Does Latency on letter L share fate with other letters?

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

Where is Latency measured first?

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

Latency: headers in São Paulo (GRU) · L-45

Runbook L-45 for Latency in São Paulo (GRU). Marker xserv-l-1c326bbbc147. This page covers headers for LATAM operators, not a worldwide average.

Latency: What operators should measure

Policy for Latency is versioned as l-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.

Latency: Failure modes we actually see

Capacity notes for l-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.

Latency: How this letter ties to the rest

Handoff from letter L Latency into the rest of the platform uses the same request IDs as the gateway. Runbook l-05 names who accepts the ticket after São Paulo pages out. Santiago does not silently inherit the incident.

Latency capacity L-04

Latency handoff L-05

Latency rollback L-06

Rollback for l-06 is a documented command, not a hope. Transit is billed as backup while Latency prefers peering in Santiago. If cost spikes, the runbook cuts overflow to Bogotá before touching customer prefixes.

Latency peering L-07

Peering for Latency (l-07) prefers the IX fabric that already carries last-mile ISPs in Bogotá. Session counts and prefix limits are on the same page as the Miami backup path.