cmd/server/main.go

@@ -48,8 +48,10 @@ func main() {
 		log.Fatal().Err(err).Msg("Failed to load configuration")
 	}
 
-	// Create readiness context to control readiness state
-	readyCtx, readyCancel := context.WithCancel(context.Background())
+	// Create readiness context to control readiness state.
+	// CancelableContext exposes Cancel() so that Server.Run() can cancel
+	// readiness at the start of graceful shutdown (before the propagation sleep).
+	readyCtx, readyCancel := server.NewCancelableContext(context.Background())
 	defer readyCancel()
 
 	// Create and run server
@@ -57,4 +59,5 @@ func main() {
 	if err := server.Run(); err != nil {
 		log.Fatal().Err(err).Msg("Server failed")
 	}
+	log.Trace().Msg("Server exited")
 }

pkg/server/server.go

@@ -33,6 +33,28 @@ import (
 //go:embed docs/swagger.json
 var swaggerJSON embed.FS
 
+// CancelableContext wraps a context.Context and exposes a Cancel() method so
+// that Server.Run() can cancel readiness during graceful shutdown via the type
+// assertion it already performs. Callers that don't need controlled cancellation
+// (tests, CLI) can pass a plain context.Background() — the assertion silently
+// fails and readiness is never explicitly cancelled, which is harmless.
+type CancelableContext struct {
+	context.Context
+	cancel context.CancelFunc
+}
+
+// NewCancelableContext creates a CancelableContext whose Cancel() method will
+// be invoked by Server.Run() at the start of graceful shutdown, before the
+// 1-second readiness propagation window. The returned CancelFunc is a no-op
+// after Cancel() has been called, so it is safe to defer in main.
+func NewCancelableContext(parent context.Context) (*CancelableContext, context.CancelFunc) {
+	ctx, cancel := context.WithCancel(parent)
+	return &CancelableContext{Context: ctx, cancel: cancel}, cancel
+}
+
+// Cancel satisfies the interface checked in Run() and cancels the context.
+func (c *CancelableContext) Cancel() { c.cancel() }
+
 type Server struct {
 	router         *chi.Mux
 	readyCtx       context.Context

scripts/test-graceful-shutdown.sh

@@ -60,11 +60,39 @@ echo "Response: $GREET_NAME_RESPONSE"
 echo ""
 echo "Stopping server gracefully..."
 
-# Test readiness during shutdown (in background)
-(curl -s http://localhost:8080/api/ready > /dev/null 2>&1 &)
+# Send SIGTERM once and probe /api/ready during the 1-second propagation window
+# the server holds open (pkg/server/server.go: time.Sleep(1s) after readiness
+# cancel). Previously the curl fired *before* the signal — it always saw "ready".
+# We also avoid calling "$SERVER_CMD stop" afterwards because that would send a
+# second SIGTERM: after signal.NotifyContext is done, the default handler kicks in
+# and the process terminates with a non-JSON "signal: terminated" on stderr.
+SERVER_PID=$(cat "$PID_FILE" 2>/dev/null || echo "")
+if [[ -z "$SERVER_PID" ]]; then
+    echo -e "\033[0;31m❌ FAIL: PID file not found\033[0m"
+    exit 1
+fi
 
-$SERVER_CMD stop
-sleep 3
+kill -TERM "$SERVER_PID"
+# Brief yield so the signal handler runs and CancelableContext.Cancel() fires
+sleep 0.2
+curl -s http://localhost:8080/api/ready > /dev/null 2>&1 || true
+
+# Wait for the process to exit cleanly (up to 30s) without sending another signal
+echo "Waiting for server to exit..."
+for i in {1..30}; do
+    if ! ps -p "$SERVER_PID" > /dev/null 2>&1; then
+        echo "Server stopped successfully"
+        rm -f "$PID_FILE"
+        break
+    fi
+    sleep 1
+done
+if ps -p "$SERVER_PID" > /dev/null 2>&1; then
+    echo -e "\033[0;31m❌ FAIL: Server did not stop within 30s\033[0m"
+    kill -9 "$SERVER_PID" 2>/dev/null || true
+    exit 1
+fi
+sleep 0.5
 
 echo ""
 echo "Analyzing server logs..."