The new docs.trellix.com offers a modernized UI and AI-powered features like conversational searches. Content is currently available only in English. Other languages will be available in mid-October 2026. We hope you enjoy the new experience.

Alerting rules

Prev Next

The Alerting Rules management screen allows you to create, edit, enable, disable and delete custom rules. Once rules are added, each incoming event is evaluated against all the enabled rules. If an event matches a rule, an alert is triggered and the relevant alert metadata is available from the main Endpoint Alerts page.

The module includes a handful of templates for detecting various forms of lateral movement.

Note

These templates are for reference only and should be customized for the specific environment.

The rule editor allows you to define one or more conditions for specific fields within a logon event. For example, to generate an alert any time an account from a list attempts to log into a system, one could create a rule with a single condition as shown in the figure.

image17.png

The rule above used the “IN” membership to specifically list what should trigger an alert. Conversely, a “NOT IN” membership condition can be used to alert if something is not in a list. For example, if a system should only be accessed by a list of authorized accounts, an “IN” condition combined with a “NOT IN” condition, can be used to generate an alert when logon with an unauthorized account is attempted, as shown in figure.

image18.png

The available event attributes for which conditions can be authored are listed below. String fields are case-insensitive and support wildcard matching with ‘*’ and single character matching with ‘?’.

  • Source/Target attributes: Attributes related to the source or target account or system

    • Accounts: Account names

    • Account Privileges: Privileges (unknown, privileged, unprivileged)

    • Hostnames: Hostnames

    • Networks: Networks (supports CIDR notation)

    • Machine Type: Machine type (server, workstation, unknown)

    • Domain Groups : Domain groups associated to the logon account

    • Local Groups: Local groups associated to the logon account

  • Additional event metadata: Metadata about the logon event

    • Logon Categories: Category of the logon event

    • Paths: Path to the logon executable from the event

    • MD5s : MD5 of the process from the event

    • Event IDs: Event ID from the Windows event

    • Logon Types: Logon Types from the Windows event

    • xAgent IDs: xAgent ID on which the event was observed

    • Logon Status: Result of the logon attempt (SUCCESSFUL, FAILED, UNKNOWN)

    • Tunneled: A boolean of whether we detected a tunneled connection

    • Type: Type of logon that occurred