. Go to your VM -> Options -> Hardware and click on Hard Disk. Now select Disk Action and choose Resize.

A modal will pop up. Here enter a value that you want your disk to be resized with. In my case I want to add 20G of space.

Once applied - we're not done yet. Start your VM and log in to terminal.

#2 extending root space

Log into your Ubuntu VM. When you enter df -h you will notice that nothing really changed.

We need to run couple of commands. Follow along.

Firstly, we need to know which partition we need to extend.
Run lsblk. You will see that ubuntu--vg-ubuntu--lv is attached to sda3, so that is the place that we should extend.

Next up, type in sudo cfdisk. Navigate down to /dev/sda3/ and choose Resize option. You will be asked to provide the new size. By default it will be set to max avaiable size. Then, choose Write option, and Quit.

Okay, change are still not visible. Let's continue with
sudo pvresize /dev/sda3.
Now before we go to the next step, we need to find out what the Logical Volume Name is (LV name). To do that enter sudo lvdisplay and copy the LV name parameter. In my case it is /dev/ubuntu-vg/ubuntu-lv.
Next up enter the follwing command:
lvextend -l +100%FREE /dev/ubuntu-vg/ubuntu-lv
We are nearly there. When we check sudo lvdisplay we can see that LV size is now what we expected. But still, we need to resize the root file system. To do that we need to know which filesystem to resize. Hit df -h and find /dev/mapper/ubuntu--vg-ubuntu--lv.

I promise it's the last step. To resize the filesystem, just run the command: resize2fs /dev/mapper/ubuntu--vg-ubuntu--lv

And that's it! Verify the changes with df -h .

Conclusion

I've showed you how to extend disk space in your Ubuntu hosted on Proxmox with just a few Linux commands. Now when your VM approaches disk limit, you can easily extend the volume.

References:
https://pve.proxmox.com/wiki/Resize_disks
https://www.youtube.com/watch?v=58m0hnt0ij8