Evidence

The pillar the others feed: an availability record you can hand to a customer, an auditor, or a regulator, and let them recompute it.

Real boundaries

An incident’s window runs from quorum confirmation to confirmed recovery, the same seconds you were paged on. Exclusions from discarded false positives are computed per region, from the last good check before to the first good check after. Availability is floored, never rounded up: an SLA figure should be a lower bound you can defend.

Corrections that survive scrutiny

A false positive is discarded with a reason and a preview of the uptime effect; every discard carries its actor and is reversible. The History view keeps three parallel stories: the raw timeline, the curated outage record, and the configuration audit. Perstat does not claim the store is tamper-proof; it claims every change is attributed, reasoned, and undoable, which is what a reviewer actually checks.

Retention, enforced twice

7 days to 2 years depending on plan, enforced by pruning and by read-clamping, so a report can never quietly draw on data the plan does not carry. Long windows serve from daily rollups without approximation.

In an audit context

Perstat produces the independent availability evidence that feeds an ISAE 3402 or IDW PS 951 assurance report and a DORA incident notification. It is not that report and does not make you compliant; the leadership page draws the boundary, and the sample SLA report shows the artifact structure with numbers that add up. A one-click report export is the stated next step; evidence is exportable on request today.

See the sample SLA report, or read evidence and retention in the docs.