Use Pro monitoring history to understand which checks changed between stored runs. Read the individual results before relying on “no regressions”.

Compare the stored runs

1. Open PD Modules → Guard → Monitoring & Alerts.
2. Review Check history and the run time, pass/warn/fail totals and What changed.
3. Expand a relevant run to read the regression detail.
4. Compare the affected finding with a fresh Report and its scope notes.

The monitoring screen shows the latest 20 stored runs; up to 50 are retained. The cards above the history are a fresh view and may differ from the most recent stored run. An older event absent from the displayed history has not necessarily never happened.

Understand what produces an alert

Status worsening from pass to warning or failure, or warning to failure, is a regression. The implementation also tracks an increased affected count for a supported count-aware check. It does not email every value change that remains at the same severity, every unchanged failure, or every improvement.

The first run has no previous comparison. Informational, unknown or omitted measurements are not proof that a previous problem was fixed. A newly appearing warning can be a regression even if that check had no prior recorded status.

Diagnose delivery independently

The new monitoring run is stored before notification is attempted. If sending fails, a following unchanged run may report no new regression and not retry the message. Check email and webhook evidence independently; “alert sent” in a run result is not confirmed inbox or downstream processing.

Avoid unredacted environment exports or customer records.

See the PD Guard Blesta guide for related tasks, or explore PD Guard.