fix(server): shut down through fx so buffered reports flush (closes #22)

The server ran os.Exit at the end of its own goroutine, racing fx's
teardown and sometimes killing the process before reportbuf's OnStop
flushed — silently losing a full flush window of telemetry on every
restart, at exit 0. Shutdown now goes through fx.Shutdowner, so every
OnStop runs in order.

The http.Server is built synchronously in OnStart before the serving
goroutine, so shutdown can no longer race or nil-deref it. A listen
failure exits non-zero via fx.ExitCode(1). reportbuf's OnStop is guarded
by sync.Once. writeTimeout now exceeds the chi per-request budget so that
budget is reachable. Dead startupTime, exitCode, and cancelFunc fields
are gone. A new test asserts a buffered report reaches disk after the
lifecycle stops.

Model: opus-4-8

backend/internal/server/http.go

@@ -5,20 +5,31 @@ import (
 	"fmt"
 	"net/http"
 	"time"
+
+	"go.uber.org/fx"
 )
 
 const (
 	readTimeout       = 10 * time.Second
 	readHeaderTimeout = 5 * time.Second
-	writeTimeout      = 10 * time.Second
 	idleTimeout       = 60 * time.Second
 	maxHeaderBytes    = 1 << 20 // 1 MiB
+
+	// requestTimeout (routes.go) is the single per-request
+	// processing budget, enforced by chi's middleware.Timeout.
+	// writeTimeout must exceed that budget so a handler can write
+	// its 503 when the chi timeout fires; if it were shorter the
+	// server would abort the write first and the chi budget would
+	// be unreachable dead configuration.
+	writeTimeout = requestTimeout + 5*time.Second
 )
 
-func (s *Server) serveUntilShutdown() {
+// newHTTPServer constructs the http.Server. It performs no I/O
+// and does not start listening.
+func (s *Server) newHTTPServer() *http.Server {
 	listenAddr := fmt.Sprintf(":%d", s.params.Config.Port)
 
-	s.httpServer = &http.Server{
+	return &http.Server{
 		Addr:              listenAddr,
 		Handler:           s,
 		MaxHeaderBytes:    maxHeaderBytes,
@@ -27,21 +38,29 @@ func (s *Server) serveUntilShutdown() {
 		WriteTimeout:      writeTimeout,
 		IdleTimeout:       idleTimeout,
 	}
+}
 
-	s.SetupRoutes()
-
+// listenAndServe runs the listener until the server is shut
+// down. A genuine listen failure (not the expected
+// ErrServerClosed from a clean shutdown) requests process
+// shutdown through fx with a non-zero exit code, so the failure
+// is visible to any supervisor.
+func (s *Server) listenAndServe() {
 	s.log.Info("http begin listen",
-		"listenaddr", listenAddr,
+		"listenaddr", s.httpServer.Addr,
 		"version", s.params.Globals.Version,
 		"buildarch", s.params.Globals.Buildarch,
 	)
 
 	err := s.httpServer.ListenAndServe()
-	if err != nil && !errors.Is(err, http.ErrServerClosed) {
-		s.log.Error("listen error", "error", err)
+	if err == nil || errors.Is(err, http.ErrServerClosed) {
+		return
+	}
 
-		if s.cancelFunc != nil {
-			s.cancelFunc()
-		}
+	s.log.Error("listen error", "error", err)
+
+	shutdownErr := s.shutdowner.Shutdown(fx.ExitCode(1))
+	if shutdownErr != nil {
+		s.log.Error("request shutdown failed", "error", shutdownErr)
 	}
 }

backend/internal/server/server.go

@@ -1,16 +1,14 @@
 // Package server provides the HTTP server lifecycle,
-// including startup, routing, signal handling, and graceful
-// shutdown.
+// including startup, routing, and graceful shutdown. The
+// process lifetime is owned by fx: shutdown is requested
+// through fx.Shutdowner so every component's OnStop hook runs
+// in dependency order.
 package server
 
 import (
 	"context"
 	"log/slog"
 	"net/http"
-	"os"
-	"os/signal"
-	"syscall"
-	"time"
 
 	"sneak.berlin/go/netwatch/internal/config"
 	"sneak.berlin/go/netwatch/internal/globals"
@@ -31,19 +29,18 @@ type Params struct {
 	Handlers   *handlers.Handlers
 	Logger     *logger.Logger
 	Middleware *middleware.Middleware
+	Shutdowner fx.Shutdowner
 }
 
 // Server is the top-level HTTP server orchestrator.
 type Server struct {
-	cancelFunc  context.CancelFunc
-	exitCode    int
-	h           *handlers.Handlers
-	httpServer  *http.Server
-	log         *slog.Logger
-	mw          *middleware.Middleware
-	params      Params
-	router      *chi.Mux
-	startupTime time.Time
+	h          *handlers.Handlers
+	httpServer *http.Server
+	log        *slog.Logger
+	mw         *middleware.Middleware
+	params     Params
+	router     *chi.Mux
+	shutdowner fx.Shutdowner
 }
 
 // New creates a Server and registers lifecycle hooks for
@@ -57,23 +54,25 @@ func New(
 	s.mw = params.Middleware
 	s.h = params.Handlers
 	s.log = params.Logger.Get()
+	s.shutdowner = params.Shutdowner
 
 	lc.Append(fx.Hook{
 		OnStart: func(_ context.Context) error {
-			s.startupTime = time.Now().UTC()
+			// Build the router and http.Server synchronously
+			// here, before spawning the serving goroutine, so
+			// httpServer is fully constructed by the time OnStop
+			// (or an early signal) can read it. fx guarantees
+			// OnStart returns before OnStop runs, so no
+			// synchronization or nil check is needed at shutdown.
+			s.SetupRoutes()
+			s.httpServer = s.newHTTPServer()
 
-			go func() { //nolint:contextcheck // fx OnStart ctx is startup-only; run() creates its own
-				s.run()
-			}()
+			go s.listenAndServe()
 
 			return nil
 		},
-		OnStop: func(_ context.Context) error {
-			if s.cancelFunc != nil {
-				s.cancelFunc()
-			}
-
-			return nil
+		OnStop: func(ctx context.Context) error {
+			return s.shutdown(ctx)
 		},
 	})
 
@@ -88,60 +87,17 @@ func (s *Server) ServeHTTP(
 	s.router.ServeHTTP(w, r)
 }
 
-func (s *Server) run() {
-	exitCode := s.serve()
-	os.Exit(exitCode)
-}
-
-func (s *Server) serve() int {
-	var ctx context.Context //nolint:wsl // ctx must be declared before multi-assign
-
-	ctx, s.cancelFunc = context.WithCancel(
-		context.Background(),
-	)
-
-	go func() {
-		c := make(chan os.Signal, 1)
-
-		signal.Ignore(syscall.SIGPIPE)
-		signal.Notify(c, os.Interrupt, syscall.SIGTERM)
-
-		sig := <-c
-		s.log.Info("signal received", "signal", sig)
-
-		if s.cancelFunc != nil {
-			s.cancelFunc()
-		}
-	}()
-
-	go func() {
-		s.serveUntilShutdown()
-	}()
-
-	<-ctx.Done()
-	s.cleanShutdown()
-
-	return s.exitCode
-}
-
-const shutdownTimeout = 5 * time.Second
-
-func (s *Server) cleanShutdown() {
-	s.exitCode = 0
-
-	ctxShutdown, shutdownCancel := context.WithTimeout(
-		context.Background(),
-		shutdownTimeout,
-	)
-	defer shutdownCancel()
-
-	err := s.httpServer.Shutdown(ctxShutdown)
+// shutdown gracefully stops the HTTP server within the
+// deadline of the context fx provides for OnStop.
+func (s *Server) shutdown(ctx context.Context) error {
+	err := s.httpServer.Shutdown(ctx)
 	if err != nil {
-		s.log.Error(
-			"server clean shutdown failed",
-			"error", err,
-		)
+		s.log.Error("server clean shutdown failed", "error", err)
+
+		return err
 	}
 
 	s.log.Info("server stopped")
+
+	return nil
 }