Configuration Guardian — It Watches TrackMe Itself

Findings reach you — even when nobody is looking at the UI

TrackMe watches your data. The Configuration Guardian watches TrackMe — proactively detecting misconfigurations and must-know conditions across your deployment, and turning them into actionable, self-healing alerts. A service account missing a capability, a remote account whose token is about to expire, a backup that has gone stale, an AI provider that stopped answering, a tenant whose operations have degraded, a maintenance window left open past its end — the Guardian surfaces each one with a plain-language message, remediation steps, and a structured list of recommended actions.

And because every check is self-healing, the moment the underlying condition is resolved the next detection cycle clears the alert automatically. No manual housekeeping, no stale warnings — findings reach you when they matter and clear themselves when the condition heals.

A full management experience

TrackMe 2.4.7 turns the Guardian from a background safeguard into a first-class, manageable surface — a dedicated page organized in three tabs, with a KPI strip summarizing critical, warning and info counts, active and snoozed alerts, and the number of monitored checks.

Snooze, clear, or disable — you decide the noise

Not every finding needs to shout at you forever. The Guardian gives you three distinct controls, so you can acknowledge what you know without going blind to what you don’t:

• Clear now re-evaluates immediately — a temporary acknowledgement that comes straight back on the next cycle if the condition still holds.
• Snooze hides a known condition for 1, 7 or 30 days while detection keeps running underneath — and a warning that escalates to critical breaks through the snooze, because you silenced a warning, not an active degradation.
• Disable a check entirely from the catalog when it doesn’t apply to your deployment — its active alerts clear, and re-enabling resumes scanning automatically.

Snoozing is a read-time overlay, never a blindfold: the check keeps detecting, the record keeps updating, and the alert returns automatically when the snooze expires if the condition is still true.

It emails you — on transitions that matter

Configure a delivery account and recipients once, and the Guardian notifies you by branded HTML email on genuine lifecycle transitions: a new alert, an escalation to critical, or a resolution when the condition self-heals. Set a minimum severity so informational notices never reach your inbox, and pick exactly which of the three events you care about.

For teams routing through their own pipelines, an opt-in saved search additionally fires on any alert that becomes critical — attach any Splunk alert action you already use (email, ITSM, Slack, PagerDuty), with per-alert throttling so a persistent condition notifies once, not every cycle.

Ask AI — remediation, in context

When the AI Assistant is configured, Ask AI opens a chat pre-loaded with the Guardian’s live state: it sees the active alerts and their recommended actions, and walks you through remediation and verification. The same Guardian context reaches the Virtual Tenants and Tenant Home assistants too, so a question like “why aren’t my entities updating?” is answered by cross-checking the relevant checks first.

Observable and accountable by design

Every Guardian state transition — created, escalated, cleared, snoozed, unsnoozed, and every catalog change — is written to the TrackMe audit index under a dedicated sourcetype, so you always have a timeline of what changed and when. All state lives in inspectable KV Store collections, and on-demand runs are dispatched as a Splunk search job so even a fleet-wide scan can never be interrupted by a web-proxy timeout. The complete REST surface is self-documented in TrackMe’s built-in API reference.

Available to every deployment — no AI required

The Configuration Guardian management experience is available to every TrackMe deployment — the detection, the catalog, snooze, tuning, email notifications and the audit trail all work with no AI provider configured. The optional Ask AI button is the only part that needs the AI Assistant; the Guardian itself simply watches your deployment, and tells you — clearly, and only when it matters.