Joseph Doherty 6733519ae5 fix(dashboard): derive gateway status from health checks; send config once per hub connection
SEC-27: DashboardSnapshot.GatewayStatus was hardcoded Healthy. It is now the
aggregate HealthCheckService status, evaluated once per snapshot tick (2s
bound, zero-wait gate) and cached for GetSnapshot; Unknown before the first
evaluation, on failure, or without a health service.

The startup-static EffectiveGatewayConfiguration is JsonIgnore'd on the
snapshot, so /hubs/snapshot ticks carry only volatile state. The hub sends it
once per connection (ConfigurationUpdated) and serves GetConfiguration().
Snapshot broadcasts are time-bounded (1s) so one slow client cannot delay
ticks. StatusBadge maps Unhealthy to Bad.
2026-10-22 06:18:02 -04:00
S
Description
No description provided
53 MiB
Languages
C# 48.4%
Java 40.1%
Python 3.2%
Rust 3%
Go 2.6%
Other 2.7%