SG SealGrid Athena Docs

Audit Event Types Reference

Every security-relevant action in Athena is written to the audit trail as an event with a stable event type — a short identifier like LoginFailure, AgentRevoked, or DeploymentExecuted. This page is the catalog of those event types, grouped by area and tagged with the severity Athena assigns. Use it to build SIEM correlation rules and alerts, to decide what to watch for, and to filter the audit log to exactly the activity you care about.

One catalog, everywhere

The same event types appear in the on-box audit log, in the console's audit view, and — when you turn on audit forwarding — in the copy streamed to your SIEM. The event-type string is stable, so a search or alert you write against it keeps working as the product evolves.

Anatomy of an event#

Beyond the event type, each audit record carries a consistent set of fields. These are the fields you filter, group, and correlate on in a SIEM:

FieldMeaning
Event typeThe stable identifier for what happened (the values catalogued below).
SeverityInformation, Warning, Error, or Critical (see Severity levels).
TimestampWhen the event occurred; a UTC form is also emitted for SIEM ingestion.
UsernameThe actor who triggered it — blank for system events.
IP addressThe client source IP — blank for agent-side or system events.
AgentAgent ID and hostname when the event concerns a specific agent.
Resource type / IDWhat was acted on — for example User, Agent, Deployment, Certificate, Settings.
ActionThe verb — Create, Update, Delete, Execute, Login, and so on.
OutcomeSuccess or failure, plus an error message when it failed.
DescriptionA human-readable summary of the event.
Old / new valueBefore-and-after values on updates, so a change is self-documenting.
Event ID & schema versionA unique ID per event and a schema version, so downstream searches stay stable.

For how those fields map onto the wire when forwarding to a collector (RFC 5424, RFC 3164, CEF, or JSON), see Message formats.

Severity levels#

Athena assigns each event one of these severities. Failures are generally escalated above their success counterpart — a successful login is Information, a failed one is Warning. In the audit view and in the PowerShell and REST filters, a severity filter is a minimum: choosing Warning returns Warning, Error, and Critical.

SeverityUse it for
InformationNormal, successful activity — routine operations and reads worth recording.
WarningSomething to notice — a failed login, a revocation or deletion, an unattended remote session, an off-box copy of credentials being used.
ErrorAn operation failed — a deployment, command, or scheduled job that did not succeed.
CriticalThe highest-risk events — uploading a new agent executable, or a blocked external script-injection attempt.

The event catalog#

The tables below list the event types Athena records, by area. Severity is the level assigned when the event is written; where an action has both a success and a failure form, both are shown.

Authentication & sessions#

Event typeSeverityWhen it fires
LoginSuccessInformationA user signs in. The provider that authenticated (Local or LDAP) is noted in the description.
LoginFailureWarningA sign-in attempt is rejected. The failure is deliberately generic to avoid revealing whether an account exists.
LogoutInformationA user signs out; the reason (for example an idle timeout) is recorded when there is one.

Users & access#

Event typeSeverityWhen it fires
UserCreatedInformationA new user account is created.
UserUpdatedInformationA user account's details are changed.
UserDeletedWarningA user account is deleted.
UserRoleChangedInformationA user's role is reassigned.
UserEnabled / UserDisabledInformationA user account is enabled or disabled.
PasswordChangedInformationA user changes their own password successfully.
PasswordChangeFailedWarningA self-service password change is rejected (for example, the current password was wrong).
PasswordResetWarningAn administrator resets another user's password.
ProfileUpdatedInformationA user edits their own profile (display name or email).
ProfileUpdateFailedWarning / ErrorA self-service profile edit fails.

Agents & certificates#

Event typeSeverityWhen it fires
AgentRegisteredInformationAn agent enrolls and receives its identity.
AgentUpdatedInformationAn agent's editable details are changed.
AgentRevokedWarningAn agent's certificate is revoked so it can no longer connect.
AgentDeletedWarningAn agent record is removed.
AgentMaintenanceEnabled / AgentMaintenanceDisabledInformationMaintenance mode is turned on or off for an agent.
AgentUnattendedChangedInformationAn agent's unattended-access setting is changed.
AgentRecordPolicyChangedInformationAn agent's session-recording policy is changed.
AgentUpdatePushed / AgentUpdatePushFailedInformation / WarningAn agent update is pushed to a machine.
AgentUpdateCompleted / AgentUpdateFailedInformation / WarningAn agent reports the result of an update.
AgentMigrationStarted / AgentMigrationStartFailedInformation / WarningAn agent is directed to move to a different server.
AgentInstallScriptGeneratedInformationAn install script for deploying the agent is generated.
AgentBundleDownloadedInformationAn agent install bundle is downloaded.
CACreatedInformationThe root certificate authority is created.
CAExtendedInformationThe root CA's validity is extended; existing agent certificates stay valid.
CARegeneratedInformationThe root CA is regenerated with a new key; existing agent certificates are revoked.
Watch these for lifecycle changes

AgentRevoked, AgentDeleted, and CARegenerated all change what can connect to your server. See Certificates & PKI and Retiring & Removing Agents for the operations behind them.

Agent binaries#

Event typeSeverityWhen it fires
AgentBinaryUploadedCriticalA new agent executable is uploaded — high-risk, because it becomes deployable code.
AgentBinaryActivatedWarningAn uploaded binary is made the active version that agents receive.
AgentBinaryDeactivatedWarningA binary is stood down from active use.
AgentBinaryDeletedWarningAn uploaded binary is removed.

Deployments, commands & jobs#

Event typeSeverityWhen it fires
DeploymentCreatedInformationA software deployment is defined.
DeploymentExecutedInformation / ErrorA deployment runs on an agent — Error on failure.
DeploymentCancelledInformationA deployment is cancelled.
CommandCreatedInformationA remote command is queued.
CommandExecutedInformation / ErrorA remote command runs on an agent — Error on failure.
CommandCancelledInformationA remote command is cancelled.
ScheduledJobCreated / ScheduledJobUpdated / ScheduledJobDeletedInformationA scheduled job is created, changed, or removed.
ScheduledJobExecutedInformation / ErrorA scheduled job runs; the affected/success/failure counts are recorded.

Compliance & scans#

Event typeSeverityWhen it fires
ComplianceRuleCreated / ComplianceRuleUpdated / ComplianceRuleDeletedWarningA compliance rule is created, changed, or removed.
ComplianceBaselineCreated / ComplianceBaselineUpdated / ComplianceBaselineDeletedWarningA compliance baseline is created, changed, or removed.
BaselineAssignmentCreated / BaselineAssignmentUpdated / BaselineAssignmentDeletedWarningA baseline is assigned to, or unassigned from, a collection.
ComplianceEvaluateNowWarningAn on-demand compliance evaluation is triggered.
ScanDefinitionCreated / ScanDefinitionUpdated / ScanDefinitionDeletedInformationA scan definition is created, changed, or removed.

Collections & credentials#

Event typeSeverityWhen it fires
CollectionCreated / CollectionUpdated / CollectionDeletedInformationA collection is created, changed, or removed.
CredentialCreatedInformationA credential is added to the vault.
CredentialUpdatedInformation / WarningA credential is edited — Warning when the secret itself changes.
CredentialDeletedWarningA credential is removed from the vault.
CredentialUsedInformationA stored credential is used to authenticate an operation.

Registration tokens#

Event typeSeverityWhen it fires
TokenCreatedInformationAn enrollment token is issued.
TokenRevokedInformationA token is revoked before it expires.
TokenDeletedInformationA token record is removed.

Remote sessions & recordings#

Event typeSeverityWhen it fires
ScreenShareSessionStartedInformation / WarningA screen-share session begins — Warning when it is an unattended session.
ScreenShareSessionEndedInformationA screen-share session ends.
ScreenShareClipboardSyncedInformationClipboard content is synced during a session.
VideoStreamRequestedInformationA remote video stream is requested.
VideoStreamDeniedWarningA remote video stream request is denied.
RecordingViewedInformationA saved session recording is viewed.
RelayProvisionedWarningA screen-share relay (TURN) is provisioned.

Server, database & security#

Event typeSeverityWhen it fires
ServerStartedInformationThe server starts up.
ServerStoppingWarningThe server begins a clean shutdown.
SettingsChangedInformationA server setting is changed; the old and new values are recorded.
DatabaseBackupInformationA database backup is taken.
DatabaseOptimizeInformationA database optimization runs.
EmergencyUnlockWarningAn emergency unlock is performed.
EmergencyIpBanWarningAn IP address is banned via the emergency controls.
CspViolationWarningA Content Security Policy violation is reported.
CspViolationAttackCriticalA CSP report is classified as an external script-injection attempt and blocked.
The catalog grows with the product

New capabilities add their own event types over time. The tables above cover the current set; treat the audit view's own event-type filter as the live source of truth for what a given build can emit (see Filtering by event type).

Filtering by event type#

In the console, open Audit Log and use the Event Type and Severity filters to narrow the view — for example, show only LoginFailure at Warning and above over the last day. The audit view populates its event-type list from the server, so it always matches your build.

To discover the commonly filtered event types programmatically, call the catalog endpoint:

GET api/Audit/event-types

This returns the frequently used event types (logins, user and role changes, agent registration and revocation, deployments, commands, scheduled jobs, settings changes, and server start/stop). The full set that can appear in the log is broader — everything catalogued on this page — because any recorded action carries its own type.

Filtering from PowerShell#

The PowerShell module queries the same trail. Filter by event type, severity, actor, text, and date range:

# Failed logins in the last 24 hours
Get-AthenaAudit -EventType "LoginFailure" -StartDate (Get-Date).AddDays(-1)

# Everything at Warning severity or above this week, all pages
Get-AthenaAudit -Severity Warning -StartDate (Get-Date).AddDays(-7) -All

# Deployment activity for a specific operator
Get-AthenaAudit -Username "jsmith" -Search "deployment"

-Severity is a minimum, mirroring the console. -StartDate and -EndDate also accept the -FromDate / -ToDate aliases. Add -All to page through every match instead of a single page.

Filtering over the REST API#

The same filters are available on the read API. Both endpoints require at least the Helpdesk role.

RequestWhat it does
GET api/Audit?eventType=LoginFailure&severity=WarningPage through the audit log filtered by event type and minimum severity. Also supports username, fromDate, toDate, success, search, page, and pageSize.
GET api/Audit/recentThe most recent events, for a quick pulse.
GET api/Audit/event-typesThe catalog of commonly filtered event types.

For the on-disk file locations, the audit providers, and how to forward this trail to a SIEM, see Audit & SIEM and Audit Forwarding to SIEM.