**Linux** is an open-source, privacy-focused desktop operating system alternative. In the face of pervasive telemetry and other privacy-encroaching technologies in mainstream operating systems, Linux desktop has remained the clear choice for people looking for total control over their computers from the ground up.
Our website generally uses the term “Linux” to describe **desktop** Linux distributions. Other operating systems which also use the Linux kernel such as ChromeOS, Android, and Qubes OS are not discussed on this page.
There are some notable privacy concerns with Linux which you should be aware of. Despite these drawbacks, desktop Linux distributions are still great for most people who want to:
It is a [common misconception](../basics/common-misconceptions.md#open-source-software-is-always-secure-or-proprietary-software-is-more-secure) that Linux and other open-source software is inherently secure simply because the source code is available. There is an expectation that community verification occurs regularly, but this isn’t always [the case](https://seirdy.one/posts/2022/02/02/floss-security/).
In reality, distro security depends on a number of factors, such as project activity, developer experience, the level of rigor applied to code reviews, and how often attention is given to specific parts of the codebase that may go untouched for years.
### Missing Security Features
At the moment, desktop Linux [falls behind alternatives](https://discussion.fedoraproject.org/t/fedora-strategy-2028-proposal-fedora-linux-is-as-secure-as-macos/46899/9) like macOS or Android when it comes to certain security features. We hope to see improvements in these areas in the future.
- **Verified boot** on Linux is not as robust as alternatives such as Apple’s [Secure Boot](https://support.apple.com/guide/security/secac71d5623/web) or Android’s [Verified Boot](https://source.android.com/security/verifiedboot). Verified boot prevents persistent tampering by malware and [evil maid attacks](https://en.wikipedia.org/wiki/Evil_Maid_attack), but is still largely [unavailable on even the most advanced distributions](https://discussion.fedoraproject.org/t/has-silverblue-achieved-verified-boot/27251/3).
- **Strong sandboxing** for apps on Linux is severely lacking, even with containerized apps like Flatpaks or sandboxing solutions like Firejail. Flatpak is the most promising sandboxing utility for Linux thus far, but is still deficient in many areas and allows for [unsafe defaults](https://flatkill.org/2020/) which allow most apps to trivially bypass their sandbox.
Additionally, Linux falls behind in implementing [exploit mitigations](https://madaidans-insecurities.github.io/linux.html#exploit-mitigations) which are now standard on other operating systems, such as Arbitrary Code Guard on Windows or Hardened Runtime on macOS. Also, most Linux programs and Linux itself are coded in memory-unsafe languages. Memory corruption bugs are responsible for the [majority of vulnerabilities](https://msrc.microsoft.com/blog/2019/07/a-proactive-approach-to-more-secure-code/) fixed and assigned a CVE. While this is also true for Windows and macOS, they are quickly making progress on adopting memory-safe languages—such as Rust and Swift, respectively—while there is no similar effort to rewrite Linux in a memory-safe language like Rust.
Not all Linux distributions are created equal. Our [Linux recommendation page](../desktop.md) is not meant to be an authoritative source on which distribution you should use, but our recommendations *are* aligned with the following guidelines. These are a few things you should keep in mind when choosing a distribution:
We highly recommend that you choose distributions which stay close to the stable upstream software releases, often referred to as rolling release distributions. This is because frozen release cycle distributions often don’t update package versions and fall behind on security updates.
For frozen distributions such as [Debian](https://www.debian.org/security/faq#handling), package maintainers are expected to backport patches to fix vulnerabilities rather than bump the software to the “next version” released by the upstream developer. Some security fixes [do not](https://arxiv.org/abs/2105.14565) receive a [CVE ID](https://en.wikipedia.org/wiki/Common_Vulnerabilities_and_Exposures) (particularly less popular software) at all and therefore do not make it into the distribution with this patching model. As a result minor security fixes are sometimes held back until the next major release.
We don’t believe holding packages back and applying interim patches is a good idea, as it diverges from the way the developer might have intended the software to work. [Richard Brown](https://rootco.de/aboutme/) has a presentation about this:
<iframewidth="560"height="315"src="https://invidious.privacyguides.net/embed/i8c0mg_mS7U?local=true"title="Regular Releases are Wrong, Roll for your life"frameborder="0"allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture"allowfullscreen></iframe>
Traditionally, Linux distributions update by sequentially updating the desired packages. Traditional updates such as those used in Fedora, Arch Linux, and Debian based distributions can be less reliable if an error occurs while updating.
Atomic updating distributions apply updates in full or not at all. Typically, transactional update systems are also atomic.
A transactional update system creates a snapshot that is made before and after an update is applied. If an update fails at any time (perhaps due to a power failure), the update can be easily rolled back to a “last known good state."
The Atomic update method is used for immutable distributions like Silverblue, Tumbleweed, and NixOS and can achieve reliability with this model. [Adam Šamalík](https://twitter.com/adsamalik) provided a presentation on how `rpm-ostree` works with Silverblue:
There is often some confusion between “security-focused” distributions and “pentesting” distributions. A quick search for “the most secure Linux distribution” will often give results like Kali Linux, Black Arch, or Parrot OS. These distributions are offensive penetration testing distributions that bundle tools for testing other systems. They don’t include any “extra security” or defensive mitigations intended for regular use.
Arch and Arch-based distributions are not recommended for those new to Linux (regardless of distribution) as they require regular [system maintenance](https://wiki.archlinux.org/title/System_maintenance). Arch does not have a distribution update mechanism for the underlying software choices. As a result you have to stay aware with current trends and adopt technologies as they supersede older practices on your own.
For a secure system, you are also expected to have sufficient Linux knowledge to properly set up security for their system such as adopting a [mandatory access control](https://en.wikipedia.org/wiki/Mandatory_access_control) system, setting up [kernel module](https://en.wikipedia.org/wiki/Loadable_kernel_module#Security) blacklists, hardening boot parameters, manipulating [sysctl](https://en.wikipedia.org/wiki/Sysctl) parameters, and knowing what components they need such as [Polkit](https://en.wikipedia.org/wiki/Polkit).
Anyone using the [Arch User Repository (AUR)](https://wiki.archlinux.org/title/Arch_User_Repository) **must** be comfortable auditing PKGBUILDs that they download from that service. AUR packages are community-produced content and are not vetted in any way, and therefore are vulnerable to software supply chain attacks, which has in fact happened [in the past](https://www.bleepingcomputer.com/news/security/malware-found-in-arch-linux-aur-package-repository/).
The AUR should always be used sparingly, and often there is a lot of bad advice on various pages which direct people to blindly use [AUR helpers](https://wiki.archlinux.org/title/AUR_helpers) without sufficient warning. Similar warnings apply to use third-party Personal Package Archives (PPAs) on Debian based distributions or Community Projects (COPR) on Fedora.
- **Manjaro**: This distribution holds packages back for 2 weeks to make sure that their own changes don’t break, not to make sure that upstream is stable. When AUR packages are used, they are often built against the latest [libraries](https://en.wikipedia.org/wiki/Library_(computing)) from Arch’s repositories.
- **Garuda**: They use [Chaotic-AUR](https://aur.chaotic.cx/) which automatically and blindly compiles packages from the AUR. There is no verification process to make sure that the AUR packages don’t suffer from supply chain attacks.
We recommend **against** using the Linux-libre kernel, since it [removes security mitigations](https://www.phoronix.com/news/GNU-Linux-Libre-5.7-Released) and [suppresses kernel warnings](https://news.ycombinator.com/item?id=29674846) about vulnerable microcode.
Most Linux distributions have an option within its installer for enabling [LUKS](../encryption.md#linux-unified-key-setup) FDE. If this option isn’t set at installation time, you will have to backup your data and re-install, as encryption is applied after [disk partitioning](https://en.wikipedia.org/wiki/Disk_partitioning), but before [file systems](https://en.wikipedia.org/wiki/File_system) are formatted. We also suggest securely erasing your storage device:
Consider using [ZRAM](https://wiki.archlinux.org/title/Zram#Using_zram-generator) instead of a traditional swap file or partition to avoid writing potentially sensitive memory data to persistent storage (and improve performance). Fedora-based distributions [use ZRAM by default](https://fedoraproject.org/wiki/Changes/SwapOnZRAM).
If you require suspend-to-disk (hibernation) functionality, you will still need to use a traditional swap file or partition. Make sure that any swap space you do have on a persistent storage device is [encrypted](https://wiki.archlinux.org/title/Dm-crypt/Swap_encryption) at a minimum to mitigate some of these threats.
We recommend using a desktop environment that supports the [Wayland](https://en.wikipedia.org/wiki/Wayland_(display_server_protocol)) display protocol, as it was developed with security [in mind](https://lwn.net/Articles/589147/). Its predecessor ([X11](https://en.wikipedia.org/wiki/X_Window_System)) does not support GUI isolation, which allows any window to [record, log, and inject inputs in other windows](https://blog.invisiblethings.org/2011/04/23/linux-security-circus-on-gui-isolation.html), making any attempt at sandboxing futile. While there are options to do nested X11 such as [Xpra](https://en.wikipedia.org/wiki/Xpra) or [Xephyr](https://en.wikipedia.org/wiki/Xephyr), they often come with negative performance consequences, and are neither convenient to set up nor preferable over Wayland.
Fortunately, common environments such as [GNOME](https://www.gnome.org), [KDE](https://kde.org), and the window manager [Sway](https://swaywm.org) have support for Wayland. Some distributions like Fedora and Tumbleweed use it by default, and some others may do so in the future as X11 is in [hard maintenance mode](https://www.phoronix.com/news/X.Org-Maintenance-Mode-Quickly). If you’re using one of those environments it is as easy as selecting the “Wayland” session at the desktop display manager ([GDM](https://en.wikipedia.org/wiki/GNOME_Display_Manager), [SDDM](https://en.wikipedia.org/wiki/Simple_Desktop_Display_Manager)).
We recommend **against** using desktop environments or window managers that do not have Wayland support, such as Cinnamon (default on Linux Mint), Pantheon (default on Elementary OS), MATE, Xfce, and i3.
Some Linux distributions (such as [Linux-libre](https://en.wikipedia.org/wiki/Linux-libre)-based or DIY distros) don’t come with the proprietary [microcode](https://en.wikipedia.org/wiki/Microcode) updates which patch critical security vulnerabilities. Some notable examples of these vulnerabilities include [Spectre](https://en.wikipedia.org/wiki/Spectre_(security_vulnerability)), [Meltdown](https://en.wikipedia.org/wiki/Meltdown_(security_vulnerability)), [SSB](https://en.wikipedia.org/wiki/Speculative_Store_Bypass), [Foreshadow](https://en.wikipedia.org/wiki/Foreshadow), [MDS](https://en.wikipedia.org/wiki/Microarchitectural_Data_Sampling), [SWAPGS](https://en.wikipedia.org/wiki/SWAPGS_(security_vulnerability)), and other [hardware vulnerabilities](https://www.kernel.org/doc/html/latest/admin-guide/hw-vuln/index.html).
We **highly recommend** that you install microcode updates, as they contain important security patches for the CPU which can not be fully mitigated in software alone. Fedora and openSUSE both have the microcode updates applied by default.
Most Linux distributions will automatically install updates or remind you to do so. It is important to keep your OS up to date so that your software is patched when a vulnerability is found.
Some distributions (particularly those aimed at advanced users) are more bare bones and expect you to do things yourself (e.g. Arch or Debian). These will require running the "package manager" (`apt`, `pacman`, `dnf`, etc.) manually in order to receive important security updates.
Additionally, some distributions will not download firmware updates automatically. For that you will need to install [`fwupd`](https://wiki.archlinux.org/title/Fwupd).
Many desktop Linux distributions (Fedora, openSUSE, etc.) come with [NetworkManager](https://en.wikipedia.org/wiki/NetworkManager) to configure Ethernet and Wi-Fi settings.
It is possible to [randomize](https://fedoramagazine.org/randomize-mac-address-nm/) the [MAC address](https://en.wikipedia.org/wiki/MAC_address) when using NetworkManager. This provides a bit more privacy on Wi-Fi networks as it makes it harder to track specific devices on the network you’re connected to. It does [**not**](https://papers.mathyvanhoef.com/wisec2016.pdf) make you anonymous.
We recommend changing the setting to **random** instead of **stable**, as suggested in the [article](https://fedoramagazine.org/randomize-mac-address-nm/).
If you are using [systemd-networkd](https://en.wikipedia.org/wiki/Systemd#Ancillary_components), you will need to set [`MACAddressPolicy=random`](https://www.freedesktop.org/software/systemd/man/systemd.link.html#MACAddressPolicy=) which will enable [RFC 7844 (Anonymity Profiles for DHCP Clients)](https://www.freedesktop.org/software/systemd/man/systemd.network.html#Anonymize=).
MAC address randomization is primarily beneficial for Wi-Fi connections. For Ethernet connections, randomizing your MAC address provides little (if any) benefit, because a network administrator can trivially identify your device by other means (such as inspecting the port you are connected to on the network switch). Randomizing Wi-Fi MAC addresses depends on support from the Wi-Fi’s firmware.
There are other system identifiers which you may wish to be careful about. You should give this some thought to see if it applies to your [threat model](../basics/threat-modeling.md):
- **Hostnames:** Your system's hostname is shared with the networks you connect to. You should avoid including identifying terms like your name or operating system in your hostname, instead sticking to generic terms or random strings.
- **Usernames:** Similarly, your username is used in a variety of ways across your system. Consider using generic terms like "user" rather than your actual name.
- **Machine ID:**: During installation a unique machine ID is generated and stored on your device. Consider [setting it to a generic ID](https://madaidans-insecurities.github.io/guides/linux-hardening.html#machine-id).
### System Counting
The Fedora Project [counts](https://fedoraproject.org/wiki/Changes/DNF_Better_Counting) how many unique systems access its mirrors by using a [`countme`](https://fedoraproject.org/wiki/Changes/DNF_Better_Counting#Detailed_Description) variable instead of a unique ID. Fedora does this to determine load and provision better servers for updates where necessary.
This [option](https://dnf.readthedocs.io/en/latest/conf_ref.html#options-for-both-main-and-repo) is currently off by default. We recommend adding `countme=false` to `/etc/dnf/dnf.conf` just in case it is enabled in the future. On systems that use `rpm-ostree` such as Silverblue, the countme option is disabled by masking the [rpm-ostree-countme](https://fedoramagazine.org/getting-better-at-counting-rpm-ostree-based-systems/) timer.
openSUSE also uses a [unique ID](https://en.opensuse.org/openSUSE:Statistics) to count systems, which can be disabled by deleting the `/var/lib/zypp/AnonymousUniqueId` file.
