Per-component health view: mirrors, login service, multisig escrow, vendor dashboard, dispute queue, in-market messaging.
Component-level health is on the homepage component panel. As of the latest probe: all three mirrors green; login service green (synthetic-account handshake completes); multisig engine green (signer round-trip under 200 ms); vendor dashboard green (synthetic vendor read/write cycles succeeding); dispute queue green (mean ticket pickup under the 6-hour SLA); in-market messaging green (synthetic message round-trip under 1 s).
Health is distinct from latency. A component can be healthy (synthetic checks complete) while running elevated latency; that's what the “degraded” classification captures on the tertiary mirror today. Health goes red only when the synthetic checks fail outright, not on slow-but-still-completing responses.
A mirror can answer instantly while the escrow engine behind it is backed up. Rolling everything into one figure hides exactly the failures that affect an order, which is why the services are listed separately.
Login and messaging affect whether you can use the site at all. Escrow and the dispute queue affect money already committed. A slow vendor dashboard is a seller problem and mostly invisible from the buyer side.
Actions that normally complete instantly start taking minutes, then look like they failed while quietly succeeding. The failure mode to avoid is repeating the action, because two submissions are considerably harder to unpick than one slow one.
If a funding or release action stalls, wait for the page to settle and check the order state before touching anything. This is the single most expensive place to be impatient on any market.
Component states, the 90 day uptime history and the incident log live on the status dashboard. This page covers one question in depth rather than repeating the board.