* feat(webui): add /health and /health/detailed readiness endpoints (archie-core-1786637496789-265-b052b772) Implements the Operations: readiness probes bead from the hermes gap analysis (docs/prds/hermes-gap-analysis.md sections 9-10). - GET /health: liveness, no auth, always 200 ok. - GET /health/detailed: authenticated (dashboard token), JSON report with overall ok/degraded status plus per-subsystem probe results. New readiness probes, each reporting real current state: - state_db: read-only StatusCounts against the store. - config: configuration.Validate against the running config. - disk: statfs, degraded above 90% used. - model: active model configured + provider endpoint reachable. - gateway: channel running/failed state + connected session count. Layering: probe contract in internal/domain/health, implementations in internal/infrastructure/readiness, wiring in internal/app/archied, HTTP surface in internal/webui. * no-mistakes(document): docs(archied): fix stale doc pointer in readiness probe comment
32 lines
1.4 KiB
Go
32 lines
1.4 KiB
Go
package webui
|
|
|
|
import "net/http"
|
|
|
|
// handleHealth is a liveness probe: the process is up and able to serve a
|
|
// request. It deliberately bypasses requireToken, like /healthz -- a local
|
|
// orchestration tool (systemd, a watchdog, a container healthcheck) must be
|
|
// able to ask "is the process alive?" without the dashboard token. It does no
|
|
// subsystem work and always answers 200 when the process can respond at all.
|
|
func (s *Server) handleHealth(w http.ResponseWriter, _ *http.Request) {
|
|
w.WriteHeader(http.StatusOK)
|
|
_, _ = w.Write([]byte("ok"))
|
|
}
|
|
|
|
// handleHealthDetailed is the authenticated readiness surface. It runs every
|
|
// registered readiness probe and returns each subsystem's live result plus
|
|
// the overall rollup, so an operator can see exactly which subsystem is
|
|
// degraded rather than only that the daemon is unhealthy.
|
|
//
|
|
// It requires the dashboard token (and is mounted behind requireToken) because
|
|
// it exposes real subsystem state -- a remote caller should not be able to
|
|
// probe the disk layout or provider reachability unauthenticated. When no
|
|
// probe registry is wired, it answers 503 rather than pretending to know.
|
|
func (s *Server) handleHealthDetailed(w http.ResponseWriter, r *http.Request) {
|
|
if s.Health == nil {
|
|
http.Error(w, "readiness probes are not configured", http.StatusServiceUnavailable)
|
|
return
|
|
}
|
|
report := s.Health.Run(r.Context())
|
|
writeJSON(w, report)
|
|
}
|