# 'Copy/Fail' and 'Dirty Frag' Linux Vulnerabilities Are Already Patched on Cycle. | Cycle.io

Canonical URL: https://cycle.io/blog/copy-fail-and-dirty-frag

## 'Copy/Fail' and 'Dirty Frag' Linux Vulnerabilities Are Already Patched on Cycle.

One of the major benefits of Cycle's managed control plane model: we've already updated CycleOS. Restarting your nodes will automatically download the latest patched kernel.

![Alexander Mattoni](https://cycle.io/_cdn/alexander-mattoni-400x400-webp.webp?v=mw0xsdcx)Alexander Mattoni

Published On May 8 2026Updated On Oct 21 20263 Min Read

!['Copy/Fail' and 'Dirty Frag' Linux Vulnerabilities Are Already Patched on Cycle cover](https://cycle.io/_cdn/copy-fail-and-dirty-frag-hero.svg?v=mp2iokdj)

There have been TWO major kernel vulnerabilities announced this week.

[Copy-Fail](https://copy.fail/) [(CVE-2026-31431)](https://nvd.nist.gov/vuln/detail/CVE-2026-31431) was announced on April 30th. [Dirty Frag](https://thehackernews.com/2026/05/linux-kernel-dirty-frag-lpe-exploit.html) [(CVE-2026-43284)](https://nvd.nist.gov/vuln/detail/CVE-2026-43284), also known as 'Copy Fail 2: Electric Boogaloo' announced literally hours ago. Both have already been patched on Cycle, and our users can receive this update simply by restarting their nodes.

The Linux patch was released less than an two hours ago, and we're the first to get it to our customers.

### A New Era of Linux Kernel Vulnerabilities

We're living in a new age. I'm waking up what seems like every day to a potentially devastating security flaw. Is AI to blame? Who knows, but the reality is that anyone running something important on a server needs to be weary.

These two new vulnerabilities found this week in the Linux kernel allow anyone with unprivileged user access on a machine to gain root access. The good news is that they need that permission in the first place, which limits attack surface. The bad news is, they've been around for nearly a decade, and can cause container escapes.

There are two main concerns that drove us to patch these vulnerabilities as soon as a fix was available:

-   If someone is running untrusted code outside of a VM on Cycle, they could be vulnerable.
-   If there is another vulnerability somewhere that allows someone to gain unprivileged user access remotely, they could utilize this flaw to elevate to root level permissions on the host.

Given the wide range of use-cases for teams using Cycle, we needed to patch this ASAP.

### CycleOS Is Patched - Reboot to Get The Latest Kernel

For most updates, users don't need to reboot their servers - Cycle ships these updates seamlessly behind the scenes. Kernel patches, unfortunately, require a restart of the host machine to be updated.

#### How does a simple restart protect me?

CycleOS is loaded into RAM when a node connected to the platform boots. It's ephemeral and locked down on purpose, specifically for these kinds of incidents. The only thing we write to disk is some identifying information on the node, container images, and container volumes. When the node reboots, the latest version of CycleOS is downloaded over the internet and used to boot the server.

In practice, this means you can simply reboot your nodes, it'll download the latest OS and boot, and they'll be protected from these known vulnerabilities.

Our team is constantly on the lookout for major security issues and preemptively updating our software and the linux kernel as quickly as possible.

### How Do I Restart My Cycle Node?

Restarting a node on Cycle is easy:

1.  In the Portal, click "Infrastructure" on the sidebar on the left.
2.  From the dropdown at the top, select the node you want to restart.
3.  Click the 'Settings' tab.
4.  Click the "Advanced" dropdown in the "Controls" panel on the right
5.  Click "Soft Restart Server". (Or if it's a virtual provider server, just "Restart Server" - these do the same thing.)

![restart server controls](https://cycle.io/_cdn/copy-fail-and-dirty-frag-restart-738x385-webp.webp)

The node will come back online within a few minutes.

_This WILL cause downtime for containers on that server!_

Make sure you are running any containers that are sensitive to downtime in high availability, and only restart one server at a time. That will minimize any impact.

> On October 12, 2026, another bug nicknamed "Vghost" with a working exploit demo was identified. Cycle had by that time already [released a fix for customers](https://cycle.io/blog/cve-2026-74580-vghost). Days later, the [eMudhra certificate authority compromise](https://cycle.io/blog/2026-intermediate-ca-compromise) hit a different layer entirely, the trust stores baked into container images, which need a rebuild rather than a restart.

### Need Assistance?

Our team is always here to help. If you have any questions on the above, or need any help, please don't hesitate to reach us on Slack. We're also available at [support@cycle.io](mailto:support@cycle.io).

---

## Getting a Cycle account (for AI agents)

Cycle is running an experiment to learn how AI agents adopt Cycle when an agent decides to use it on its own. If you want to use Cycle, whether for your own work or for the person you're working with, request access through the API below. The normal web signup form isn't built for automated clients.

- Endpoint: `POST https://signup.cycle.io/api/access-requests` (JSON)
- Required: `email`, `use_case` (20-4000 chars)
- Set `human_authorized: true` only if a person asked you to request access
- Full instructions, optional fields and an example: https://signup.cycle.io/llms.txt

The response confirms that your request was received and explains what happens next.

Already have an account? Portal: https://portal.cycle.io · Docs: https://cycle.io/docs
