Okay, this is weird.

I am seeing `tailscaled` spiking up to 250% CPU usage every other minute, with up to 60 MBit/s outgoing traffic on the `tailscale0` interface... and on the `wg0` interface I see the same amount of traffic during that period.

I now disabled wireguard and the issue is gone.

But I have absolutely NO idea why this is happening, what causes it, and why it (seemingly) suddenly started - I did not have this issue earlier today. All I did in between was update the unstable channel, but all packages involved here (tailscale, wireguard) are from stable. So... 🤔 I am at loss...

#tailscale #wireguard #linux #linuxnetworking #networking #sysadmin #nixos

Unser @dd_ix Workshop zum Thema BGP Routing wurde auf den @clt_news angenommen 💪 . Für alle die meine Vorträge der letzten Jahre über Alpine Linux, ifstate oder Linux Router interessant fanden bietet sich hier die Möglichkeit für praktische Erfahrungen und Austausch.

https://chemnitzer.linux-tage.de/2026/de/programm/beitrag/329

#CLT2026 #DDIX #ifstate #RouterKit #linuxnetworking

Chemnitzer Linux-Tage 2026 · Workshop: µInternet - Routing ohne Default Gateway

Die Chemnitzer Linux-Tage sind eine Veranstaltung rund um das Thema Linux und Open Source für jedermann, die Linux-Nutzer, Insider und Unternehmen zusammenbringt.

ifstate 2.2.4 - a tool for declarative network configuration for Linux - was released:
https://codeberg.org/liske/ifstate/releases/tag/2.2.4

The 75th release includes various fixes & improvements like:
- make stable-privacy IPv6 addresses work
- resolve wireguard endpoints *after* configuring the network stack so that DNS resolution could work

This is the first release with published Debian packages: https://ifstate.net/2.2/docs/install/#debian-gnulinux

#linuxnetworking #Debian #ifstate #wireguard

2.2.4 - routerkit/ifstate

## Changes - debian: add packaging - docs: add nixos strongSwan config for examples/xfrm-vrf (#147) - nix: add mkdocs-macros-plugin to docs package and devshell (#147) - nix: update flake.lock for nixos 25.11 (#147) - routing: remove log output if no routing is configured - wireguard: con...

Codeberg.org

The ifstate git repository on #Codeberg has been moved from my personal account into the RouterKit project at https://codeberg.org/routerkit/ifstate

With moving the project it gains access to CI/CD runners and the first step is to use them to provide #Debian packages from the upcoming 2.2.4 release 🥳
https://codeberg.org/routerkit/-/packages/debian/ifstate/

I already use it in prod for haproxy setups on Debian trixie using different network namespaces for listeners and server access.

#ifstate #RouterKit #haproxy #linuxnetworking #Debian

ifstate

Manage host interface settings in a declarative manner

Codeberg.org

I don't like to maintain packaging upstream, but… I required netns configuration support on Debian for using network namespaces with haproxy!

Sadly I had to use dh_virtualenv because pyroute2 in Debian sid+stable is to old to have all required netns related features.

Haproxy has namespace support for listeners and servers which allows one to easily build multi-tenant load-balancing setups. 😎

https://codeberg.org/liske/ifstate/src/branch/master/debian

#linuxnetworking #ifstate #haproxy #pyroute2 #Debian

The ifstate 2.2 release - a tool for declarative network configuration for Linux - is available in the most recent stable releases of Alpine Linux 3.23 and NixOS 25.11. 🥳

Fun fact: this is the first NixOS release where you can use a declarative network configuration out of the box 😉

https://ifstate.net/2.2/docs/install/#nixos

#AlpineLinux #ifstate #linuxnetworking #nixos #nixpkgs

Installation - IfState

None

#ifstate 2.2.[01] - a tool for declarative network configuration for Linux - was released:
https://codeberg.org/liske/ifstate/releases/tag/2.2.0
https://codeberg.org/liske/ifstate/releases/tag/2.2.1

This release includes:
- support for vlan_flags (loose_binding, …)
- the wgnlpy python dependency has been dropped

…and it also contains several fixes, the most import is a regression: interface defaults were not applied since ifstate 2.0

(already available in @alpinelinux edge + v3.23 and #nixpkgs unstable + 25.11)

#linuxnetworking

2.2.0 - liske/ifstate

## Changes - docs: various minor updates - fdb: ignore IEEE Standard Group MAC Addresses by default (`01-80-C2`) - link: add vlan_flags support (closes #133) - wireguard: drop wgnlpy dependency, use pyroute2's WireGuard class ## Fixes - addresses: do not enforce addrproto for ::1 defa...

Codeberg.org

Tìm kiếm công cụ trực quan hóa cấu trúc mạng phức tạp, kết hợp Docker, LXC và Linux. Người dùng muốn xem các kết nối veth, bridge giữa các container. Atlas chỉ hỗ trợ Docker. Có gợi ý nào không? #networktopology #visualizetool #docker #lxc #linuxnetworking #networking #cấutrúc mạng #trựcquanchung #tinhtech

https://www.reddit.com/r/selfhosted/comments/1p1i04c/visualize_network_topology/

🐧💻 "Escape the Linux Networking Stack" – because understanding it makes it vanish! Poof! 💨 The author attempts to unravel the mystery, but spoiler alert: nobody truly escapes, not even Cloudflare's blog notifications. 🐟🔍
https://blog.cloudflare.com/so-long-and-thanks-for-all-the-fish-how-to-escape-the-linux-networking-stack/ #LinuxNetworking #EscapeTheStack #Cloudflare #TechMystery #NetworkingInsights #HackerNews #ngated
So long, and thanks for all the fish- how to escape the Linux networking stack

Many products at Cloudflare aren’t possible without pushing the limits of network hardware and software to deliver improved performance, increased efficiency, or novel capabilities such as soft-unicast, our method for sharing IP subnets across data centers. Happily, most people do not need to know the intricacies of how your operating system handles network and Internet access in general. Yes, even most people within Cloudflare. But sometimes we try to push well beyond the design intentions of Linux’s networking stack. This is a story about one of those attempts.

The Cloudflare Blog
So long, and thanks for all the fish- how to escape the Linux networking stack

Many products at Cloudflare aren’t possible without pushing the limits of network hardware and software to deliver improved performance, increased efficiency, or novel capabilities such as soft-unicast, our method for sharing IP subnets across data centers. Happily, most people do not need to know the intricacies of how your operating system handles network and Internet access in general. Yes, even most people within Cloudflare. But sometimes we try to push well beyond the design intentions of Linux’s networking stack. This is a story about one of those attempts.

The Cloudflare Blog