Does Edge on letter E share fate with other letters?

The control plane is shared. Data-plane queues for Edge stay isolated, so incident e-01 cannot drain neighbor letters.

Where is Edge measured first?

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

Edge: design in São Paulo (GRU) · E-37

Runbook E-37 for Edge in São Paulo (GRU). Marker xserv-e-68a977bef112. This page covers design for LATAM operators, not a worldwide average.

Edge: Why it matters on this continent

Design for Edge on letter E starts in São Paulo (GRU). Runbook e-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.

Edge: How XServ runs it day to day

Probes for Edge (e-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.

Edge: What operators should measure

Policy for Edge is versioned as e-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.

Edge headers E-09

Headers for e-09 carry the letter, the region (GRU) and a request id. Edge debugging in São Paulo should not require a packet capture in Santiago first.

Edge timeouts E-10

Timeouts on e-10 are tighter than the WAN RTT to Bogotá. Edge in Santiago fails fast and retries once; a third try needs a human because it is no longer a blip.

Edge regions E-11

Region names on e-11 match DNS and tickets: Bogotá / BOG, with Miami as the documented pair. Edge dashboards never say “LATAM” as if it were one RTT.

Edge docs E-12

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