Sergio Talens-Oliag Technical Blog

With all the recent buzz about DeepSeek and its capabilities, I’ve decided to give it a try using Ollama and Open WebUI on my work laptop which has an NVIDIA GPU: $ lspci | grep NVIDIA 0000:01:00.0 3D controller: NVIDIA Corporation GA107GLM [RTX A2000 8GB Laptop GPU] (rev a1) For the installation I initially I looked into the approach suggested on this article, but after reviewing it I decided to go for a docker only approach, as it leaves my system clean and updates are easier. Step 0: Install dockerI already had it on my machine, so nothing to do here. Step 1: Install the nvidia-container-toolkit packageAs it is needed to use the NVIDIA GPU with docker I followed the instructions to install the package using apt from the NVIDIA website. Step 2: Run the Open WebUI container bundled with OllamaI could install ollama directly on linux or run it on docker, but I found out that there is a container with Open WebUI bundled with Ollama, so I decided to use it instead. To start the container I’ve executed the following command: docker run -d \ -e OLLAMA_HOST="0.0.0.0:11434" -p 127.0.0.1:11434:11434 \ -p 127.0.0.1:3000:8080 \ -v ollama:/root/.ollama \ -v open-webui:/app/backend/data \ --gpus=all --name open-webui --restart always \ ghcr.io/open-webui/open-webui:ollama...

Although I am a Debian Developer (not very active, BTW) I am using Ubuntu LTS (right now version 24.04.1) on my main machine; it is my work laptop and I was told to keep using Ubuntu on it when it was assigned to me, although I don’t believe it is really necessary or justified (I don’t need support, I don’t provide support to others and I usually test my shell scripts on multiple systems if needed anyway). Initially I kept using Debian Sid on my personal laptop, but I gave it to my oldest son as the one he was using (an old Dell XPS 13) was stolen from him a year ago. I am still using Debian stable on my servers (one at home that also runs LXC containers and another one on an OVH VPS), but I don’t have a Debian Sid machine anymore and while I could reinstall my work machine, I’ve decided I’m going to try to use a system container to run Debian Sid on it. As I want to use a container instead of a VM I’ve narrowed my options to lxc or systemd-nspawn (I have docker and podman installed, but I don’t believe they are good options for running system containers). As I will want to take snapshots of the container filesystem I’ve decided to try incus instead of systemd-nspawn (I already have experience with it and while it works well it has less features than incus). Installing incusAs this is a personal system where I want to try things, instead of using the packages included with Ubuntu I’ve decided to install the ones from the zabbly incus stable repository. To do it I’ve executed the following as root: # Get the zabbly repository GPG key curl -fsSL https://pkgs.zabbly.com/key.asc -o /etc/apt/keyrings/zabbly.asc # Create the zabbly-incus-stable.sources file sh -c 'cat <<EOF > /etc/apt/sources.list.d/zabbly-incus-stable.sources Enabled: yes Types: deb URIs: https://pkgs.zabbly.com/incus/stable Suites: $(. /etc/os-release && echo ${VERSION_CODENAME}) Components: main Architectures: $(dpkg --print-architecture) Signed-By: /etc/apt/keyrings/zabbly.asc EOF'...

For a long time I’ve been using the Terminator terminal emulator on Linux machines, but last week I read a LWN article about a new emulator called Ghostty that looked interesting and I decided to give it a try. The author sells it as a fast, feature-rich and cross-platform terminal emulator that follows the zero configuration philosophy. Installation and configurationI installed the debian package for Ubuntu 24.04 from the ghostty-ubuntu project and started playing with it. The first thing I noticed is that the zero configuration part is true; I was able to use the terminal without a configuration file, although I created one to change the theme and the font size, but other than that it worked OK for me; my $HOME/.config/ghostty/config file is as simple as: font-size=14 theme=/usr/share/ghostty/themes/iTerm2 Solarized Light...

I’ve always been a fan of template engines that work with text files, mainly to work with static site generators, but also to generate code, configuration files, and other text-based files. For my own web projects I used to go with Jinja2, as all my projects were written in Python, while for static web sites I used the template engines included with the tools I was using, i.e. Liquid with Jekyll and Go Templates (based on the text/template and the html/template go packages) for Hugo. When I needed to generate code snippets or configuration files from shell scripts I used to go with sed and/or envsubst, but lately things got complicated and I started to use a command line application called tmpl that uses the Go Template Language with functions from the Sprig library. tmplI’ve been using my fork of the tmpl program to process templates on CI/CD pipelines (gitlab-ci) to generate configuration files and code snippets because it uses the same syntax used by helm (easier to use by other DevOps already familiar with the format) and the binary is small and can be easily included into the docker images used by the pipeline jobs. One interesting feature of the tmpl tool is that it can read values from command line arguments and from multiple files in different formats (YAML, JSON, TOML, etc) and merge them into a single object that can be used to render the templates. There are alternatives to the tmpl tool and I’ve looked at them (i.e. simple ones like go-template-cli or complex ones like gomplate), but I haven’t found one that fits my needs. For my next project I plan to evaluate a move to a different tool or template format, as tmpl is not being actively maintained (as I said, I’m using my own fork) and it is not included on existing GNU/Linux distributions (I packaged it for Debian and Alpine, but I don’t want to maintain something like that without an active community and I’m not interested in being the upstream myself, as I’m trying to move to Rust instead of Go as the compiled programming language for my projects). Mini JinjaLooking for alternate tools to process templates on the command line I found the minijinja rust crate, a minimal implementation of the Jinja2 template engine that also includes a small command line utility (minijinja-cli) and I believe I’ll give it a try on the future for various reasons: I’m already familiar with the Jinja2 syntax and it is widely used on the industry.On my code I can use the original Jinja2 module for Python projects and MiniJinja for Rust programs.The included command line utility is small and easy to use, and the binaries distributed by the project are good enough to add them to the docker container images used by CI/CD pipelines.As I want to move to Rust I can try to add functionalities to the existing command line client or create my own version of it if they are needed (don’t think so, but who knows)....

On recent weeks I’ve had some time to scratch my own itch on matters related to tools I use daily on my computer, namely the desktop / window manager and my text editor of choice. This post is a summary of what I tried, how it worked out and my short and medium-term plans related to them. Desktop / WMOn the desktop / window manager front I’ve been using Cinnamon on Debian and Ubuntu systems since Gnome 3 was published (I never liked version 3, so I decided to move to something similar to Gnome 2, including the keyboard shortcuts). In fact I’ve never been a fan of Desktop environments, before Gnome I used OpenBox and IceWM because they where a lot faster than desktop systems on my hardware at the time and I was using them only to place one or two windows on multiple workspaces using mainly the keyboard for my interactions (well, except for the web browsers and the image manipulation programs). Although I was comfortable using Cinnamon, some years ago I tried to move to i3, a tilling window manager for X11 that looked like a good choice for me, but I didn’t have much time to play with it and never used it enough to make me productive with it (I didn’t prepare a complete configuration nor had enough time to learn the new shortcuts, so I went back to Cinnamon and never tried again). Anyway, some weeks ago I updated my work machine OS (it was using Ubuntu 22.04 LTS and I updated it to the 24.04 LTS version) and the Cinnamon systray applet stopped working as it used to do (in fact I still have to restart Cinnamon after starting a session to make it work) and, as I had some time, I decided to try a tilling window manager again, but now I decided to go for SwayWM, as it uses Wayland instead of X11. Sway configurationOn my ~/.config/sway/config I tuned some things: Set fuzzel as the application launcher.Installed manually the shikane application and created a configuration to be executed always when sway is started / reloaded (I adjusted my configuration with wdisplays and used shikanectl to save it).Added support for running the xdg-desktop-portal-wlr service.Enabled the swayidle command to lock the screen after some time of inactivity.Adjusted the keyboard to use the es key mapAdded some keybindings to make my life easier, including the use of grimm and swappy to take screenshotsConfigured waybar as the environment bar.Added a shell script to start applications when sway is started (it uses swaymsg to execute background commands and the i3toolwait script to wait for the #!/bin/sh # VARIABLES CHROMIUM_LOCAL_STATE="$HOME/.config/google-chrome/Local State" I3_TOOLWAIT="$HOME/.config/sway/scripts/i3-toolwait" # Functions chromium_profile_dir() { jq -r ".profile.info_cache|to_entries|map({(.value.name): .key})|add|.\"$1\" // \"\"" "$CHROMIUM_LOCAL_STATE" } # MAIN IGZ_PROFILE_DIR="$(chromium_profile_dir "sergio.talens@intelygenz.com")" OURO_PROFILE_DIR="$(chromium_profile_dir "sergio.talens@nxr.global")" PERSONAL_PROFILE_DIR="$(chromium_profile_dir "stalens@gmail.com")" # Common programs swaymsg "exec nextcloud --background" swaymsg "exec nm-applet" # Run spotify on the first workspace (it is mapped to the laptop screen) swaymsg -q "workspace 1" ${I3_TOOLWAIT} "spotify" # Run tmux on the swaymsg -q "workspace 2" ${I3_TOOLWAIT} -- foot tmux a -dt sto wp_num="3" if [ "$OURO_PROFILE_DIR" ]; then swaymsg -q "workspace $wp_num" ${I3_TOOLWAIT} -m ouro-browser -- google-chrome --profile-directory="$OURO_PROFILE_DIR" wp_num="$((wp_num+1))" fi if [ "$IGZ_PROFILE_DIR" ]; then swaymsg -q "workspace $wp_num" ${I3_TOOLWAIT} -m igz-browser -- google-chrome --profile-directory="$IGZ_PROFILE_DIR" wp_num="$((wp_num+1))" fi if [ "$PERSONAL_PROFILE_DIR" ]; then swaymsg -q "workspace $wp_num" ${I3_TOOLWAIT} -m personal-browser -- google-chrome --profile-directory="$PERSONAL_PROFILE_DIR" wp_num="$((wp_num+1))" fi # Open the browser without setting the profile directory if none was found if [ "$wp_num" = "3" ]; then swaymsg -q "workspace $wp_num" ${I3_TOOLWAIT} google-chrome wp_num="$((wp_num+1))" fi swaymsg -q "workspace $wp_num" ${I3_TOOLWAIT} evolution wp_num="$((wp_num+1))" swaymsg -q "workspace $wp_num" ${I3_TOOLWAIT} slack wp_num="$((wp_num+1))" # Open a private browser and a console in the last workspace swaymsg -q "workspace $wp_num" ${I3_TOOLWAIT} -- google-chrome --incognito ${I3_TOOLWAIT} foot # Go back to the second workspace for keepassxc swaymsg "workspace 2" ${I3_TOOLWAIT} keepassxc...