Fleet announced a new suite of enhanced, multi-cloud features for DevOps and cloud security teams, adding observability tools for servers and Kubernetes containers. These capabilities simplify access to previously hard-to-reach data from sensitive production environments, such as gaming servers and edge caching nodes.
Industry pioneer Mark Burgess, author of CFEngine and initiator of Promise Theory put it like this: “Configuring stuff is easy, but understanding the monster you’ve created is hard. This is why the challenges of infrastructure are all about knowledge management. Getting data is half the battle, and that’s a big scaling challenge.”
But the need for good data from production systems has never been greater. Servers and containers have many of the same basic security visibility needs as laptops, like intrusion detection (HIDS), vulnerability reporting, and live investigation during incidents (DFIR). They just require more painstaking steps to ensure the mere act of gathering the data doesn’t cause negative performance impact or expensive downtime. As companies increasingly rely on vast networks of servers to run their businesses, many opt not to take the risk of installing commercial endpoint detection and response (EDR) software on their production environments, worried about the performance “tax” and stability issues like the recent worldwide “blue screen of death” that shut down some airports in the United States.
By contrast, Fleet is designed to scale seamlessly from tens of servers, to thousands of servers, to hundreds of thousands of servers, with minimal performance impact. This dramatically simplifies gathering data for compliance audits, and it makes it possible to build more advanced security paradigms. Now, Fleet is drawing increasing attention from companies that need to maximize both security and performance from their production infrastructures.
“We’re able to address reliability and compliance concerns without sacrificing a single point-of-a-percent of performance for our servers. All of this done consistently and continuously,” said Charles Zaffery, principal infrastructure engineer at Roblox. “Fleet is getting shown to our company board.”
Fleet’s history and open-source nature give it a unique advantage. Built atop osquery, a popular open-source agent developed by Fleet co-founder Zach Wasserman during his time on Facebook’s cybersecurity team, it has since been deployed across millions of devices and adopted widely in large enterprises and hobbyist setups worldwide. This existing install base, coupled with the thousands of people reading its source code, make Fleet and osquery highly customizable and provide a firm, scalable foundation for shipping new, enterprise-ready features quickly. For example, in just the last couple of months Fleet quietly trickled out multiple new features for infrastructure teams, including Arm64 support, BPF events on Arch Linux, remote installation of RPM packages, OVAL vulnerability scanning, auto-patching, zsh shell scripting, and AI-powered explanations of security policies for developers who build on company infrastructure.
“Fleet’s extremely wide and diverse set of data allows us [Roblox] to answer questions that we didn’t even know we had,” said Charles. “On top of that, the experience is near instantaneous: Literally, seconds to sort through billions of data points and return the exact handful that we need, with complete auditing and transparency.”
This data is one of Fleet’s superpowers. One example: Dre, a security engineer at a top electric vehicle manufacturer, used Fleet to identify which of their computers had a vulnerable TPM chip, an obscure hardware component. Even when the computer manufacturer couldn’t give him a straight answer over the phone, Fleet identified the vulnerable chip in seconds, across thousands of machines – from employee laptops to vehicle production lines.
“We picked Fleet for the simplicity of rolling it out and the ability to integrate into our environment,” said another engineer at a different Fortune 100 computer and networking company. “Now, it is running on around 80,000 hosts internally at my company for threat detection, security reporting, and vulnerability management.”
Atlassian, Dropbox, Roblox, Nubank, Fastly, and others have increased their adoption over time, standardizing on Fleet to easily get data and maintain surgical control across their entire infrastructure. For example, at Fastly, the cybersecurity department first purchased Fleet to replace a legacy, proprietary product installed on their servers, including high-performance caching nodes, where even a small performance hit is critical. Then one year later, new features in Fleet allowed it to spread to the IT department and replace a legacy device management vendor, eliminating more tool overlap while unifying security data across servers and laptops.
The growth and IPOs of open-core companies like GitLab and Hashicorp have paved the way for open-core companies like Fleet. Unlike with proprietary, “black box” security software, which can be difficult to troubleshoot and can raise suspicions from engineers, Fleet customers retain access to 100% of the source code that runs on their servers and containers. This makes Fleet easier to procure and easier for anyone to adopt and use, inside and outside of the enterprise.
“Five years ago, I worked as a backend developer on an in-store payments product for the Fortune 1 company, and back then, there were a lot of late nights. We had nothing but buggy, in-house tools for doing deployments and checking up on servers, plus a few commercial products that someone had bought, but that no one actually used,” said Mike McNeil, CEO and co-founder of Fleet. “The first time I met Zach and saw osquery, I realized this was going to change everything.”
