1.0.0 marks FleetOps as stable. Docker, Docker Swarm, remote agents, Kubernetes, Helm and per-environment roles have settled, and releases now follow semantic versioning: breaking changes only in a new major version. Update the image tag and restart; the database migrates on start.
One image, two plans
Until now there were two images, fleets/fleetops for Community and fleets/fleetops-ee for Enterprise. From 1.0 there is one, fleets/fleetops, and Community and Enterprise are plans rather than editions:
- Every install starts on Community. The sidebar shows the plan and version under the logo.
- To move to Enterprise, buy a plan on the pricing page and paste the emailed key in Admin → License. The Enterprise features turn on without a reinstall or a restart, and admins on Community see an Upgrade card that says what it adds.
- If a license expires, Enterprise features keep working for 14 days, then turn off. Community features and data are never touched.
fleets/fleetops-ee is now a deprecated alias of the same image. It is still published with 1.x for existing installs, but a later release stops publishing it, so switch the image name to fleets/fleetops on the same data volume. The installer does this for you: curl -fsSL https://fleetops.run/install.sh | sh -s -- --upgrade. The fleetops-ee-* binaries on the releases page are copies of fleetops for the same reason.
Free to run, including in production
The FleetOps Enterprise License has a new section for Community use: you may run unmodified official builds for any purpose, production included, without a subscription, as long as no license key is installed and the Enterprise features stay off. Everything outside ee/ is still Apache-2.0. Using the Enterprise features in production, or running a modified ee/, still needs a subscription. See Open source.
Image vulnerability scanning (Enterprise)
FleetOps now scans the images on your Docker hosts, Swarm nodes and Kubernetes clusters with Trivy. Scanning is off until an admin turns it on in Admin → Security scanning, for every environment or one at a time; turning it on pulls the Trivy image (pinned by digest) and its vulnerability database.
Images are scanned when they arrive (pulled, built, loaded or tagged, from FleetOps or not), once a day, and on Scan now. Results are kept per digest, so the same image on ten hosts is scanned once. On Docker hosts and Swarm nodes, a short-lived scan container reads the image straight from the host’s storage, limited to 1 CPU and 1 GiB, one at a time per host. On Kubernetes, a Job per image runs in the fleetops-scan namespace.
Security in the sidebar lists every image across the fleet, with fixable critical vulnerabilities on images that have running containers at the top: a patch queue. Each image has a Vulnerabilities tab and badges in the image lists. Air-gapped environments point Trivy and its databases at a mirror. People see only the environments and namespaces they can access. The security scanning guide has the details.
A new look
The web UI was redesigned around a three-panel shell: an environment library on the left, the page in the middle, and a now-playing panel on the right that follows the container, service or pod you select, with its live log. A live event tape runs along the bottom, figures tick in place as they change, and every button carries its icon. The UI is now dark only, and has phone and tablet layouts.
Admin pages ask before destructive actions, keep one action per row with the rest in a menu, and the audit log is live and filterable. Admins can disable an account, and replacing a license previews what it gains and loses before installing it.
Next
Notifications for new CVEs, then port-forward, API tokens, SIEM forwarding and long audit retention. See the roadmap.