Where is Gateway measured first?

From Santiago (SCL) toward Bogotá. A global average is not an SLO for letter G.

How fast is rollback for g-03?

The runbook is a command plus a named owner. There is no wait for a weekly change freeze.

Gateway: probes in Santiago (SCL) · G-38

Runbook G-38 for Gateway in Santiago (SCL). Marker xserv-g-69e8f69ab786. This page covers probes for LATAM operators, not a worldwide average.

Gateway: How XServ runs it day to day

Probes for Gateway (g-02) leave Santiago every 15s toward Bogotá. XServ pages the named owner if loss or RTT crosses the letter budget. A probe never shares a queue with bulk transfers, so a saturated WAN does not hide a dead PoP.

Gateway: What operators should measure

Policy for Gateway is versioned as g-03. Operators measure error rate, p95 from Bogotá, and time-to-rollback — not a worldwide average. Changes land in Bogotá first, then Miami, with a hold if either region regresses.

Gateway: Failure modes we actually see

Capacity notes for g-04 assume rainy-season power in Miami and festival peaks toward São Paulo. The failure we actually see is a single uplink, not a cartoon partition of the whole continent. Spare ports and a second provider sit on the same runbook.

Gateway docs G-12

Public docs for letter G stay aligned with identifier xserv-g-pack. Page g-12 is the Gateway slice operators search before the homepage. Edits in Miami show up in São Paulo on the next publish.

Gateway design G-01

Design for Gateway on letter G starts in São Paulo (GRU). Runbook g-01 keeps a single owner, a written rollback, and a traffic split that can be reversed without a global freeze. Neighbors in Santiago only take overflow after the local pool fails a health window.