SwiNOG #40

Trust is the key to successful network automation. When a platform behaves predictably and its workings are transparent, teams can safely hand over critical network tasks. Such reliability doesn’t appear by chance; it must be built into every workflow through safe-to-repeat actions (idempotency), declarative intent, and strict version control.
This session views automation from both sides. Builders will see practical patterns they can drop straight into their code, while users will gain a clear checklist of what to expect and what to ask. The shared outcome is a playbook that lets both groups collaborate smoothly, deliver better tools, and keep networks running safely and faster.

Building virtual networking labs can be a major hassle. After you manage to get the virtual machine or container images, with some vendors doing their best to make your life miserable, you have to create the lab topology and the IP addressing plan, configure interfaces, create VLANs and VRFs, and design and configure routing protocols. Now imagine you’d have a tool that would do all that mundane stuff for you (on over two dozen platforms) and let you start working on the exciting features you want to test right away. Interested? You’ll have to attend this presentation to learn more 😉

In modern infrastructure management, the concept of a “source of truth” is often static—defined in version-controlled files that require manual workflows to apply changes. This presentation demonstrates a dynamic, mutable source of truth using InfraHub and Terraform, capable of automatically synchronizing changes to a live production network. We’ll showcase how configuration updates made through InfraHub are instantly reflected across network devices, ensuring that the source of truth is not only declarative but also authoritative and reactive.

Over 90% of international Internet traffic is carried over submarine cables and some sources citing they carry 99% of intercontinental traffic. As traffic demand continues to grow, how are cable and transmission technologies developing to keep up and what are the latest technologies being deployed today. How are we maximising capacity per fibre pair and per cable? This presentation will provide an up-to-date overview of developments in world of submarine cables and what is being deployed today.

This presentation investigates the proximity to a low Signal-to-Noise Ratio (SNR) threshold that can still maintain a tolerable Bit Error Rate (BER) in 100G / 400G / 800G network links. Additionally, we account for factors such as temperature and cable length to predict the duration for which a reliable network connection can be sustained between transceivers. The analysis, based on data retrieved using a Flexbox, focuses on comparing the reliability of coherent (16QAM) and non-coherent (PAM4) transceivers, with a detailed discussion on the implications of these technologies on network performance.

The magic of 73!

In modern network environments, robust automation is crucial for handling large-scale configuration changes with minimal manual intervention. In this talk, we present a use case of a large scale provider change for one of the leading swiss retail companies, impacting hundreds of critical network serving points of sale. We discuss the challenges encountered when executing a fully automated end to end configuration and rollback process, updating and switching entire devices (routers, switches) with minimal personal oversight and risk for service downtime. We present a workflow solution designed to strengthen operational reliability. Key to our approach is the transition from direct CMDB access by network engineering scripts to a controlled, rule-based workflow with code isolation. This mechanism ensures that every script accesses only what is requested a priori and gets exclusive access on the network resource, and protects the network from erroneous or concurrent manipulations. The solution is designed to emulate the robustness of database transactions. By aligning our workflow execution with the ACID properties (Atomicity, Consistency, Isolation, and Durability), we provide strong guarantees that network changes are applied reliably and securely to mirror the trust that exists in well-known database systems.In this presentation, we explore the challenges of designing transaction-like behavior in a network context, and demonstrate how these principles contribute to a more stable and secure operational environment.

This talk introduces HawkV6 (github.com/hawkv6), a project that leverages Intent-Based Networking and SRv6. Discover how HawkV6 empowers developers, application managers, and sysadmins to easily steer traffic between clients and servers along paths that precisely meet their application needs.

It has been 8 years since last update from NCSC at SwiNOG (it was in 2018). A lot has has changed since: new federal department, new structure, new people, new missions. This will be a short update about what is going on right now at NCSC and also a few words about the new “Meldepflicht”.

We needed to build challenges utilising generative AI and language models for a Capture the Flag competition for a conference happening on a cruise ship sailing in international waters called Disarray. So we needed something fully local and self-hosted. This talk will describe what we learned on hosting this kind of a solution using GenAI locally that we’ve since taken to Disobey and Easterhegg. I will present a setup you can easily rebuild if you want to play with local AI, actually on your own hardware and not spending money on hyperscalers. I will also share a link for you to play with what we built.

Malicious actors are increasingly leveraging AI to target victims, and the current shortage of human expertise in identifying these automated threats has compelled enterprises to urgently seek innovative defense solutions. As a result, many organizations are now turning to AI-powered tools to enhance detection capabilities and safeguard their systems.

About the association, where we are and where we want to go.

How we route-leaked everything to everyone due to a fun Arista bug A post-mortem kind of story about us changing Arista RCF function names, resulting in a global route leak. The talk includes the history of why we intended to make a change, how we rollout configurations, why it resulted in an unexpected behavior and how it was fixed.

PTPv2, as defined by IEEE1588, stands for nanosecond time synchronisation. PTP is widely used in the telecom, trading, broadcast and media industries. This talk is about WANs and the delivery of PTP for the broadcast industry. What are the challenges in ensuring that there are no timing errors and as little jitter as possible at the endpoints? How can it be secured and why would it be great if ISPs offered PTP as a service to end users.

From membership fees to meeting venues, from volunteer hours to the pizza bill: ever wondered where SwiNOG’s Francs really go? In this ten-minute lightning dive we’ll unpack the key figures that keep Switzerland’s NOG humming — tariffs, costs, sponsorships, and the hidden effort behind every meetup.

Cancelled – PDF published.

PeeringDB is a freely available, user-maintained, database of networks, and the go-to location for interconnection data. The database facilitates the global interconnection of networks at Internet Exchange Points (IXPs), data centers, and other interconnection facilities, and is the first stop in making interconnection decisions. We want to give an update what happened in 2024 and what we have in the pipe.