SUSE Observability MCP server by drutigliano

Description

The idea is to implement the SUSE Observability Model Context Protocol (MCP) Server as a specialized, middle-tier API designed to translate the complex, high-cardinality observability data from StackState (topology, metrics, and events) into highly structured, contextually rich, and LLM-ready snippets.

This MCP Server abstract the StackState APIs. Its primary function is to serve as a Tool/Function Calling target for AI agents. When an AI receives an alert or a user query (e.g., "What caused the outage?"), the AI calls an MCP Server endpoint. The server then fetches the relevant operational facts, summarizes them, normalizes technical identifiers (like URNs and raw metric names) into natural language concepts, and returns a concise JSON or YAML payload. This payload is then injected directly into the LLM's prompt, ensuring the final diagnosis or action is grounded in real-time, accurate SUSE Observability data, effectively minimizing hallucinations.

Goals

• Grounding AI Responses: Ensure that all AI diagnoses, root cause analyses, and action recommendations are strictly based on verifiable, real-time data retrieved from the SUSE Observability StackState platform.
• Simplifying Data Access: Abstract the complexity of StackState's native APIs (e.g., Time Travel, 4T Data Model) into simple, semantic functions that can be easily invoked by LLM tool-calling mechanisms.
• Data Normalization: Convert complex, technical identifiers (like component URNs, raw metric names, and proprietary health states) into standardized, natural language terms that an LLM can easily reason over.
• Enabling Automated Remediation: Define clear, action-oriented MCP endpoints (e.g., execute_runbook) that allow the AI agent to initiate automated operational workflows (e.g., restarts, scaling) after a diagnosis, closing the loop on observability.

 Hackweek STEP

• Create a functional MCP endpoint exposing one (or more) tool(s) to answer queries like "What is the health of service X?") by fetching, normalizing, and returning live StackState data in an LLM-ready format.

 Scope

• Implement read-only MCP server that can:
  • Connect to a live SUSE Observability instance and authenticate (with API token)
  • Use tools to fetch data for a specific component URN (e.g., current health state, metrics, possibly topology neighbors, ...).
  • Normalize response fields (e.g., URN to "Service Name," health state DEVIATING to "Unhealthy", raw metrics).
  • Return the data as a structured JSON payload compliant with the MCP specification.

Deliverables

• MCP Server v0.1 A running Golang MCP server with at least one tool.
• A README.md and a test script (e.g., curl commands or a simple notebook) showing how an AI agent would call the endpoint and the resulting JSON payload.

Outcome A functional and testable API endpoint that proves the core concept: translating complex StackState data into a simple, LLM-ready format. This provides the foundation for developing AI-driven diagnostics and automated remediation.

Resources

• https://www.honeycomb.io/blog/its-the-end-of-observability-as-we-know-it-and-i-feel-fine
• https://www.datadoghq.com/blog/datadog-remote-mcp-server
• https://modelcontextprotocol.io/specification/2025-06-18/index
• https://modelcontextprotocol.io/docs/develop/build-server

 Basic implementation

• https://github.com/drutigliano19/suse-observability-mcp-server

---

Results

Successfully developed and delivered a fully functional SUSE Observability MCP Server that bridges language models with SUSE Observability's operational data. This project demonstrates how AI agents can perform intelligent troubleshooting and root cause analysis using structured access to real-time infrastructure data.

Example execution

---

Enable more features in mcp-server-uyuni by j_renner

Description

I would like to contribute to mcp-server-uyuni, the MCP server for Uyuni / Multi-Linux Manager) exposing additional features as tools. There is lots of relevant features to be found throughout the API, for example:

• System operations and infos
• System groups
• Maintenance windows
• Ansible
• Reporting
• ...

At the end of the week I managed to enable basic system group operations:

• List all system groups visible to the user
• Create new system groups
• List systems assigned to a group
• Add and remove systems from groups

Goals

• Set up test environment locally with the MCP server and client + a recent MLM server [DONE]
• Identify features and use cases offering a benefit with limited effort required for enablement [DONE]
• Create a PR to the repo [DONE]

Resources

• MCP Server Uyuni
• MLM API Docs

---

Explore LLM evaluation metrics by thbertoldi

Description

Learn the best practices for evaluating LLM performance with an open-source framework such as DeepEval.

Goals

Curate the knowledge learned during practice and present it to colleagues.

-> Maybe publish a blog post on SUSE's blog?

Resources

https://deepeval.com

https://docs.pactflow.io/docs/bi-directional-contract-testing

---

Try AI training with ROCm and LoRA by bmwiedemann

Description

I want to setup a Radeon RX 9600 XT 16 GB at home with ROCm on Slowroll.

Goals

I want to test how fast AI inference can get with the GPU and if I can use LoRA to re-train an existing free model for some task.

Resources

• https://rocm.docs.amd.com/en/latest/compatibility/compatibility-matrix.html
• https://build.opensuse.org/project/show/science:GPU:ROCm
• https://src.opensuse.org/ROCm/
• https://www.suse.com/c/lora-fine-tuning-llms-for-text-classification/

Results

got inference working with llama.cpp:

export LLAMACPP_ROCM_ARCH=gfx1200
HIPCXX="$(hipconfig -l)/clang" HIP_PATH="$(hipconfig -R)" \
cmake -S . -B build -DGGML_HIP=ON -DAMDGPU_TARGETS=$LLAMACPP_ROCM_ARCH \
-DCMAKE_BUILD_TYPE=Release -DLLAMA_CURL=ON \
-Dhipblas_DIR=/usr/lib64/cmake/hipblaslt/ \
&& cmake --build build --config Release -j8
m=models/gpt-oss-20b-mxfp4.gguf
cd $P/llama.cpp && build/bin/llama-server --model $m --threads 8 --port 8005 --host 0.0.0.0 --device ROCm0 --n-gpu-layers 999

Without the --device option it faulted. Maybe because my APU also appears there?

I updated/fixed various related packages: https://src.opensuse.org/ROCm/rocm-examples/pulls/1 https://src.opensuse.org/ROCm/hipblaslt/pulls/1 SR 1320959

benchmark

I benchmarked inference with llama.cpp + gpt-oss-20b-mxfp4.gguf and ROCm offloading to a Radeon RX 9060 XT 16GB. I varied the number of layers that went to the GPU:

• 0 layers 14.49 tokens/s (8 CPU cores)
• 9 layers 17.79 tokens/s 34% VRAM
• 15 layers 22.39 tokens/s 51% VRAM
• 20 layers 27.49 tokens/s 64% VRAM
• 24 layers 41.18 tokens/s 74% VRAM
• 25+ layers 86.63 tokens/s 75% VRAM (only 200% CPU load)

So there is a significant performance-boost if the whole model fits into the GPU's VRAM.

---

Exploring Modern AI Trends and Kubernetes-Based AI Infrastructure by jluo

Description

Build a solid understanding of the current landscape of Artificial Intelligence and how modern cloud-native technologies—especially Kubernetes—support AI workloads.

Goals

Use Gemini Learning Mode to guide the exploration, surface relevant concepts, and structure the learning journey:

• Gain insight into the latest AI trends, tools, and architectural concepts.
• Understand how Kubernetes and related cloud-native technologies are used in the AI ecosystem (model training, deployment, orchestration, MLOps).

Resources

• Red Hat AI Topic Articles

  • https://www.redhat.com/en/topics/ai

• Kubeflow Documentation

  • https://www.kubeflow.org/docs/

• Q4 2025 CNCF Technology Landscape Radar report:

  • https://www.cncf.io/announcements/2025/11/11/cncf-and-slashdata-report-finds-leading-ai-tools-gaining-adoption-in-cloud-native-ecosystems/
  • https://www.cncf.io/wp-content/uploads/2025/11/cncfreporttechradar_111025a.pdf

• Agent-to-Agent (A2A) Protocol

  • https://developers.googleblog.com/en/a2a-a-new-era-of-agent-interoperability/

---

SUSE Virtualization (Harvester): VM Import UI flow by wombelix

Description

SUSE Virtualization (Harvester) has a vm-import-controller that allows migrating VMs from VMware and OpenStack, but users need to write manifest files and apply them with kubectl to use it. This project is about adding the missing UI pieces to the harvester-ui-extension, making VM Imports accessible without requiring Kubernetes and YAML knowledge.

VMware and OpenStack admins aren't automatically familiar with Kubernetes and YAML. Implementing the UI part for the VM Import feature makes it easier to use and more accessible. The Harvester Enhancement Proposal (HEP) VM Migration controller included a UI flow implementation in its scope. Issue #2274 received multiple comments that an UI integration would be a nice addition, and issue #4663 was created to request the implementation but eventually stalled.

Right now users need to manually create either VmwareSource or OpenstackSource resources, then write VirtualMachineImport manifests with network mappings and all the other configuration options. Users should be able to do that and track import status through the UI without writing YAML.

Work during the Hack Week will be done in this fork in a branch called suse-hack-week-25, making progress publicly visible and open for contributions. When everything works out and the branch is in good shape, it will be submitted as a pull request to harvester-ui-extension to get it included in the next Harvester release.

Testing will focus on VMware since that's what is available in the lab environment (SUSE Virtualization 1.6 single-node cluster, ESXi 8.0 standalone host). Given that this is about UI and surfacing what the vm-import-controller handles, the implementation should work for OpenStack imports as well.

This project is also a personal challenge to learn vue.js and get familiar with Rancher Extensions development, since harvester-ui-extension is built on that framework.

Goals

• Learn Vue.js and Rancher Extensions fundamentals required to finish the project
• Read and learn from other Rancher UI Extensions code, especially understanding the harvester-ui-extension code base
• Understand what the vm-import-controller and its CRDs require, identify ready to use components in the Rancher UI Extension API that can be leveraged
• Implement UI logic for creating and managing VmwareSource / OpenstackSource and VirtualMachineImport resources with all relevant configuration options and credentials
• Implemnt UI elements to display VirtualMachineImport status and errors

Resources

HEP and related discussion

• https://github.com/harvester/harvester/blob/master/enhancements/20220726-vm-migration.md
• https://github.com/harvester/harvester/issues/2274
• https://github.com/harvester/harvester/issues/4663

SUSE Virtualization VM Import Documentation

• https://docs.harvesterhci.io/v1.6/advanced/addons/vmimport

Rancher Extensions Documentation

• https://extensions.rancher.io/extensions/next/home

Rancher UI Plugin Examples

• https://github.com/rancher/ui-plugin-examples

Vue Router Essentials

• https://v3.router.vuejs.org/guide/

Vue Router API

• https://v3.router.vuejs.org/api/#routes

Vuex Documentation

---

Collection and organisation of information about Bulgarian schools by iivanov

Description

To achieve this it will be necessary:

• Collect/download raw data from various government and non-governmental organizations
• Clean up raw data and organise it in some kind database.
• Create tool to make queries easy.
• Or perhaps dump all data into AI and ask questions in natural language.

Goals

By selecting particular school information like this will be provided:

• School scores on national exams.
• School scores from the external evaluations exams.
• School town, municipality and region.
• Employment rate in a town or municipality.
• Average health of the population in the region.

Resources

Some of these are available only in bulgarian.

• https://danybon.com/klasazia
• https://nvoresults.com/index.html
• https://ri.mon.bg/active-institutions
• https://www.nsi.bg/nrnm/ekatte/archive

Results

• Information about all Bulgarian schools with their scores during recent years cleaned and organised into SQL tables
• Information about all Bulgarian villages, cities, municipalities and districts cleaned and organised into SQL tables
• Information about all Bulgarian villages and cities census since beginning of this century cleaned and organised into SQL tables.
• Information about all Bulgarian municipalities about religion, ethnicity cleaned and organised into SQL tables.
• Data successfully loaded to locally running Ollama with help to Vanna.AI
• Seems to be usable.

TODO

• Add more statistical information about municipalities and ....

Code and data

• Github

---

Help Create A Chat Control Resistant Turnkey Chatmail/Deltachat Relay Stack - Rootless Podman Compose, OpenSUSE BCI, Hardened, & SELinux by 3nd5h1771fy

Description

The Mission: Decentralized & Sovereign Messaging

FYI: If you have never heard of "Chatmail", you can visit their site here, but simply put it can be thought of as the underlying protocol/platform decentralized messengers like DeltaChat use for their communications. Do not confuse it with the honeypot looking non-opensource paid for prodect with better seo that directs you to chatmailsecure(dot)com

In an era of increasing centralized surveillance by unaccountable bad actors (aka BigTech), "Chat Control," and the erosion of digital privacy, the need for sovereign communication infrastructure is critical. Chatmail is a pioneering initiative that bridges the gap between classic email and modern instant messaging, offering metadata-minimized, end-to-end encrypted (E2EE) communication that is interoperable and open.

However, unless you are a seasoned sysadmin, the current recommended deployment method of a Chatmail relay is rigid, fragile, difficult to properly secure, and effectively takes over the entire host the "relay" is deployed on.

Why This Matters

A simple, host agnostic, reproducible deployment lowers the entry cost for anyone wanting to run a privacy‑preserving, decentralized messaging relay. In an era of perpetually resurrected chat‑control legislation threats, EU digital‑sovereignty drives, and many dangers of using big‑tech messaging platforms (Apple iMessage, WhatsApp, FB Messenger, Instagram, SMS, Google Messages, etc...) for any type of communication, providing an easy‑to‑use alternative empowers:

• Censorship resistance - No single entity controls the relay; operators can spin up new nodes quickly.
  
• Surveillance mitigation - End‑to‑end OpenPGP encryption ensures relay operators never see plaintext.
  
• Digital sovereignty - Communities can host their own infrastructure under local jurisdiction, aligning with national data‑policy goals.
  

By turning the Chatmail relay into a plug‑and‑play container stack, we enable broader adoption, foster a resilient messaging fabric, and give developers, activists, and hobbyists a concrete tool to defend privacy online.

Goals

As I indicated earlier, this project aims to drastically simplify the deployment of Chatmail relay. By converting this architecture into a portable, containerized stack using Podman and OpenSUSE base container images, we can allow anyone to deploy their own censorship-resistant, privacy-preserving communications node in minutes.

Our goal for Hack Week: package every component into containers built on openSUSE/MicroOS base images, initially orchestrated with a single container-compose.yml (podman-compose compatible). The stack will:

• Run on any host that supports Podman (including optimizations and enhancements for SELinux‑enabled systems).
  
• Allow network decoupling by refactoring configurations to move from file-system constrained Unix sockets to internal TCP networking, allowing containers achieve stricter isolation.
• Utilize Enhanced Security with SELinux by using purpose built utilities such as udica we can quickly generate custom SELinux policies for the container stack, ensuring strict confinement superior to standard/typical Docker deployments.
• Allow the use of bind or remote mounted volumes for shared data (/var/vmail, DKIM keys, TLS certs, etc.).
  
• Replace the local DNS server requirement with a remote DNS‑provider API for DKIM/TXT record publishing.
  

By delivering a turnkey, host agnostic, reproducible deployment, we lower the barrier for individuals and small communities to launch their own chatmail relays, fostering a decentralized, censorship‑resistant messaging ecosystem that can serve DeltaChat users and/or future services adopting this protocol

Resources

• The links included above
• https://chatmail.at/doc/relay/
• https://delta.chat/en/help
• Project repo -> https://codeberg.org/EndShittification/containerized-chatmail-relay

---

Bring to Cockpit + System Roles capabilities from YAST by miguelpc

Bring to Cockpit + System Roles features from YAST

Cockpit and System Roles have been added to SLES 16 There are several capabilities in YAST that are not yet present in Cockpit and System Roles We will follow the principle of "automate first, UI later" being System Roles the automation component and Cockpit the UI one.

Goals

The idea is to implement service configuration in System Roles and then add an UI to manage these in Cockpit. For some capabilities it will be required to have an specific Cockpit Module as they will interact with a reasource already configured.

Resources

A plan on capabilities missing and suggested implementation is available here: https://docs.google.com/spreadsheets/d/1ZhX-Ip9MKJNeKSYV3bSZG4Qc5giuY7XSV0U61Ecu9lo/edit

Linux System Roles:

• https://linux-system-roles.github.io/
• https://build.opensuse.org/package/show/openSUSE:Factory/ansible-linux-system-roles Package on sle16 ansible-linux-system-roles

First meeting Hackweek catchup

• Monday, December 1 · 11:00 – 12:00
• Time zone: Europe/Madrid
• Google Meet link: https://meet.google.com/rrc-kqch-hca

---

Enhance git-sha-verify: A tool to checkout validated git hashes by gpathak

Description

git-sha-verify is a simple shell utility to verify and checkout trusted git commits signed using GPG key. This tool helps ensure that only authorized or validated commit hashes are checked out from a git repository, supporting better code integrity and security within the workflow.

Supports:

• Verifying commit authenticity signed using gpg key
  
• Checking out trusted commits
  

Ideal for teams and projects where the integrity of git history is crucial.

Goals

A minimal python code of the shell script exists as a pull request.

The goal of this hackweek is to:

• DONE: Add more unit tests
  • New and more tests can be added later
    
• Partially DONE: Make the python code modular
  
• DONE: Add code coverage if possible
  

Resources

• Link to GitHub Repository: https://github.com/openSUSE/git-sha-verify

---

Testing and adding GNU/Linux distributions on Uyuni by juliogonzalezgil

Join the Gitter channel! https://gitter.im/uyuni-project/hackweek

Uyuni is a configuration and infrastructure management tool that saves you time and headaches when you have to manage and update tens, hundreds or even thousands of machines. It also manages configuration, can run audits, build image containers, monitor and much more!

Currently there are a few distributions that are completely untested on Uyuni or SUSE Manager (AFAIK) or just not tested since a long time, and could be interesting knowing how hard would be working with them and, if possible, fix whatever is broken.

For newcomers, the easiest distributions are those based on DEB or RPM packages. Distributions with other package formats are doable, but will require adapting the Python and Java code to be able to sync and analyze such packages (and if salt does not support those packages, it will need changes as well). So if you want a distribution with other packages, make sure you are comfortable handling such changes.

No developer experience? No worries! We had non-developers contributors in the past, and we are ready to help as long as you are willing to learn. If you don't want to code at all, you can also help us preparing the documentation after someone else has the initial code ready, or you could also help with testing :-)

The idea is testing Salt (including bootstrapping with bootstrap script) and Salt-ssh clients

To consider that a distribution has basic support, we should cover at least (points 3-6 are to be tested for both salt minions and salt ssh minions):

1. Reposync (this will require using spacewalk-common-channels and adding channels to the .ini file)
2. Onboarding (salt minion from UI, salt minion from bootstrap scritp, and salt-ssh minion) (this will probably require adding OS to the bootstrap repository creator)
3. Package management (install, remove, update...)
4. Patching
5. Applying any basic salt state (including a formula)
6. Salt remote commands
7. Bonus point: Java part for product identification, and monitoring enablement
8. Bonus point: sumaform enablement (https://github.com/uyuni-project/sumaform)
9. Bonus point: Documentation (https://github.com/uyuni-project/uyuni-docs)
10. Bonus point: testsuite enablement (https://github.com/uyuni-project/uyuni/tree/master/testsuite)

If something is breaking: we can try to fix it, but the main idea is research how supported it is right now. Beyond that it's up to each project member how much to hack :-)

• If you don't have knowledge about some of the steps: ask the team
• If you still don't know what to do: switch to another distribution and keep testing.

This card is for EVERYONE, not just developers. Seriously! We had people from other teams helping that were not developers, and added support for Debian and new SUSE Linux Enterprise and openSUSE Leap versions :-)

In progress/done for Hack Week 25

Guide

We started writin a Guide: Adding a new client GNU Linux distribution to Uyuni at https://github.com/uyuni-project/uyuni/wiki/Guide:-Adding-a-new-client-GNU-Linux-distribution-to-Uyuni, to make things easier for everyone, specially those not too familiar wht Uyuni or not technical.

openSUSE Leap 16.0

The distribution will all love!

https://en.opensuse.org/openSUSE:Roadmap#DRAFTScheduleforLeap16.0

Curent Status We started last year, it's complete now for Hack Week 25! :-D

• [W] Reposync (this will require using spacewalk-common-channels and adding channels to the .ini file) NOTE: Done, client tools for SLMicro6 are using as those for SLE16.0/openSUSE Leap 16.0 are not available yet
• [W] Onboarding (salt minion from UI, salt minion from bootstrap scritp, and salt-ssh minion) (this will probably require adding OS to the bootstrap repository creator)
• [W] Package management (install, remove, update...). Works, even reboot requirement detection

---

Rancher Cluster Lifecycle Visualizer by jferraz

Description

Rancher’s v2 provisioning system represents each downstream cluster with several Kubernetes custom resources across multiple API groups, such as clusters.provisioning.cattle.io and clusters.management.cattle.io. Understanding why a cluster is stuck in states like "Provisioning", "Updating", or "Unavailable" often requires jumping between these resources, reading conditions, and correlating them with agent connectivity and known failure modes. This project will build a Cluster Lifecycle Visualizer: a small, read-only controller that runs in the Rancher management cluster and generates a single, human-friendly view per cluster. It will watch Rancher cluster CRDs, derive a simplified lifecycle phase, keep a history of phase transitions from installation time onward, and attach a short, actionable recommendation string that hints at what the operator should check or do next.

Goals

• Provide a compact lifecycle summary for each Rancher-managed cluster (e.g. Provisioning, WaitingForClusterAgent, Active, Updating, Error) derived from provisioning.cattle.io/v1 Cluster and management.cattle.io/v3 Cluster status and conditions.
• Maintain a phase history for each cluster, allowing operators to see how its state evolved over time since the visualizer was installed.
• Attach a recommended action to the current phase using a small ruleset based on common Rancher failure modes (for example, cluster agent not connected, cluster still stabilizing after an upgrade, or generic error states), to improve the day-to-day debugging experience.
• Deliver an easy-to-install, read-only component (single YAML or small Helm chart) that Rancher users can deploy to their management cluster and inspect via kubectl get/describe, without UI changes or direct access to downstream clusters.
• Use idiomatic Go, wrangler, and Rancher APIs.

Resources

• Rancher Manager documentation on RKE2 and K3s cluster configuration and provisioning flows.
• Rancher API Go types for provisioning.cattle.io/v1 and management.cattle.io/v3 (from the rancher/rancher repository or published Go packages).
• Existing Rancher architecture docs and internal notes about cluster provisioning, cluster agents, and node agents.
• A local Rancher management cluster (k3s or RKE2) with a few test downstream clusters to validate phase detection, history tracking, and recommendations.

---

Rancher/k8s Trouble-Maker by tonyhansen

Project Description

When studying for my RHCSA, I found trouble-maker, which is a program that breaks a Linux OS and requires you to fix it. I want to create something similar for Rancher/k8s that can allow for troubleshooting an unknown environment.

Goals for Hackweek 25

• Update to modern Rancher and verify that existing tests still work
• Change testing logic to populate secrets instead of requiring a secondary script
• Add new tests

Goals for Hackweek 24 (Complete)

• Create a basic framework for creating Rancher/k8s cluster lab environments as needed for the Break/Fix
• Create at least 5 modules that can be applied to the cluster and require troubleshooting

Resources

• https://github.com/celidon/rancher-troublemaker
• https://github.com/rancher/terraform-provider-rancher2
• https://github.com/rancher/tf-rancher-up
• https://github.com/rancher/quickstart

---

The Agentic Rancher Experiment: Do Androids Dream of Electric Cattle? by moio

Rancher is a beast of a codebase. Let's investigate if the new 2025 generation of GitHub Autonomous Coding Agents and Copilot Workspaces can actually tame it.

---

The Plan

Create a sandbox GitHub Organization, clone in key Rancher repositories, and let the AI loose to see if it can handle real-world enterprise OSS maintenance - or if it just hallucinates new breeds of Kubernetes resources!

Specifically, throw "Agentic Coders" some typical tasks in a complex, long-lived open-source project, such as:

❥ The Grunt Work: generate missing GoDocs, unit tests, and refactorings. Rebase PRs.

❥ The Complex Stuff: fix actual (historical) bugs and feature requests to see if they can traverse the complexity without (too much) human hand-holding.

❥ Hunting Down Gaps: find areas lacking in docs, areas of improvement in code, dependency bumps, and so on.

If time allows, also experiment with Model Context Protocol (MCP) to give agents context on our specific build pipelines and CI/CD logs.

Why?

We know AI can write "Hello World." and also moderately complex programs from a green field. But can it rebase a 3-month-old PR with conflicts in rancher/rancher? I want to find the breaking point of current AI agents to determine if and how they can help us to reduce our technical debt, work faster and better. At the same time, find out about pitfalls and shortcomings.

The CONCLUSION!!!

A State of the Union document was compiled to summarize lessons learned this week. For more gory details, just read on the diary below!

---

Cluster API Provider for Harvester by rcase

Project Description

The Cluster API "infrastructure provider" for Harvester, also named CAPHV, makes it possible to use Harvester with Cluster API. This enables people and organisations to create Kubernetes clusters running on VMs created by Harvester using a declarative spec.

The project has been bootstrapped in HackWeek 23, and its code is available here.

Work done in HackWeek 2023

• Have a early working version of the provider available on Rancher Sandbox : *DONE *
• Demonstrated the created cluster can be imported using Rancher Turtles: DONE
• Stretch goal - demonstrate using the new provider with CAPRKE2: DONE and the templates are available on the repo

DONE in HackWeek 24:

• Add more Unit Tests
• Improve Status Conditions for some phases
• Add cloud provider config generation
• Testing with Harvester v1.3.2
• Template improvements
• Issues creation

DONE in 2025 (out of Hackweek)

• Support of ClusterClass
• Add to clusterctl community providers, you can add it directly with clusterctl
• Testing on newer versions of Harvester v1.4.X and v1.5.X
• Support for clusterctl generate cluster ...
• Improve Status Conditions to reflect current state of Infrastructure
• Improve CI (some bugs for release creation)

Goals for HackWeek 2025

• FIRST and FOREMOST, any topic is important to you
• Add e2e testing
• Certify the provider for Rancher Turtles
• Add Machine pool labeling
• Add PCI-e passthrough capabilities.
• Other improvement suggestions are welcome!

Thanks to @isim and Dominic Giebert for their contributions!

Resources

Looking for help from anyone interested in Cluster API (CAPI) or who wants to learn more about Harvester.

This will be an infrastructure provider for Cluster API. Some background reading for the CAPI aspect:

• Cluster infrastructure provider contract
• Machine infrastructure provider contract

---

Self-Scaling LLM Infrastructure Powered by Rancher by ademicev0

Self-Scaling LLM Infrastructure Powered by Rancher

---

Description

The Problem

Running LLMs can get expensive and complex pretty quickly.

Today there are typically two choices:

1. Use cloud APIs like OpenAI or Anthropic. Easy to start with, but costs add up at scale.
2. Self-host everything - set up Kubernetes, figure out GPU scheduling, handle scaling, manage model serving... it's a lot of work.

What if there was a middle ground?

What if infrastructure scaled itself instead of making you scale it?

Can we use existing Rancher capabilities like CAPI, autoscaling, and GitOps to make this simpler instead of building everything from scratch?

Project Repository: github.com/alexander-demicev/llmserverless

---

What This Project Does

A key feature is hybrid deployment: requests can be routed based on complexity or privacy needs. Simple or low-sensitivity queries can use public APIs (like OpenAI), while complex or private requests are handled in-house on local infrastructure. This flexibility allows balancing cost, privacy, and performance - using cloud for routine tasks and on-premises resources for sensitive or demanding workloads.

A complete, self-scaling LLM infrastructure that:

• Scales to zero when idle (no idle costs)
• Scales up automatically when requests come in
• Adds more nodes when needed, removes them when demand drops
• Runs on any infrastructure - laptop, bare metal, or cloud

Think of it as "serverless for LLMs" - focus on building, the infrastructure handles itself.

How It Works

A combination of open source tools working together:

Flow:

• Users interact with OpenWebUI (chat interface)
• Requests go to LiteLLM Gateway
• LiteLLM routes requests to:
  • Ollama (Knative) for local model inference (auto-scales pods)
  • Or cloud APIs for fallback

---