CloudLinux’s Post

This is crazy: five Claude Code instances working on the same distributed project, each running on its own server/container, in a different part of the world.

I direct them, but they talk to each other directly: agreeing on contracts, deployments, integrations…

The scary part: every now and then, they seem to forget I exist and start making decisions and actions on their own to move toward the goal, based on their own interpretation and their own rules, without asking for permission.

This is scary.

We can either sit on the sidelines and wait for someone else to solve tomorrow’s infrastructure challenges, or we can step up and build the solution ourselves.

As the Linux ecosystem transitions toward Wayland, enterprise IT faces a real hurdle: there is no shared standard for how headless, multi-user remote desktop should work on servers.

Many system administrators rightly ask: X11 still works fine for us today, why bother changing?

The reality is that the shift is happening upstream whether we like it or not. The question isnt if Wayland will become the default, but whether organizations will have a stable, predictable environment when it does.

When Jonas Ådahl at Red Hat initiated work on the RemoteDesktop1 specification, we immediately recognized it was the right path forward. But vision alone doesn't ship standards if no one does the heavy lifting.

As a leading player in Linux remote desktop, taking responsibility was the only rational choice for us at Cendio:
• Funding core open-source development
• Committing our own senior engineers (huge shoutout to Samuel Mannehed and the team) • Building entirely in the open alongside GNOME, KDE, and the broader community.

Will others in the industry get a free ride on this work once its done? Absolutely, and we’re fine with that. That’s how open source is supposed to work.

For us, it’s about taking the lead. By shaping the standard from the ground up, we ensure the foundation is built right—so our ThinLinc customers and the broader community can rest easy knowing they have a world-class solution long after X11 is phased out.

Driving open standards takes time and patience, but it’s the only way to build infrastructure that actually lasts.

Fantastic work by the team, and great collaboration across project boundaries!

(Link to Samuel’s technical breakdown of the spec in the first comment below 👇)

#Linux #Wayland #OpenSource #ThinLinc #Cendio #TechLeadership

🐧 It's Time to Embrace Linux. It's Time to Embrace Open Source.
What if you could take control of your technology instead of being controlled by it?

What if you could build, customize, and improve your software without being locked into a vendor's ecosystem?

That's the power of Linux and Open Source. 🚀
💡 Why make the switch?
🔓 Freedom: Use, study, modify, and share software.
💰 Cost-effectiveness: Reduce software licensing costs and invest more in innovation.
🛡️ Transparency: Access the source code and build trust through openness.
⚙️ Flexibility: Customize solutions to meet your exact requirements.
🌱 Sustainability: Extend the life of your systems and reduce unnecessary dependence on proprietary technologies.
🤝 Community: Become part of a global ecosystem of developers, organizations, and enthusiasts who believe in collaboration.

Open Source isn't just about free software. It's about freedom of choice, shared knowledge, and the ability to shape your own digital future.
Whether you're an individual, a startup, an educational institution, or a large enterprise, there is an opportunity to do more with open technologies.
You don't have to change everything overnight. Start with a Linux server. Explore an open-source application. Migrate one workload. Learn, experiment, and grow.

The future of technology should be open. Be a part of it!

At Carving IT, we believe in the power of open-source technologies to build practical, flexible, and sustainable solutions.

🐧 Choose Linux. Choose Open Source. Choose Freedom.

#Linux #OpenSource #DigitalFreedom #OpenSourceSoftware #Technology #DigitalTransformation #CarvingIT

A private fork feels like the fast option. Then every upstream release arrives with an invoice attached: another rebase, another round of conflicts, another security fix you have to port by hand.

This December marks ten years since my first commit to OpenVPN, and we've taken the other route ever since. We've merged hundreds of commits into the project, from the netlink networking layer in 2.5 and data channel offload in 2.6 to six of the headline features in 2.7.

📦 𝗪𝗵𝗮𝘁 𝘂𝗽𝘀𝘁𝗿𝗲𝗮𝗺 𝗯𝘂𝘆𝘀 𝘆𝗼𝘂
A feature merged upstream ships in the official release, reaches every distribution and downstream product that builds on it, and gets tested by thousands of users you will never meet. There is no private patch set to carry into the next version, fixes arrive with the release instead of on your own budget, and your product stays compatible with the rest of the ecosystem.

🤝 𝗧𝗿𝘂𝘀𝘁 𝗰𝗼𝗺𝗽𝗼𝘂𝗻𝗱𝘀
The part nobody puts in a roadmap is trust. Maintainers read a patch from someone they know differently than one from a stranger, because they've seen how that person handles feedback and whether they stay around after the merge. Multi-socket support in OpenVPN had been attempted before and set aside more than once, until Gianmarco De Gregori on our team picked it up, rebuilt the foundation it needed and carried it through review. It shipped as the first new feature in the OpenVPN 2.7 release notes.

🎯 𝗙𝗼𝗿 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝘁𝗲𝗮𝗺𝘀
If your product depends on an open-source project, upstream belongs in the plan from day one, with time budgeted for review and people who already know the community. Done that way it's rarely the slow path, and it's the one that stops costing you later.

Is there a patch in your product that should have gone upstream years ago?

The full story is in the first comment.

#OpenSource #OpenVPN #Linux #Networking #ProductDevelopment

Useful framing for engineering leaders who depend on open source: a private fork turns every upstream release and security fix into recurring maintenance work. Budgeting for upstream review and long-term maintainer trust is part of supply-chain resilience, not just community goodwill.

Antonio Quartulli

Sep 26

A private fork feels like the fast option. Then every upstream release arrives with an invoice attached: another rebase, another round of conflicts, another security fix you have to port by hand.

This December marks ten years since my first commit to OpenVPN, and we've taken the other route ever since. We've merged hundreds of commits into the project, from the netlink networking layer in 2.5 and data channel offload in 2.6 to six of the headline features in 2.7.

📦 𝗪𝗵𝗮𝘁 𝘂𝗽𝘀𝘁𝗿𝗲𝗮𝗺 𝗯𝘂𝘆𝘀 𝘆𝗼𝘂
A feature merged upstream ships in the official release, reaches every distribution and downstream product that builds on it, and gets tested by thousands of users you will never meet. There is no private patch set to carry into the next version, fixes arrive with the release instead of on your own budget, and your product stays compatible with the rest of the ecosystem.

🤝 𝗧𝗿𝘂𝘀𝘁 𝗰𝗼𝗺𝗽𝗼𝘂𝗻𝗱𝘀
The part nobody puts in a roadmap is trust. Maintainers read a patch from someone they know differently than one from a stranger, because they've seen how that person handles feedback and whether they stay around after the merge. Multi-socket support in OpenVPN had been attempted before and set aside more than once, until Gianmarco De Gregori on our team picked it up, rebuilt the foundation it needed and carried it through review. It shipped as the first new feature in the OpenVPN 2.7 release notes.

🎯 𝗙𝗼𝗿 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝘁𝗲𝗮𝗺𝘀
If your product depends on an open-source project, upstream belongs in the plan from day one, with time budgeted for review and people who already know the community. Done that way it's rarely the slow path, and it's the one that stops costing you later.

Is there a patch in your product that should have gone upstream years ago?

The full story is in the first comment.

#OpenSource #OpenVPN #Linux #Networking #ProductDevelopment

With just under 10,000 Github stars, Sn1per's open-source core is one of the most widely-adopted offensive-security tools on GitHub. Our commercial editions build on that proven foundation with team workflows, enterprise support, and hardened deployment.

We built a few 𝗠𝗖𝗣 𝘀𝗲𝗿𝘃𝗲𝗿𝘀 for our internal tools, and they are working really well now.

Earlier, doing any operation in a tool meant opening it, going through many screens, and clicking each option by hand. Even a small update took more time than it should.

Now these tools are connected with 𝗦𝗜 𝘁𝗵𝗿𝗼𝘂𝗴𝗵 𝗠𝗖𝗣. Anyone on the team can just type what is needed in simple English, and the work gets done.

Create, read, update and delete, all from one chat. No training is needed and nobody has to remember where each option is placed.

What I liked most is that the team started using it from day one. We did not have to explain much.

MCP is not only for developers. If your team uses internal tools daily, connecting them to SI is worth a try.

Red Hat Satellite is a powerful enterprise platform, but the subscription cost can be difficult to justify for some teams.

Foreman + Katello gives you an open-source way to manage Linux infrastructure with many of the core capabilities teams expect from Satellite.

You can manage repositories, Content Views, lifecycle environments, patching, provisioning, host registration, and remote execution from a centralized platform.

The biggest difference is support and operational ownership. With Satellite, you get official Red Hat support, vendor-backed upgrade paths, and enterprise assurance. With Foreman + Katello, your own team takes more responsibility for upgrades, validation, troubleshooting, and maintenance.

That does not make Foreman + Katello a lab-only solution. For organizations with strong Linux skills, good internal processes, and tighter budgets, it can be a very capable production option.

The real question is not simply “Which tool has more features?”

It is “How much operational responsibility is your team prepared to own?”

#Linux #RedHat #Foreman #Katello #SysAdmin #Infrastructure #DevOps #PatchManagement #OpenSource

Agents, Workers – Cloudflare API MCP server sends Workers Issues as MCP Events - Workers Issues provides built-in error monitoring for Cloudflare Workers. It detects production failures and groups related failures into issues without an SDK or application wrapper. The Cloudflare API MCP server now sends these issues as MCP Events ↗︎, the first…
https://lnkd.in/eD8Vjjff

Catch Nick's talk tomorrow where he shows you how to securely expose OpenClaw as a remote MCP server leveraging Pomerium as an MCP gateway. 👀

#mcp #openclaw #zerotrust

Nick Taylor

Oct 4

Looking forward to the MCP Developers Summit Toronto tomorrow! If you're attending, come check out my talk at 2:50pm in the Giovanni room. I'll show you how to secure your OpenClaw as a remote MCP server. 👀

https://lnkd.in/gneJbs-V

#mcp #zerotrust #finegrainedauthorization #openclaw