EmoTracker 3.0.3.2 becomes unresponsive during SNI autotracking.

[akadubss]

EmoTracker runs normally for approximately 30–60 minutes while autotracking ALttPR.

The UI eventually freezes completely.

The tracker window no longer responds to input.
Windows reports the application as "Not Responding."
SNI continues running.
The game continues running on the SNES.

The issue is reproducible after longer play sessions.

• Installed the latest NDI Runtime (fixed missing NDILib.dll errors)
• Fresh EmoTracker logs
• Fresh SNI session
• Rebooted PC
• Captured a full memory dump while frozen

If more information is needed please let me know.

EmoTracker: 3.0.3.2 (official x64)
CodeTracker package
SNI: 0.103
Hardware: Original SNES
Flash Cart: FXPak Pro / SD2SNES
Firmware: 1.11.1
Windows 11

EmoTracker.Extensions.AutoTracker.AutoTrackerExtension.ReadU8

Call stack:

AutoTrackerExtension.ReadU8
System.Threading.Tasks.Task.InternalWait
System.Threading.Monitor.Wait
NLua
lua54.dll
ScriptManager.SafeCall
LuaMemorySegment.OnSegmentDataUpdated
Avalonia Dispatcher (UI thread)

APPLICATION_HANG_BusyHang

Failure Bucket:
APPLICATION_HANG_BusyHang_cfffffff_win32u.dll!NtUserMsgWaitForMultipleObjectsEx

---

Raw Logs
PROCESS_NAME: EmoTracker.exe

ERROR_CODE: (NTSTATUS) 0xcfffffff

FAILURE_BUCKET_ID:
APPLICATION_HANG_BusyHang_cfffffff_win32u.dll!NtUserMsgWaitForMultipleObjectsEx

...

System_Private_CoreLib!System.Threading.Tasks.Task.InternalWait
System_Private_CoreLib!System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification

EmoTracker!EmoTracker.Extensions.AutoTracker.AutoTrackerExtension.ReadU8

System.Reflection.RuntimeMethodInfo.Invoke
NLua!NLua.Method.LuaMethodWrapper.CallInvoke
NLua!NLua.Method.LuaMethodWrapper.CallMethodFromName
NLua!NLua.Method.LuaMethodWrapper.Call

lua54.dll

EmoTracker.Data.ScriptManager.SafeCall

EmoTracker.Data.AutoTracking.LuaMemorySegment.OnSegmentDataUpdated

Avalonia.Threading.DispatcherOperation.InvokeCore
Avalonia.Threading.Dispatcher.ExecuteJobsCore

Avalonia.Win32.Win32Platform.WndProc
Avalonia.Win32.Win32DispatcherImpl.RunLoop
Avalonia.Threading.Dispatcher.MainLoop

EmoTracker.Program.Main

---

PROCESS_NAME: EmoTracker.exe

ERROR_CODE: (NTSTATUS) 0xcfffffff

SYMBOL_NAME:
win32u!NtUserMsgWaitForMultipleObjectsEx+14

FAILURE_BUCKET_ID:
APPLICATION_HANG_BusyHang_cfffffff_win32u.dll!NtUserMsgWaitForMultipleObjectsEx

[emosaru]

This should be fixed by the SNI/NWA autotracker connection-loss handling work (commit bb61fc4, merged to main).

What changed:

• all SNI gRPC calls are now bounded with a 2s deadline and the connection is demoted on connection-level read/write failure, so a dropped/stalled SNI server can no longer block reads or leave the tracker non-responsive
• concurrent reconnect attempts are serialized via a connect lock on both SNI and NWA devices
• the tracker stays armed and transparently auto-reconnects in the background with a transient Reconnecting warning state, instead of stopping + erroring

This is on main and will ship in the next release after v3.0.3.2. If you can, please retest against the latest dev build.