Skip to main content

Key concepts

Severity​

Each finding has one severity, from the vulnerability data:

SeverityMeaning in the interface
CriticalAct now. Usually remotely exploitable with little effort.
HighAct soon.
MediumPlan the fix in the normal cycle.
LowFix when convenient.
NegligibleVery minor; listed for completeness.
UnknownThe data has no severity for this vulnerability. Review it by hand.

The severity bars on the dashboard and the Severity filter use the same six levels.

CVSS​

CVSS is the base score of the vulnerability, from 0 to 10, when the data has one. Use the Minimum CVSS filter in More filters to hide low scores. A finding without a score shows an empty CVSS cell.

Fixable and fix state​

A finding is fixable when a fixed version of the package is known. The Fixable only filter shows just those findings. In the findings table the Fix column shows one of four states:

StateMeaning
Fixed, with a versionUpgrade to the version shown, or a later one.
No fix yetThe vulnerability is known, but no fixed version exists yet.
Won't fixThe upstream project decided not to fix it.
UnknownThe data does not say.

Findings​

A finding connects one vulnerability to one installed package on one asset. The same vulnerability on five hosts is five findings. Each finding has a first seen time, the first scan that reported it. Use the First seen since filter to see what is new.

Assets​

An asset is something that has an inventory:

  • a host: a whole machine, reported by a Linux, Windows or macOS agent, or the node of a Kubernetes cluster;
  • a container: one container of a pod, reported by a Kubernetes agent, with its cluster, namespace, pod, container and image.

Use the Kind filter to show hosts, containers or both.

Dependencies​

A dependency is a package found in an inventory: a library, an operating system package or an application. It has a name, a version, a type such as deb, rpm, npm, golang, maven or pypi, and a language where one applies. Its Package URL (purl) identifies it across tools. The Dependencies page lists every package, and each package links to the assets it is installed on.

Clusters, namespaces and nodes​

For Kubernetes assets the interface shows the cluster, the namespace, the pod and the container. A node is the host name of the agent that reported a row: a node agent and the host agent of the same Kubernetes node count as one node. Hosts without a cluster are reached through their node.

An image that runs in several containers is one asset, scanned once. Its findings are listed once, and the interface names every place it runs.

Scope​

Your scope decides which clusters and nodes you see. An administrator sets it per user. Your scope applies to findings, dependencies, assets, agents, the dashboard, exports and reports. A user with the whole scope sees everything; a scoped user sees the findings of their clusters and the hosts of their nodes, and nothing else.

Scans and inventories​

An agent sends its inventory to the server on a schedule, and after a package-manager change on Linux hosts. The server then scans the new inventory: it matches the packages against the vulnerability data. A new inventory shows as Scan pending on the Assets page until its scan has run. Scans run on a schedule and can be started by an administrator. See Scans and tasks.