Last time we discussed how to set up Lemmy locally, this time we will discuss setting up Lemmy in production mode on a Rasberry Pi with functioning image upload by using Docker. This time we have to deviate more from the official guide as some things don’t seem to work. To follow this guide, you will need a basic understanding of the terminal and a Raspberry Pi 3 or 4 (I have only tested this on the Raspberry Pi 4). If you are on Windows 10 or 11 you can use OpenSSH in PowerShell.

Setting up the Raspberry Pi

To prepare an SD card for the Raspberry Pi, download the Raspberry Pi Imager. Insert the SD card, select the Raspberry Pi OS Lite (64-bit) and make sure you pick the SD card for Storage. You could pick the full version of the OS, but make sure you pick a 64-bit version of Debian Bullseye. Before clicking “Write”, go click on the settings icon and enable ssh. You can also set up a user, hostname, authorization keys and WiFi.

Now insert the card into your Raspberry Pi, connect power and you should be able to ssh to the pi. So, with the default pi user, that would be ssh pi@raspberrypi.

Installing Docker

To install Docker we have to follow the Docker Debian installation guide (The Raspian guide leads to a configuration that won’t be able to find any stable docker installation).

First, we have to install the dependencies for adding the new repository:

sudo apt-get update

sudo apt-get install ca-certificates curl gnupg

Add Docker’s official GPG key:

sudo install -m 0755 -d /etc/apt/keyrings

curl -fsSL https://download.docker.com/linux/debian/gpg | sudo gpg --dearmor -o /etc/apt/keyrings/docker.gpg

sudo chmod a+r /etc/apt/keyrings/docker.gpg

And set up the Docker repository: echo \ "deb [arch="$(dpkg --print-architecture)" signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/debian \ "$(. /etc/os-release && echo "$VERSION_CODENAME")" stable" | \ sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

Now we can install docker and docker-compose:

sudo apt-get update

sudo apt-get install docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin docker-compose

To be able to run docker command without using sudo we have to add our current user to the Docker group:

sudo groupadd docker

`sudo usermod -aG docker $USERz

newgrp docker

Configuring Lemmy

We need to download a few configuration files. The configuration files listed in the guide don’t support Arm64, so I took the files from Lemmy 1.17.3 and modified them, so they pick the ARM version of the docker images. The NGINX configuration does work, but it is included to make the download simpler:

curl https://gist.githubusercontent.com/Fireblade75/95a0dfa7abbedff554eb9109434060cd/raw/5cf6eddbe706dd25b84234ce619f18a4faca854a/docker-compose.yml -o docker-compose.yml

curl https://gist.githubusercontent.com/Fireblade75/95a0dfa7abbedff554eb9109434060cd/raw/5cf6eddbe706dd25b84234ce619f18a4faca854a/lemmy.hjson -o lemmy.hjson

curl https://gist.githubusercontent.com/Fireblade75/95a0dfa7abbedff554eb9109434060cd/raw/5cf6eddbe706dd25b84234ce619f18a4faca854a/nginx.conf -o nginx.conf

If you want to change the default password of the database, make sure that you change it both in the docker-compose file and the lemmy.hjson configuration.

Now we can run docker-compose up, this downloads all the containers and starts the Lemmy server. Check the logs for errors and see if there is anything we still need to solve. When the services are done starting, we can stop the cluster again by pressing control + C.

A problem I had was that the image server did not get the right permissions to the location where it wants to store its files. To solve this, we simply have to run the following command:

sudo chown -R 991:991 volumes/pictrs/

Running Lemmy

When all errors are solved, we can start the cluster in detached mode. Let’s first destroy the containers by using docker-compse down. And after that we can run docker-compose up -d. The containers should start now, but this time docker-compose is running in detached mode, this mode does not block the terminal and lets Docker run in the background.

You now have a working installation of Lemmy on a Raspberry Pi. It listens to port 80, so you should be able to navigate to it from other devices in your network. For example, by going to http://raspberrypi/ . The default user is lemmy and its password is lemmylemmy, this is configured inside the lemmy.hjson file. If you later want to update Lemmy to a newer version, you can just change the version of the Docker images inside the docker-compose file.

Hopefully this helped you understand how to set up Lemmy, if you have any question please ask.

Ik zag dat ze bij Beehaw nu één stijl aan icoontjes aan houden. Dat ziet er erg leuk en herkenbaar uit. Natuurlijk is het ook goed dat je gewoon een community kan oprichten zonder dat je je daar super druk over hoeft te maken.

Ik vond het zelf alleen wel leuk om iets van een gezamenlijke stijl aan te houden voor de icoontjes die ik maakte, daarom koos ik voor de stijl van onze nu al meestgebruikte community [email protected]. Wil je ook net zoals !nieuws en [email protected] deze stijl aanhouden, dan kan je het SVG-bestand van de Tech community gebruiken als basis: https://gist.github.com/Fireblade75/005f4d398eb67c970bbd2e3f5d77b24f

Theo, a former Twitch employee, that now is one of the larger tech streamers on Twitch, made a video where he quickly goes over both react-email and Resend. Resend is a new service hat makes it easy to set up email for your website, and it is very affordable for small projects. It even comes with a free tier.

cross-posted from: https://beehaw.org/post/574562

Here's a laundry list of sort with tons of tools we'd like to see

• Role for approval of applications (to delegate)
• Site mods (to delegate from admins)
• Auto-report posts with certain keywords or domains (for easier time curating without reports)
• Statistics on growth (user, comments, posts, reports)
  • User total
  • MUA
  • User retention
  • Number of comments
  • Number of posts
  • Number of reports open
  • Number of reports resolved
• Sort reports
  • by resolved/open
  • by local/remote
• Different ways to resolved a report
  • Suspend account for a limited amount of time rather than just banning
  • Send warning
• Account mod info
  • Number of 'strikes' (global and local) and reports
  • Moderation notes
  • Change email
  • Change password
  • Change role
• Ability to pin messages in a post
• Admins should be able to purge
• Filter modlog to local
• Better federation tools (applications to communities, limiting)
  • Applications to communities to allow safe spaces to exist (people should not be able to just "walk in" on a safe space - similarly to follow requests in Mastodon in a way)
  • Limiting (Lock our communities down from certain instances but still allow people using our instance to talk to people from those instances)

Obviously considering the moment when this is being made - federation tools are our highest priority.

ChatGPT, Microsoft Bing, Google Bard - al die chatbots die afgelopen jaar zijn verschenen, werken op basis van grote taalmodellen. Het lijkt zo simpel: je stelt een vraag en je krijgt antwoord. Maar onder de motorkap zijn het razend ingewikkelde systemen, die alleen zo goed kunnen werken omdat ze zoveel training hebben gehad. Hoe werkt dat precies? En wat zijn precies de risico's en bezwaren van deze nieuwe technologie?

We all know that Lemmy is part of the Fediverse, but how does it do that. This is done trough federating with both other Lemmy servers, but also by implementing the ActivityStreams protocol sot it can communicate with other applications on the Fediverse.

The linked document describes the protocol and how it should work.

De Nederlandse Kiesraad gaat kijken of de software die bij verkiezingen wordt gebruikt in de toekomst niet meer via een cd-rom hoeft te worden verstuurd. Dat gebeurt nu wel met de testsoftware, maar veel computers hebben geen cd-romlades meer.

If you want to help with the development or just want to test things with your own Lemmy instance, you will have to set up a local instance on your own PC. This is not that hard, but it is not uncommon that you will do something wrong and if you are not, that experienced with the technology that is used, it can be hard to understand the error messages that you receive. That’s why I wrote this blog to help developers to run their own local instance.

So when setting up your local instance, it is a good idea to read the official guide for local development. We will now set up both the API/back-end and the front-end.

The back-end

First, we need the rust toolchain. The easiest way is to just get Rustup by following the installation command you find on this website.

Now before we start checking or building the back-end we need to install all required libraries.

For Debian-based (like Ubuntu) this is:

sudo apt install git cargo libssl-dev pkg-config libpq-dev curl postgresql

For Arch-based this is:

sudo pacman -S git cargo libssl-dev pkg-config libpq-dev curl postgresql

For macOS, you can just install postgresql:

brew install postgresql brew services start postgresql /usr/local/opt/postgres/bin/createuser -s postgres

Now we need to add a db user for Lemmy to the database. Sometimes psql cannot be found, in those cases you can often just switch to the postgresql user with sudo su postgres

psql -c "create user lemmy with password 'password' superuser;" -U postgres psql -c 'create database lemmy with owner lemmy;' -U postgres

You can change the password if you want, in that case remember the password you entered.

Now we have everything we need for the back-end, it is time to download the Lemmy project.

git clone https://github.com/LemmyNet/lemmy.git --recursive git clone https://github.com/LemmyNet/lemmy-ui.git --recursive

Make sure you don’t forget the --recursive flag, it is required to download all the code.

Now we can have a look at the configuration of the back-end. In the “lemmy” project there should be a folder named “config”, in this config file are 2 files, defaults.hjson and config.hjson. If you need to make settings to your server, you can make those in the config.hjson file. You can use this to change the password of the database, for example. The defaults.hjson file should help with finding out how this can be done.

Now we can check if everything works correctly, open a shell in the “lemmy” project (this is the back-end). There should be a Cargo.toml file in this folder.

Here you can now run cargo check to check if everything compiles. This should run fine, and then you can run cargo run. Now you should have a running server.

After making changes, you need to format the code with cargo +nightly fmt --all and run the linter with ./scripts/fix-clippy.sh.

The front-end

To get started with the front-end we need both Node and Yarn. Node is available through brew with “brew install node”, but you can also install it from the Node.js website. For many Linux distributions, it is also possible to use your package manager.

Then we still need Yarn, there are again multiple ways to install Yarn, the recommended way to install yarn is trough corepack, this is explained on the yarn website. You can also install it through brew with “brew install yarn” or simply trough npm with npm install -g yarn. I went for the npm route.

After installing yarn, you can install all node dependencies with yarn install and start the development server with yarn start.

Image uploads

We did not set up an image server, so you won’t be able to upload images. The docker setup does support this, but for general development building the docker containers is too slow.

Windows

I haven’t tried this out on Windows, but you should be able to follow all the Linux steps with Windows subsystem for Linux. You might be able to get it to work natively, but some installation steps will be different.

Google may soon be ordered to break up its lucrative ad business, which amounted to nearly $225 billion in 2022 and represented nearly 80 percent of Google's total revenue.

Met een inkoopprijs van 23 euro per megawattuur (MWh) op 1 juni leek de gascrisis voorbij, maar nu blijkt hoe grillig de gasprijzen zijn. De prijs van gas sloot gisteren op ruim 40 euro per MWh en naderde eerder op de dag zelfs even de 50 euro per MWh. Oorzaken: de stijgende vraag naar gas in China en minder aanbod uit Noorwegen.

De koninklijke trein gaat met pensioen. Na bijna 160 jaar zullen het koningshuis en buitenlandse staatshoofden niet meer met het rijtuig reizen. Het vervoersmiddel is aan het einde van zijn levensduur, meldt de NS. In de toekomst reist de koninklijke familie met reguliere treinen.