Tannat cable: latency up to 3.9× normal, 28 Apr 2026
What we measured
| Cable | Tannat |
| Segment | Las Toninas, AR → Santos, BR |
| Detected | 2026-04-28 05:01 UTC |
| Resolved | 2026-04-28 08:30 UTC |
| Duration | 3h 29m |
| Observed latency | 125.8 ms |
| Baseline | 156.1 ms · 3.94× |
| Increase | +294% |
| Severity | critical |
Who is affected
Countries where Tannat comes ashore, how much of their submarine diversity this cable represents, and what our measurements into them show.
| Country | Dependence on this cable | Other submarine cables | What the measurements show |
|---|---|---|---|
| Argentina resilience index: 70 · moderate |
1 of 8 cables (13% of submarine diversity) | ARSAT Submarine Fiber Optic Cable, Bicentenario, Firmina +4 more | median 270 ms today vs 214 ms over 30 days (96 routes) - slower |
| Brazil resilience index: 100 · resilient |
1 of 26 cables (4% of submarine diversity) | America Movil Submarine Cable System-1 (AMX-1), Brazilian Festoon, BRUSA +22 more | median 144 ms today vs 152 ms over 30 days (378 routes) - unchanged |
| Uruguay resilience index: 44 · vulnerable |
1 of 4 cables (25% of submarine diversity) | Bicentenario, Firmina, Unisur | no measurement today; median 115 ms over 30 days (22 routes) |
Dependence is structural (share of the country's submarine cables), not a traffic share: real traffic distribution across cables is not public.
Path analysis
Traffic kept the same route. Latency rose without a visible path change, which points at congestion or a fault on the existing route rather than a reroute.
Cross-probe verdict. Other measurements did not confirm the event. We treat this as a probable false positive and label it as such.
Timeline
- Alert 2026042801: warning - ratio 3.94x
- Alert #96 improving: 1/3 consecutive OK
- Alert #96 improving: 2/3 consecutive OK
- Alert #96 resolved: RTT 125.79ms (baseline 156.05ms), 3 consecutive OK
Detection method
Latency is sampled continuously from our probe network. An alert opens when round-trip time to a monitored endpoint holds above its rolling baseline, and closes automatically once it returns. Full method is described on the methodology page.