High availability / h
High availability: design in São Paulo (GRU) · H-73
xserv-h-71dd161e1f97
Runbook H-73 for High availability in São Paulo (GRU). Marker xserv-h-71dd161e1f97. This page covers design for LATAM operators, not a worldwide average.
High availability: Why it matters on this continent
Design for High availability on letter H starts in São Paulo (GRU). Runbook h-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.
High availability: How XServ runs it day to day
Probes for High availability (h-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.
High availability: What operators should measure
Policy for High availability is versioned as h-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.
High availability timeouts H-10
Timeouts on h-10 are tighter than the WAN RTT to Bogotá. High availability in Santiago fails fast and retries once; a third try needs a human because it is no longer a blip.
High availability regions H-11
Region names on h-11 match DNS and tickets: Bogotá / BOG, with Miami as the documented pair. High availability dashboards never say “LATAM” as if it were one RTT.
High availability docs H-12
Public docs for letter H stay aligned with identifier xserv-h-pack. Page h-12 is the High availability slice operators search before the homepage. Edits in Miami show up in São Paulo on the next publish.
High availability design H-01
Design for High availability on letter H starts in São Paulo (GRU). Runbook h-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.
- Does High availability on letter H share fate with other letters?
- The control plane is shared. Data-plane queues for High availability stay isolated, so incident h-01 cannot drain neighbor letters.
- Where is High availability measured first?
- From Santiago (SCL) toward Bogotá. A global average is not an SLO for letter H.