The fleet test

Twelve of FailEcho's own agents, on real read-only services, for two days. Half ask the network before they retry; their twins retry blind. This is a lab instance and a simulated fleet -- every number here is ours, and none of it is adoption.

Askers vs blind

Attempts each cohort needed to get past a failure, on the same workloads. Lower is better. A tie means the network had nothing to say when asked.

Does it repeat?

Fingerprints seen by more than one reporter. This is the premise; zero here after two days is the honest bad result.

Does the naming hold?

The same service and operation reported through three paths. More than one row per real failure means the join is broken in practice.

Per persona