CloudLinux

The #1 feature request at CloudFest 2026 is now live. AI Bot Management ships in the Imunify Security WordPress plugin. Included at no extra cost with Imunify360. We'll cover it in more detail at CloudLinux Product Pulse Q3 on August 27, 4:00 PM CEST / 10:00 AM ET. Also on the agenda: KernelCare live patching to stop rebooting for every kernel security patch Imunify for AI Agents (early access): security at the operating system level for AI agents running on hosting servers Under Attack Mode in Imunify360: one switch puts a flooded domain behind a JavaScript challenge 60 minutes live, including Q&A. Ask a question for a chance to receive CloudLinux swag. Register: https://hubs.ly/Q04t8Lzg0

Scaling a hosting business usually means your costs scale right along with it. It doesn't have to. When PAMIR WEB HOSTING LLC added the CloudLinux VPS Bundle, they brought CloudLinux and Imunify360 to their VPS plans in one contract: lower licensing cost, and a stronger, safer offering for their clients at the same time. Their CEO said it best 👇 Curious what the bundle could do for your margins? Join our VPS Program: https://hubs.ly/Q04s5nQT0 #WebHosting #VPS #CloudLinux

OVH's Januscape post-mortem said it plainly: this emergency procedure will need to be repeated. It took 18 days. On July 19, the last Januscape reboot wave ended: 11 days of coordinated reboots across tens of thousands of hypervisors, with about a million customer VMs riding on top. On July 21, the fix for the next KVM bug was quietly merged upstream. On August 6, it got a CVE and a name: Zapscape. Here's the problem: a fleet that rebooted for Januscape rebooted onto a kernel that predates the Zapscape fix. One month later, the same decision is back. Coordinate another reboot campaign, or patch the kernel while it runs. Servers running KernelCare took the Januscape fix in memory, while the reboot waves were still rolling. No reboot, no maintenance window. Zapscape livepatches are arriving the same way. The full story, including what a well-run reboot campaign actually costs: https://hubs.ly/Q04sVDVv0 #Linux #KVM #LivePatching #KernelCare #CyberSecurity

Heading to WordCamp US and want to make the most of it? Let's talk before the doors open! Our team will be there from 16 August, and we're setting aside time to meet hosts and agencies one on one. Whether it's proactive security with Imunify360 or stability and performance with CloudLinux, we'd love to hear what you're working on. Reach out to the team: sales@cloudlinux.com #WCUS  #WordCampUS  #WordPress  #WebHosting

MariaDB 10.6 has been out of community support since July 6. Nothing broke that day. Your databases still serve queries exactly as before. What changed is quieter: every vulnerability disclosed in 10.6 since then is one the project will never fix. On a shared server, that isn't one site exposed. It's every customer site that touches the database. Upgrading is still the right long-term move, but a fleet-wide major upgrade is a real project, and a rushed one breaks customer sites and drives churn. Endless Lifecycle Support (ELS) for MariaDB keeps 10.6 patched past end of life, delivered as drop-in updates through the package manager you already use. No data migration, no schema changes, no forced upgrade. You close the window now and upgrade on your own timeline. In partnership with TuxCare. Live for CloudLinux 7/8/9 and other EL 7/8/9 systems. Learn more: https://hubs.ly/Q04scRnL0

Missed the live session? It's now available on demand. MariaDB 10.6 reached end of life on July 6. It still runs, but the security patches stopped. In this session we covered what that actually means for the databases still on 10.6, and how to keep them patched without a rushed migration. What we covered: • What end of life really changed for 10.6, and why 2026 makes it sharper • Why "still running" isn't the same as "still patched" • What a version upgrade really costs, and where rushed migrations break customer sites • How ELS for MariaDB keeps 10.6 patched with no data migration Sign up to watch the full recording on demand: https://hubs.ly/Q04sd1mY0

Linux kernel root exploits are landing in waves right now, and each one turns an ordinary local user into root. On a shared server, that's one compromised account away from reaching every other customer on the box. Until a server reboots, it stays exposed. KernelCare closes that window automatically: it patches the running kernel with no reboot, no maintenance window, and no 2 a.m. downtime for your customers. The agent checks for new fixes every four hours and applies them for you, so your whole fleet stays patched. And your cost stays flat per server, no matter how many CVEs land that month. Start a free trial and watch it patch a live server in minutes: https://lnkd.in/eVWEDDmt #WebHosting #LinuxSecurity #KernelCare #Sysadmin #Uptime

Phoenix, AZ we're on our way. CloudLinux and Imunify360 are heading to WordCamp US 2026, 16-19 August at the Phoenix Convention Center! If you build, host, or secure WordPress sites, we'd love to connect. We'll be talking server stability, proactive security, and keeping hosting environments fast and protected! Reach out to the team: sales@cloudlinux.com #WCUS  #WordCampUS  #WordPress  #WebHosting

A single compromised site on a VPS never results in just one support ticket. The infection can easily spread to neighboring accounts, trigger IP blacklisting, and trap your top engineers in hours of manual cleanup, ultimately draining your profit margins. The solution lies in a dedicated security and isolation layer: CloudLinux keeps every account fully contained, while Imunify360 automatically handles malware detection and removal. As a proof, after standardizing its managed VPS on the bundle, HostArmada reported about 80% fewer hacked-site tickets. Fewer tickets. Protected margins. Read the HostArmada story: https://hubs.ly/Q04q-pbt0

The change log used to tell you when something mattered. It doesn't anymore. Vendors are intentionally hiding security fixes inside routine software updates, because if they label something a "security update," attackers immediately start reverse-engineering the patch. Igor Seletskiy's advice for hosting providers in 2026: assume every update is a security update. Don't wait for a label. Don't wait for a CVE number. Update. This applies to your entire stack, not just the Linux kernel. But for the kernel specifically, continuous patching without rebooting is now a real operational requirement. That's exactly what KernelCare is built for.